UNIVERSIDAD COMPLUTENSE DE MADRID 

FACULTAD DE MEDICINA 

 

 
 

 

TESIS DOCTORAL 

 

 
Estrategias para la mejora de la persistencia y la eficacia de las células CAR 

NK para el tratamiento del mieloma múltiple refractario y en recaída 
 

Strategies to improve persistence and efficacy of CAR NK cells for the 
treatment of relapsed and refractory multiple myeloma 

 

 
MEMORIA PARA OPTAR AL GRADO DE DOCTOR 

 

PRESENTADA POR 
 

 

 
Jessica Encinas Mayoral 

 
 
 

DIRECTORES 
 
 
 

Antonio Valeri Lozano 
Daniel Primo Ramos 

Joaquín Martínez López 
 

 

 

 

© Jessica Encinas Mayoral, 2024 



 

 
 

  



 

 
 

UNIVERSIDAD COMPLUTENSE DE MADRID 

FACULTAD DE MEDICINA 

DOCTORADO EN INVESTIGACIÓN BIOMÉDICA 

 

 

 

TESIS DOCTORAL 

 

ESTRATEGIAS PARA LA MEJORA DE LA PERSISTENCIA Y LA EFICACIA DE LAS 

CÉLULAS CAR NK PARA EL TRATAMIENTO DEL MIELOMA MÚLTIPLE 

REFRACTARIO Y EN RECAÍDA 

 

STRATEGIES TO IMPROVE PERSISTENCE AND EFFICACY OF CAR NK CELLS FOR 

THE TREATMENT OF RELAPSED AND REFRACTORY MULTIPLE MYELOMA 

 

MEMORIA PARA OPTAR AL GRADO DE DOCTOR 

PRESENTADA POR 

 

Jessica Encinas Mayoral 

 

DIRECTORES 

Antonio Valeri Lozano 

Daniel Primo Ramos 

Joaquín Martínez López 

 

 

Madrid, 2023 



 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

 
 

 

 

 

 

 

 

 

 

 

A mis padres, Asun y Jose, 

A mi hermano Iván, 

A mi familia, 

 

 

 

 

 

 

 

 

 

 

 

 



 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

 
 

AGRADECIMIENTOS / ACKNOWLEDGMENTS 

 

Sin duda una tesis no es un trabajo que implique la dedicación de una sola persona y en 

ésta han participado muchísimas personas con las que he tenido la suerte de trabajar estos 

últimos años. 

En primer lugar, quiero agradecer la cesión de muestras por parte de pacientes de mieloma 

múltiple y donantes sanos a este proyecto, porque sin su colaboración este trabajo no 

hubiera sido posible.  

También quiero agradecer la dedicación de mis directores de tesis: Antonio, Joaquín y 

Dani. Muchas gracias, Joaquín, por darme la oportunidad de trabajar en el grupo de 

hematología y poder desarrollar esta tesis que, sin duda, tampoco habría sido posible sin 

la ayuda de Antonio y de Almu. Muchas gracias, Antonio, por guiarme, por enseñarme 

todo sobre las NKs, por corregirme los “le”, que todavía sigo sin decir bien, y por tu 

apoyo estos años. Almu, que “como llegué tarde al reparto de postdocs, me tocó ella”, y 

la verdad es que no podría haber tenido una mejor. Ha sido un pilar fundamental de esta 

tesis, de estos dos proyectos, y la que nos ha enseñado todo sobre citometría y cultivos. 

Muchísimas gracias por ayudarme y por cuidarme, y recuerda que vales mucho. 

Muchas gracias también a Dani y a Joan por hacerme un hueco en Vivia, guiarme con los 

experimentos y apoyarme en esta tesis.  

Elena y Eva, mis compañeras de tesis. 

Elena, aquella muchacha a la que sentamos en una silla en mitad del despacho, porque no 

teníamos más sitio y que rápidamente se hizo hueco con sus “fantásticas” NK-92. Juntas 

aprendimos a hacer “calceínas”, expandir NKs, pasar citometría y a infundir a los ratones 

por cola. Eso sí, si alguien necesita optimizar algún protocolo que no dude en hablar con 

ella. Mil gracias por tu ayuda durante toda la tesis, por tus consejos (sobre todo para 

pintarme las uñas permanentes, que obviamente no me quedan igual que a ti). Me hace 

muchísima ilusión volver a trabajar contigo. 

Eva la pobre tampoco llegó en un buen día, por motivos que no voy a comentar aquí, pero 

¿qué pensaría ese primer día como predoc? Menos mal que la cosa fue mejorando, o no. 

Muchas gracias mujer por tu apoyo durante estos años, por esas primeras producciones 



 

 
 

de virus, por ayudarme con los experimentos cuando ya era tarde, por pesar a los ratones 

cuando me daba miedo y por enseñarme Villaviciosa junto a tu perrito, menos mal que le 

caí bien. 

Raquel, sin duda mi mayor descubrimiento, y es que, aunque estudiamos juntas la carrera 

en Alcalá, creemos que no llegamos a hablar ni una sola vez. Yo no creo en el destino, si 

no en las casualidades, y me ha encantado que, “por casualidad”, haya tenido la 

oportunidad de conocerte y compartir este último año y medio contigo. Gracias por 

ayudarme con las expansiones de las NKs, las producciones de virus, los western y por 

las recomendaciones de restaurantes, porque si hay alguien que más disfrute comiendo, 

es ella. 

Elena y Marta, nuestras chicas de TFM y TFG. Mil gracias, chicas, por ayudarnos con 

nuestras tesis y espero que os hayamos enseñado alguna cosa útil. Yo creo que en el fondo 

echan de menos cultivos y el citómetro, podéis volver cuando queráis. Y, sobre todo, 

gracias, Marta, por amar mis audios (los que los sufren saben que mis audios, no son 

audios, si no podcasts). 

Rebe, no te lo vas a creer, pero “la princesa de Asturias” va a defender por fin la tesis. 

Muchas gracias por cuidarme durante el TFM y ser uno de mis referentes durante mis 

primeros meses en el labo, te hemos echado mucho de menos. 

Ana, muchas gracias por tu apoyo mientras escribía la tesis y por el tiempo que hemos 

compartido estos últimos meses. 

Mis Lauris, Laura Córdoba y Laura Sánchez. No sé cuántas veces habremos reído/llorado 

en el despacho o en el comedor, y es que sois de lo mejor. Laura Córdoba sabe bien lo 

que cuesta expandir unas NKs, y ya si encima esa semana se transducen bien, todo un 

logro. Y es que ha sido una de mis compañeras de campana durante toda la tesis. ¡Qué 

maravilla eso de tener una campana de cultivos para dos! Y también de citómetro, sobre 

todo pasando Legendplex, una de nuestras actividades favoritas. Me ha encantado 

conocerte y compartir todos estos años contigo mujer. Laura Sánchez, no sabes lo que te 

hemos echado de menos estos meses, muchísimas gracias por tu apoyo y me alegro 

muchísimo de haber compartido este tiempo contigo. 

La otra pecera, Roberto, Raquel, Irene, Ale Ortiz, Andrés, Alba, Richard y María. A María 

la conocí por primera vez cuando nos dio una clase en el máster y poco a poco fui 



 

 
 

conociendo al resto del grupo, y la verdad es que sois todos estupendos. Muchas gracias, 

Richard, jefe del Hemato Beer Team, por ayudarnos con la secuenciación de ARN (y a 

Andrés por explicármela jaja). A Roberto, por ser como es, no se puede explicar, pero sin 

duda una de las personas a las que más le cunde el día, yo creo que el giratiempo de 

Hermione lo tiene él, porque si no, es imposible. Y por supuesto, para persona especial 

Raquel, reina de la fiesta, de los brindis y de los chupitos, que solo con encontrártela por 

el labo ya te alegra el día. 

Natalia y Noe, desterradas en la tercera planta, pero parte esencial del grupo. Muchas 

gracias, chicas, por vuestro apoyo estos años, por las risas y por todos los momentos que 

hemos compartido. Sigue quedando pendiente una merienda en Alcorcón. 

Isa, la primera persona que conocí del labo y que casualmente estaba estudiando la 

enfermedad de la que murió mi abuelo años antes, amiloidosis. Ojalá pronto se desarrollen 

también más terapias para las enfermedades raras. Ha sido un placer compartir estos años 

contigo. 

Mariluz, de las veteranas del labo cuando aparecí por allí y a la que tenía frita preguntando 

donde estaba todo, pero bien que me dejó sola para que me entrevistaran los de Cris y les 

contara lo maravillosa que es la terapia CAR. Por cierto, no te preocupes que ya me voy 

del labo, dejo sitio en la campana para que puedan trabajar los demás y dejo de romper 

cámaras de Neubauer, aunque últimamente me están quitando el puesto. 

Mis chicas de biobanco. Yo creo que a veces me temían cuando me veían aparecer, porque 

sabían que las iba a pedir algún favor y es que todas las muestras de pacientes que se han 

usado en esta tesis, y en otros tantos experimentos, han pasado por sus manos o me han 

ayudado a buscarlas en la Intranet, ya fueran de extracciones, de la HUNET o de Hospital 

de Día. Mil gracias, Vane, Ali, Laura Carneros (puberta y pepinera, a partes iguales), 

Laura Moreno, Lucía, María, Irene y Sandra, por ayudarme a conseguirlas prácticamente 

todas y por vuestras charlas y apoyo mientras marcaba citometría en la poyata a vuestro 

lado. 

Paqui, el alma del labo y madre de todos. Mil gracias por cuidarme durante toda la tesis, 

por hacerme reír (sobre todo desde la mesa) y contarme anécdotas de Timoteo. Marijose, 

que escondía las puntas por todo el laboratorio y luego nos las iba repartiendo como si 

fuera contrabando. Paloma, Merche y Manu, los del turno de tarde, que menos mal que 

estaban por allí para amenizarnos esos ratos.  



 

 
 

Citometría, Teresa Cedena, Fátima, Elvira, Lorena y Teresas (técnicos). Seguramente 

haya pasado la mitad de la tesis en su planta, pasando citometría. Muchas gracias a todas 

por enseñarme citometría, prestarme anticuerpos, hacerme “spoiler” del porcentaje de 

plasmáticas que tenían las muestras para poder poner los experimentos pronto, y por 

ayudarme a arreglar el citómetro cada vez que no funcionaba. 

Hematólogos. Muchas gracias a todos por ayudarnos a conseguir muestras y en especial 

a Rafa, por ayudarnos siempre con los proyectos y los datos de los pacientes. 

Sala blanca, Adri, Patri C, Patri del Moral, Ale Leivas y Fran. Ha sido un placer conoceros 

a todos y compartir estos años con vosotros, desde luego a vosotros tampoco os lo han 

puesto fácil. Muchas gracias, Ale, por tu ayuda desde que entré al labo, con la citometría, 

el sorter y el Legendplex. 

Esther. Menuda guerra te hemos dado entre todos. Muchas gracias por los tantísimos 

pedidos que te hemos mandado tramitar y también por las charlas. 

Mis chicos de Vivia, Miguel, Adri, Ana, Chechu, Vero y David. También se alegraban 

mucho de verme, sobre todo los días que tenían muchas muestras y yo me pasaba la 

mañana en el citómetro. Muchas gracias, chicos, por enseñarme a trabajar en vuestro labo, 

por las risas, los desayunos, la música de fondo (casi siempre canciones de verbena de 

pueblo), los puzles del Layton y sobre todo por vuestro apoyo, ya que la mayoría de las 

veces los experimentos no salían como esperaba, pero allí estabais vosotros para 

alegrarme el día. 

CNIO, Pedro, Michel, Álvaro, Lara, Miguel y María Velasco. A pesar de que no he 

coincidido mucho con vosotros, los ratos que hemos pasado juntos han sido de los 

mejores, sois fantásticos. Muchísimas gracias por vuestro apoyo y por ayudarnos con las 

irradiaciones. 

CIEMAT, Paula, Lauras, Irene, Marta y Bea. Muchísimas gracias, Paula, por ayudarnos 

con estos proyectos durante estos años y también a vosotras chicas, por tratarme tan bien 

y ayudarme cada vez que iba por el CIEMAT, sois estupendas. 

Grupos UNICA y mitocondriales. Menos mal que no estábamos solos en la planta y 

hemos tenido la suerte de compartir cultivos con gente tan maravillosa y tan maja, para 

poder hacer intercambios de reactivos. Muchas gracias, chicos, por vuestro apoyo y ha 



 

 
 

sido un placer conoceros y compartir tantas mañanas, y otras tantas tardes, con vosotros 

en cultivos. Y en especial, a Rocío y Pablo, por los experimentos de metabolismo. 

Dana Farber team. First, I would like to thank Dr. Munshi and Mariateresa for taking me 

in and let me be part of their team during those months. It has been a great experience and 

I have enjoyed working with all of you. Thank you so much, Yao, for teaching me new 

techniques, helping me every time I needed it and taking care of me. Jessica, it has been 

a pleasure working with you, thank you for your help and your tips, above all related to 

good restaurants. The Italian team: Anna, Eugenio, Domenico and Marcello. Thank you 

so much for your help and the time we spent together. My dear Dori, thank you so much 

for taking care of me, for touring me around Boston and took me to incredible places, I 

have been very lucky to meet you. 

Rocío y Ana. ¡Qué haría yo sin estas muchachas! Y es que valen para todo, da igual cual 

sea el problema que tenga, que me van a ayudar igual. Eso sí, primero se tienen que 

escuchar un audio de unos 10 minutos aproximadamente en el que yo, tranquilamente, 

explico los hechos acontecidos. Muchísimas gracias, chicas, por ser como sois, por 

vuestro apoyo incondicional y por vuestros consejos. 

Mis biosanis, Marina, Elena, Rosa, Cris, Judith Abarca, Judith Ortega, Natalia, Luis y 

Javier. Ellos también saben lo que implica una tesis y lo difícil que es trabajar en ciencia. 

Muchas gracias, chicos, por apoyarme estos años y por todos los momentos que hemos 

compartido. Sobre todo, a Rosa, que menos mal que nos teníamos la una a la otra en el 

mismo huso horario cuando estuvimos de estancia para apoyarnos mutuamente. 

Mis torrecillanos, María, Lidia, Andrea, Nuria, Raquel, Judith, Carlos, Víctor, Sergio y 

Miguel. Muchísimas gracias, a mi pequeña familia, por vuestro apoyo siempre, 

especialmente a Lidia, por esta maravillosa portada. 

Mi familia. Muchísimas gracias a mis padres, Asun y Jose, a mi hermano Iván, a mis 

abuelos/as, tíos/as y primos/as, por su apoyo, no solo en estos años de tesis, si no siempre, 

porque a pesar de que no entendían muy bien que hacía tantas horas en el laboratorio o 

porque estaba tanto tiempo con el ordenador escribiendo la tesis, han hecho todo lo 

posible por ayudarme. Sobre todo, mi abuela Asu, que va a ser la que más se alegre de 

que, su nieta, la científica, haya terminado de escribir. Ahora cada vez que leen o ven por 

la tele algo sobre terapia CAR, se acuerdan de que está relacionado con mi trabajo y me 

mandan la noticia. 



 

 
 

Y finalmente, muchas gracias a Perfecto, mi psicólogo y médico de mis abuelos desde 

que llegaron a Madrid. Gracias por ayudarme a entender lo que implica trabajar en ciencia 

y que, aunque es sacrificado, intentar desarrollar nuevas terapias para mejorar la vida de 

los pacientes merece la pena. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

RESUMEN 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



RESUMEN 

 
 

Introducción 

El mieloma múltiple (MM) es una neoplasia caracterizada por la proliferación clonal de 

células plasmáticas patológicas, principalmente en la médula ósea. Constituye el 10% de 

todos los tumores hematológicos y el 1,8% de todas las neoplasias. La sintomatología 

clásica se engloba en las siglas CRAB, hipercalcemia, fallo renal, anemia y enfermedad 

lítica ósea. Además, los pacientes con este tumor presentan una severa inmunosupresión, 

especialmente en las fases finales de la enfermedad. Aunque en los últimos años se han 

desarrollado una multitud de terapias para el tratamiento del MM, sigue siendo una 

enfermedad incurable. Este escenario justifica la necesidad de generar nuevos fármacos 

con distintos mecanismos de acción, especialmente en el contexto del MM refractario o 

en recaída. 

La terapia con células CAR T autólogas ha revolucionado el tratamiento de tumores 

hematológicos y actualmente cuenta con 6 productos aprobados en el mercado, dos de 

ellos en MM. Sin embargo, su uso clínico conlleva toxicidades graves, dificultades de 

producción, un coste elevado y un control subóptimo de la enfermedad en una proporción 

destacable de pacientes. La generación de células CAR NK ha emergido como una 

estrategia más segura, con menores dificultades de producción y con un coste inferior. 

Además, la posibilidad de administración en contexto alogénico facilita la accesibilidad 

de los pacientes a este tratamiento. Sin embargo, estudios preclínicos y los primeros 

ensayos clínicos han mostrado que la inmunoterapia con células CAR NK también 

presenta una serie de impedimentos relacionados con una baja persistencia y eficacia 

limitada en el paciente, que pueden condicionar las respuestas clínicas.  

Por un lado, una de las estrategias para aumentar la eficacia es la disrupción de puntos de 

control inmunológicos. Entre ellos, el eje HLA-E/NKG2A juega un papel importante en 

la inhibición de la célula NK. HLA-E, una molécula de histocompatibilidad de clase I no 

clásica, se encuentra sobre-expresada en una gran cantidad de tumores, en los que actúa 

como factor pronóstico. Estudios preclínicos han mostrado que el bloqueo de NKG2A 

con anticuerpos neutralizantes en combinación o la supresión del gen que codifica para 

este receptor inhibitorio aumentan la eficacia de las células NK en otros tipos de cáncer. 

Por otro lado, a pesar de que las células NK pertenecen a la inmunidad innata, son capaces 

de adquirir propiedades de memoria en respuesta a virus, haptenos o citoquinas. En 

concreto, la breve exposición de células NK a IL-12, IL-15 e IL-18 induce cambios a 



RESUMEN 

 
 

nivel transcripcional, epigenético, metabólico e inmunofenotípico, que incrementan la 

actividad antitumoral de las células NK, su capacidad proliferativa y su persistencia in 

vivo. 

Estrategias en combinación para aumentar la eficacia mediante el bloqueo de puntos de 

control inmunes, así como para potenciar la capacidad citotóxica y la persistencia a través 

de la inducción de un fenotipo de memoria, podrían mejorar la actividad de las células 

CAR NK para el tratamiento del MM refractario o en recaída. 

Objetivos 

El objetivo principal de esta tesis es la generación de terapias CAR NK optimizadas en 

combinación para mejorar el tratamiento del MM refractario o en recaída: i) analizar la 

relevancia del eje HLA-E/NKG2A en MM; ii) potenciar la actividad anti-MM de las 

células NKG2D- y BCMA-CAR NK activadas y expandidas (NKAE) mediante la 

disrupción del eje HLA-E/NKG2A con un nuevo anticuerpo α-NKG2A; iii) incrementar 

la persistencia y eficacia de la inmunoterapia mediante la reprogramación de las células 

NKG2D-CAR NKAE hacia un fenotipo de memoria inducido por citoquinas. 

Resultados 

Nuestros resultados muestran que la expresión de HLA-E en células plasmáticas de 

pacientes de MM está incrementada con respecto a las de donantes sanos. Además, las 

células plasmáticas patológicas presentan una mayor expresión de HLA-E con respecto a 

células plasmáticas sanas y células CD34+ del mismo paciente, especialmente en 

progresión. Aunque la supervivencia libre de progresión es menor en pacientes con un 

ratio alto de expresión de HLA-E en células plasmáticas patológicas con respecto a 

células CD34+, no se observan diferencias significativas en este parámetro, ni en la 

supervivencia general, en nuestra cohorte de pacientes. 

La concentración de IFN-γ en el plasma de médula ósea de pacientes de MM, sin signos 

analíticos ni clínicos de infección, es mayor que la de donantes sanos. El IFN-γ aumenta 

la expresión de HLA-E en células de líneas de MM y células primarias de pacientes al 

diagnóstico o en progresión. El bortezomib, inhibidor del proteasoma administrado como 

tratamiento de primera línea en pacientes de MM, disminuye la expresión de HLA-E 

incluso cuando es aumentada con IFN-γ, en células sensibles a este fármaco, pero no en 

las células resistentes. 



RESUMEN 

 
 

Las células NK de los pacientes de MM presentan niveles incrementados del receptor 

inhibitorio de HLA-E, NKG2A, especialmente en pacientes en progresión. Sin embargo, 

el bloqueo de NKG2A con anticuerpos neutralizantes de ratón (Z199) o humano (BMS-

936815) en muestras primarias de pacientes con MM, no impacta en la eliminación de 

células plasmáticas, lo que sugiere la falta de eficacia de estos tratamientos en 

monoterapia. 

Durante la expansión y activación de las células NKG2D- o BCMA-CAR NKAE aumenta 

la expresión de NKG2A como respuesta compensatoria. Las células CAR NKAE 

producen altas concentraciones de IFN-γ en co-cultivo con líneas de MM, que a su vez 

aumenta la expresión de HLA-E como resultado de una retroalimentación positiva. La 

combinación de anticuerpos α-NKG2A con las terapias NKG2D- o BCMA-CAR NKAE 

aumenta significativamente la actividad antitumoral sobre líneas celulares de MM y 

células primarias, en cultivos en contexto nativo, sin toxicidad relevante sobre células 

CD34+ del mismo paciente o PBMCs de donantes sanos. Asimismo, este aumento de 

función de las células CAR NKAE pretratadas con anticuerpos α-NKG2A está 

relacionado con mayores niveles de degranulación y producción de IFN-γ, y también se 

asocia a un balance positivo de la señal intracelular, que se refleja en una mayor 

fosforilación de ZAP70/Syk, con respecto a las células pretratadas con los isotipos 

correspondientes. No obstante, el bloqueo de NKG2A no altera significativamente la 

expresión fenotípica de receptores activadores e inhibidores en la membrana de la célula 

CAR NKAE. En los modelos de ratón de MM diseminado y ortotópico, la administración 

intraperitoneal del anticuerpo de bloqueo BMS-986315 en los animales reduce el 

porcentaje de células NKG2D- o BCMA-CAR NKAE, lo que elimina la eficacia de la 

combinación. Sin embargo, la infusión de células NKG2D-CAR NKAE pretratadas con 

el anticuerpo neutralizante no afecta a la proporción de células efectoras y disminuye 

significativamente la carga tumoral en los ratones. 

Por otra parte, a partir de volúmenes reducidos de sangre periférica de donantes sanos se 

han podido generar células cytokine-induced memory-like (CIML)-NKAE con una 

expresión estable del CAR de NKG2D. La inducción de un fenotipo de memoria a través 

de la exposición breve de las células NKG2D-CAR NKAE a las citoquinas IL-12, IL-15 

e IL-18 resulta en una mayor citotoxicidad, degranulación, producción de IFN-γ 

intracelular y un aumento en la liberación de IFN-γ, perforina y granzima A solubles, 

cuando las células de memoria se co-cultivan con líneas celulares de MM. Además, las 



RESUMEN 

 
 

células NKG2D-CAR CIML-NKAE exhiben una mayor capacidad citotóxica sobre 

células primarias de pacientes de MM, sin afectar significativamente a la supervivencia 

de células CD34+ del mismo paciente o procedentes de cordón umbilical. Asimismo, esta 

población de memoria tiene una mayor capacidad proliferativa y presenta cambios a nivel 

epigenético, transcriptómico, metabólico y fenotípico, comparado con las células 

NKG2D-CAR NKAE. En el modelo de ratón de MM U-266 ffLucGFP, las células 

NKG2D-CAR CIML-NKAE han demostrado un mayor control del tumor in vivo, que se 

asocia a un aumento en la proporción de las células NKG2D-CAR CIML-NKAE en los 

ratones y no a una persistencia superior de esta población, porque no se observa un mayor 

sostenimiento de esta terapia frente a las células NKG2D-CAR NKAE. 

Finalmente, hemos demostrado in vitro que el pretratamiento de las células NKG2D-CAR 

CIML-NKAE con el anticuerpo de bloqueo de NKG2A supera la eficacia anti-MM de las 

células NKG2D-CAR NKAE neutralizadas con este anticuerpo. 

Conclusiones 

La sobreexpresión de HLA-E en células plasmáticas patológicas y de NKG2A en células 

NK de pacientes de MM en progresión sugiere un papel relevante del eje HLA-E/NKG2A 

en el desarrollo del tumor, aunque el tratamiento con el anticuerpo α-NKG2A, BMS-

986315, en monoterapia no restaura la actividad de la célula NK. 

Las células CAR NKAE destinadas a inmunoterapia sobre-expresan NKG2A, que 

aumenta la susceptibilidad a la inhibición por la célula plasmática de MM, por lo que la 

combinación de terapias CAR NKAE con anticuerpos α-NKG2A es capaz de superar la 

resistencia a MM mediada por HLA-E in vitro e in vivo. 

La reprogramación de memoria inducida por citoquinas en combinación con la terapia 

NKG2D-CAR NKAE supone la generación de un nuevo efector con mayor eficacia anti-

MM in vitro e in vivo comparado con las células NKG2D-CAR NKAE. 

Tomadas en conjunto, ambas estrategias de tratamiento pueden ser potencialmente 

utilizadas no solo para aumentar la vulnerabilidad de las células de MM refractario o en 

recaída, si no administradas en el contexto de otro tipo de tumores, fundamentalmente 

HLA-E positivos.

 

 



 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

ABSTRACT 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



ABSTRACT 

 
 

Introduction 

Multiple myeloma (MM) is a neoplasm characterized by the clonal proliferation of 

pathologic plasma cells (PC), mainly in the bone marrow (BM). It constitutes 10% of 

hematologic tumors and 1.8% of all neoplasms. Classic symptomatology is included in 

CRAB acronym, hypercalcemia, renal failure, anemia, and bone lytic lesions. Moreover, 

MM patients have severe immunosuppression, especially at the late stages of the illness. 

Although in the last decades, a plethora of therapies for MM treatment have been 

developed, it is still an incurable disease. This scenario justifies the unmet need to 

generate new drugs with different mechanisms of action, especially in the 

relapsed/refractory MM (RRMM) context. 

Autologous CAR T therapy has revolutionized the treatment of hematologic tumors and 

currently, there are 6 market-approved treatments, two of them in MM. However, CAR T 

cell clinical use implies severe toxicities, manufacturing challenges, a high cost and 

suboptimal control of the disease in a remarkable proportion of patients. CAR NK cell 

generation has emerged as a safer, less cumbersome production and cost-effective 

strategy. Furthermore, the possibility of administering CAR NK cells in allogeneic 

context facilitates the accessibility of every patient to this treatment. However, preclinical 

studies and first clinical trials have shown that CAR NK immunotherapy also has certain 

shortcomings, regarding the low persistence and limited efficacy in patients, which may 

affect clinical responses. 

On the one hand, one of the strategies to enhance efficacy is the disruption of immune 

checkpoints. In concrete, the HLA-E/NKG2A axis plays a key role in NK cell inhibition. 

HLA-E, a non-classical class I histocompatibility molecule, is overexpressed in a high 

number of tumors, acting as a prognostic factor. Preclinical studies have shown that 

NKG2A blockade with neutralizing Abs in combination or the deletion of the gene that 

encodes NKG2A increases the efficacy of NK cells in other types of cancer. 

On the other hand, although NK cells belong to the innate immunity, they are able to 

acquire memory properties in response to viruses, haptens or cytokines. In concrete, a 

brief exposure of NK cells to IL-12, IL-15, and IL-18 induces transcriptional, epigenetic, 

metabolic, and immunophenotypic changes, which increase the NK cell antitumor 

activity, proliferative capacity, and persistence in vivo. 



ABSTRACT 

 
 

Strategies in combination to augment the efficacy through immune checkpoint blockade, 

as well as to potentiate the cytotoxic capacity and persistence through memory-phenotype 

induction, could enhance CAR NK cell activity for RRMM. 

Objectives 

The main objective of this thesis is the generation of optimized therapeutic combinations 

for CAR NK cells to enhance RRMM treatment: i) to analyze the relevance of the HLA-

E/NKG2A axis in MM; ii) to potentiate the anti-MM activity of NKG2D- and BCMA-

CAR activated and expanded NK (NKAE) cells by disrupting the HLA-E/NKG2A axis 

with a novel α-NKG2A antibody; iii) to increase the immunotherapy persistence and 

efficacy by reprogramming NKG2D-CAR NKAE cells towards a cytokine-induced 

memory-like phenotype. 

Results 

Our results show that HLA-E expression on MM patient PC is increased with respect to 

healthy donor (HD) PC. Moreover, pathologic PC have higher HLA-E expression 

compared to normal PC and CD34+ cells from the same patient, especially at progression. 

Although PFS is lower in patients with a high HLA-E expression ratio with respect to 

CD34+ cells, no significant differences are observed regarding this parameter or OS in 

our patient cohort. 

BM plasma IFN-γ concentration from MM patients, without analytic or clinical signs of 

infection, is higher than HD’. IFN-γ increases HLA-E expression on MM cell lines and 

primary PC from patients at diagnosis or progression. Bortezomib (BTZ), a proteasome 

inhibitor administered in first-line for MM treatment, reduced HLA-E expression even if 

it is increased with IFN-γ in BTZ sensitive cells, but not in resistant cells. 

NK cells from MM patients have augmented levels of the HLA-E inhibitory receptor, 

NKG2A, especially in patients at progression. However, NKG2A blockade with 

neutralizing mouse (Z199) or human (BMS-986315) Abs over MM primary samples does 

not impact on PC elimination, suggesting the lack of efficacy of these treatments in 

monotherapy. 

Along NKG2D- or BCMA-CAR NKAE cell expansion and activation, the expression of 

NKG2A augments as a compensatory response. CAR NKAE cells produce high 

concentration of IFN-γ in co-culture with MM cells, which at the same time increases 



ABSTRACT 

 
 

HLA-E expression as positive feedback. Combination of α-NKG2A Abs with NKG2D- 

or BCMA-CAR NKAE cells significantly increases antitumor activity against MM cells 

and primary cells, in native context cultures, without producing relevant toxicity against 

CD34+ cells from the same patient or HD PBMCs. Furthermore, this enhanced function 

of CAR NKAE cells pretreated with α-NKG2A Abs is related to superior degranulation 

and IFN-γ levels, and it is also associated with a positive balance of the intracellular 

signaling, reflected by a higher ZAP70/Syk phosphorylation, compared to CAR NKAE 

cells pretreated with the corresponding isotypes. Nevertheless, NKG2A blockade does 

not alter activating or inhibiting receptor phenotypic expression on CAR NKAE cell 

surface. In disseminated and orthotopic MM-bearing mice, the intraperitoneal 

administration of BMS-986315 blocking Ab reduce NKG2D- or BCMA-CAR NKAE cell 

percentage, which eliminates the combination efficacy. However, pretreated NKG2D-

CAR NKAE cell with the neutralizing Ab infusion does not affect effector cell proportion 

and significantly reduce tumor burden in mice.  

Additionally, generation of cytokine-induced memory-like (CIML)-NKAE cells with a 

NKG2D-CAR stable expression is feasible starting from low volumes of HD PB. The 

cytokine-induced memory phenotype through a brief IL-12, IL-15, and IL-18 exposure 

on NKG2D-CAR NKAE cells provokes higher cytotoxicity, degranulation, intracellular 

IFN-γ production and augments IFN-γ, perforin and granzyme A release, when memory 

cells are co-cultured with MM cell lines. Moreover, NKG2D-CAR CIML-NKAE cells 

exhibit an increased cytotoxic capacity over primary MM cells, without significantly 

affecting CD34+ cell survival from the same patients or cord blood samples. Furthermore, 

this memory population has a higher proliferative capacity and experiment epigenetic, 

transcriptomic, metabolic and phenotypic changes, compared to NKG2D-CAR NKAE 

cells. In a U-266 ffLucGFP-bearing mouse model, NKG2D-CAR CIML-NKAE cells 

have demonstrated an increased tumor control in vivo, associated with a higher NKG2D-

CAR CIML-NKAE proportion and not with a superior persistence, as there is not a higher 

sustenance of this therapy compared to NKG2D-CAR NKAE cells. 

Finally, we have demonstrated that NKG2D-CAR CIML-NKAE cell pretreatment with a 

NKG2A blocking Ab improves α-NKG2A blocked NKG2D-CAR NKAE cell anti-MM 

efficacy. 

 



ABSTRACT 

 
 

Conclusions 

Overexpression of HLA-E on pathologic PC and NKG2A on NK cells from RRMM 

patients suggest a relevant role of HLA-E/NKG2A in tumor progression, although BMS-

986315 α-NKG2A Ab monotherapy treatment does not restore NK cell activity. 

CAR NKAE cells intended for immunotherapy overexpress NKG2A, augmenting their 

susceptibility to MM PC inhibition, thus CAR NKAE therapy combination with α-

NKG2A Abs is able to overcome HLA-E mediated MM resistance in vitro and in vivo. 

Cytokine-induced memory reprogramming in combination with the NKG2D-CAR 

NKAE therapy implies the generation of a new effector with higher anti-MM efficacy in 

vitro and in vivo, compared to NKG2D-CAR NKAE cells. 

Taken together, both treatment strategies can be potentially used not only to increase 

RRMM cells vulnerability, but also to be administered in other tumors, principally the 

HLA-E positive ones. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

INDEX 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



INDEX 

 
 

RESUMEN 

ABSTRACT 

1. INTRODUCTION ........................................................................................................ 1 

1.1. Multiple myeloma .................................................................................................. 3 

1.1.1. Definition and general features ....................................................................... 3 

1.1.2 Epidemiology ................................................................................................... 3 

1.1.3. Etiology ........................................................................................................... 4 

1.1.4. Physiopathogenesis ......................................................................................... 4 

1.1.5. Clinical development....................................................................................... 8 

1.1.6. Diagnosis ......................................................................................................... 9 

1.1.7. Staging and prognosis ................................................................................... 10 

1.1.8. Treatment for newly diagnosed MM (NDMM) patients ............................... 11 

1.1.9. Relapsed/refractory treatments in MM (RRMM) ......................................... 16 

1.1.10. RRMM immunosuppression ....................................................................... 19 

1.2. Immunotherapy in MM ........................................................................................ 22 

1.2.1. MM immunotherapeutic targets .................................................................... 23 

1.2.2. Monoclonal antibodies (mAbs) ..................................................................... 30 

1.2.3. Antibody-drug conjugate (ADC) .................................................................. 33 

1.2.4. Bispecific T-cell engagers (BiTEs) and bispecific antibodies (BiAbs) ........ 34 

1.2.5. Oncolytic viruses, vaccine peptides and marrow-infiltrating lymphocytes .. 36 

1.2.6. CAR adoptive immunotherapy...................................................................... 37 

1.2.7. CAR T immunotherapy in MM ..................................................................... 39 

1.3. CAR NK immunotherapy in MM ........................................................................ 43 

1.3.1. Unique biological properties of NK cells over T cells for CAR engineering 44 

1.3.2. NK cell sources for CAR NK therapy ........................................................... 49 

1.3.3. CAR NK preclinical and clinical studies in MM .......................................... 51 

1.4. Combination of CAR NK cells with other strategies in MM .............................. 58 

1.4.1. Strategies to enhance CAR NK cell efficacy ................................................ 58 

1.4.1.1. Immune checkpoint inhibitors ................................................................ 59 

1.4.1.1.1. Description and modulation of the HLA-E/NKG2A axis ................ 61 

1.4.1.1.2. Preclinical studies targeting NKG2A ............................................... 64 

1.4.1.1.3. Clinical trials with α-NKG2A neutralizing antibodies .................... 66 

1.4.2. Strategies to enhance CAR NK cell persistence ........................................... 68 

1.4.2.1 Cytokine-induced memory-like NK cells ................................................ 69 

2. HYPOTHESIS AND OBJECTIVES.......................................................................... 75 

2.1. Hypothesis............................................................................................................ 77 



INDEX 

 
 

2.2. Objectives ............................................................................................................ 78 

3. MATERIALS AND METHODS ............................................................................... 79 

3.1. Cell lines and primary samples ............................................................................ 81 

3.1.1. Cell lines ........................................................................................................ 81 

3.1.1.1. Cell lines modified in the laboratory ...................................................... 82 

3.1.2. Primary samples ............................................................................................ 82 

3.2. Expansion, culture and CAR transduction of activated and expanded NK cells 

(NKAE) ....................................................................................................................... 82 

3.2.1. Generation of CAR NKAE and CAR CIML-NKAE cells from PB samples 82 

3.2.2. CAR molecule design and transfer plasmid amplification............................ 84 

3.2.3. Production of 3rd generation lentivector supernatants ................................... 85 

3.2.4. Titration of lentivector supernatants ............................................................. 86 

3.2.5. Lentiviral transduction .................................................................................. 86 

3.2.6. Study of vector copy number (VCN) per cell in transduced NK cells .......... 87 

3.3. Multiparametric flow cytometry and cell sorting ................................................ 88 

3.3.1. Multiparametric flow cytometry ................................................................... 88 

3.3.1.1. Immunophenotype .................................................................................. 90 

3.3.2. U-266 and RPMI-8226 ffLucGFP cell line sorting ....................................... 91 

3.4. Cytotoxicity assays .............................................................................................. 92 

3.4.1. Calcein release cytotoxicity assay against MM cell lines ............................. 92 

3.4.2. Clonogenic cytotoxicity assay against MM cell lines ................................... 93 

3.4.3. Cytotoxicity assay against primary MM cells through flow cytometry ........ 93 

3.4.4. Degranulation assay ...................................................................................... 94 

3.5. Cytokine detection assays .................................................................................... 95 

3.5.1. Intracellular IFN-γ detection assay ............................................................... 95 

3.5.2. Soluble cytokines quantification assay ......................................................... 96 

3.6. Toxicity assays ..................................................................................................... 97 

3.6.1. Toxicity against PBMCs ............................................................................... 97 

3.6.2. Toxicity against CD34+ hematopoietic cells from CB .................................. 97 

3.7. Proliferation assays .............................................................................................. 98 

3.7.1. Cell cycle analysis ......................................................................................... 98 

3.7.2. Ki67 cell proliferation analysis ..................................................................... 98 

3.8. Metabolism assays ............................................................................................... 98 

3.8.1. MitoTracker staining ..................................................................................... 98 

3.8.2. Mitochondrial Oxidative Phosphorylation (OXPHOS) system function ...... 99 

3.9. Western Blot (WB) ............................................................................................ 101 

3.9.1. Detection of CAR expression on CAR NKAE cells ................................... 102 



INDEX 

 
 

3.9.2. NKG2A signaling pathway ......................................................................... 102 

3.10. Mass cytometry ................................................................................................ 102 

3.11. Transcriptomic analysis by RNA sequencing (RNA-seq) ............................... 103 

3.12. Bulk DNA methylation analysis ...................................................................... 107 

3.13. Mouse models .................................................................................................. 107 

3.13.1. In vivo tumor biodistribution analysis by bioluminescence imaging (BLI)

 ............................................................................................................................... 108 

3.13.2. CAR NK cell infiltration analysis in PB samples ..................................... 108 

3.13.3. Human population engraftment analysis in PB and BM at necropsy ....... 108 

3.14. Statistical analysis ............................................................................................ 109 

4. RESULTS ................................................................................................................. 111 

4.1. HLA-E/NKG2A immune checkpoint is upregulated in MM patients but 

α-NKG2A blocking antibodies are not efficient as monotherapy treatment ............ 113 

4.1.1. HLA-E is overexpressed across the progression of MM ............................ 113 

4.1.2. IFN-γ is highly expressed in BM MM patients and increases HLA-E levels, 

while BTZ decreases its expression on MM cells ................................................. 115 

4.1.3. NKG2A is overexpressed on MM cytotoxic NK cells at progression but α-

NKG2A treatment does not reactivate their cytotoxicity against tumor cells in 

native cultures ....................................................................................................... 118 

4.2. NKG2A blockade improves NKG2D-CAR and BCMA-CAR NKAE cell anti-

MM cytotoxic activity............................................................................................... 120 

4.2.1 NKG2D- and BCMA-CAR NKAE cells are efficiently generated .............. 120 

4.2.2. NKG2A is highly expressed on CAR NKAE cells ..................................... 124 

4.2.3. CAR NKAE cells produce a high concentration of IFN-γ and increase   

HLA-E tumor expression after co-culture with MM cells .................................... 127 

4.2.4. NKG2A blockade enhances CAR NKAE cell cytotoxicity against MM cell 

lines ....................................................................................................................... 128 

4.2.5. NKG2A blockade increases CAR NKAE degranulation levels and IFN-γ 

release during co-culture with MM cell lines........................................................ 131 

4.2.6. α-NKG2A treatment does not significantly affect CAR NKAE cell activation 

or inhibition marker expression, but increases ZAP70/Syk phosphorylation ....... 134 

4.2.7. Pretreatment with α-NKG2A Abs enhances the cytotoxicity of NKG2D- and 

BCMA-CAR NKAE therapies over MM primary cells without compromising 

CD34+ population .................................................................................................. 135 

4.2.8. Pretreatment of NKG2D-CAR NKAE cells with the BMS-986315 α-NKG2A 

blocking Ab increases survival in disseminated MM xenograft mouse model ..... 138 

4.3. NKG2D-CAR CIML-NKAE cells have superior efficacy against MM cells in 

vitro and in vivo ........................................................................................................ 149 

4.3.1. Generation of NKG2D-CAR NKAE and CIML-NKAE cells from HD PB 

samples is feasible ................................................................................................. 149 



INDEX 

 
 

4.3.2. NKG2D-CAR CIML-NKAE cells exhibit higher cytotoxicity and 

degranulation capacity against MM cell lines ....................................................... 152 

4.3.3. Cytotoxic cytokine production is increased in NKG2D-CAR CIML-NKAE 

cells........................................................................................................................ 154 

4.3.4. NKG2D-CAR CIML-NKAE cells have enhanced proliferation activity in 

vitro ....................................................................................................................... 156 

4.3.5. NKG2D-CAR CIML-NKAE cells have increased glycolysis .................... 157 

4.3.6. NKG2D-CAR NKAE and CIML-NKAE cells have a differential 

transcriptome and epigenome profile .................................................................... 158 

4.3.7. NKG2D-CAR NKAE and CIML-NKAE cells exhibit differential expression 

of extracellular markers ......................................................................................... 160 

4.3.8. NKG2D-CAR CIML-NKAE cells display higher cytotoxicity against 

primary MM cells with low toxicity over CD34+ hematopoietic stem cells ......... 161 

4.3.9. NKG2D-CAR CIML-NKAE cells display heightened antitumor efficacy in a 

disseminated MM xenograft mouse model ........................................................... 162 

4.3.10. NKG2A blockade enhances NKG2D-CAR CIML-NKAE cells cytotoxicity 

against MM cells in vitro ...................................................................................... 165 

5. DISCUSSION ........................................................................................................... 167 

5.1. HLA-E is overexpressed in pathologic plasma cells from RRMM patients 

compared to healthy populations .............................................................................. 170 

5.1.1. α-NKG2A Ab monotherapy does not restore immune effector activity in BM 

MM patient samples .............................................................................................. 172 

5.2. NKG2D- and BCMA-CAR NKAE cells are efficiently generated ................... 174 

5.3. The blockade of NKG2A enhances NKG2D- and BCMA-CAR NKAE efficacy 

against MM cells ....................................................................................................... 177 

5.3.1. Pretreatment of NKG2D-CAR NKAE cells with BMS-986315 α-NKG2A Ab 

significantly reduces tumor burden in vivo ........................................................... 181 

5.4. NKG2D-CAR CIML-NKAE cells exhibit a higher efficacy than NKG2D-CAR 

NKAE cells in vitro and in vivo ................................................................................ 184 

6. CONCLUSIONES .................................................................................................... 195 

7. CONCLUSIONS ...................................................................................................... 199 

8. ABBREVIATIONS .................................................................................................. 203 

9. BIBLIOGRAPHY .................................................................................................... 213 

 

 

 

 



INDEX 

 
 

LIST OF FIGURES 

Figure 1. Development of MM and genetic alteration events .......................................... 6 

Figure 2. MM physiopathogenesis ................................................................................... 8 

Figure 3. Schematic description of NDMM treatments ................................................. 12 

Figure 4. Bortezomib modulates HLA-E expression ..................................................... 15 

Figure 5. Current FDA-approved treatments for NDMM and RRMM patients ............ 17 

Figure 6. Immunosupression in MM .............................................................................. 22 

Figure 7. MM targets and associated immunotherapies ................................................. 23 

Figure 8. BCMA signaling ............................................................................................. 24 

Figure 9. NKG2D-L/NKG2D axis. ................................................................................ 26 

Figure 10. Mechanisms of action of mAbs..................................................................... 31 

Figure 11. Track record of CAR immune effectors in preclinical and clinical studies. . 38 

Figure 12. NK cell differentiation and maturation ......................................................... 45 

Figure 13. Main NK cell activating and inhibiting receptors ......................................... 48 

Figure 14. NK cell sources for immunotherapy ............................................................. 51 

Figure 15. Main strategies to enhance CAR NK cells efficacy and persistence. ........... 59 

Figure 16. Modulation and signaling of HLA-E/NKG2A axis. ..................................... 64 

Figure 17. CIML-NK cells display a different phenotype compared to conventional NK 

cells. ................................................................................................................................ 71 

Figure 18. CAR NKAE cell generation protocol ........................................................... 84 

Figure 19. Schematic representation of lentivector production. ..................................... 86 

Figure 20. Schematic representation of calcein release cytotoxicity assay. ................... 93 

Figure 21. Schematic representation of LEGENDplex™ assay ..................................... 96 

Figure 22. Schematic representation of Seahorse OCR test. ........................................ 100 

Figure 23. Schematic representation of Seahorse ECAR test ...................................... 101 

Figure 24. Schematic representation of RNA-seq protocol. ......................................... 105 

Figure 25. HLA-E is overexpressed on PC from MM patients, especially at progression, 

but it does not significantly correlate with a higher OS or PFS. .................................. 114 



INDEX 

 
 

Figure 26. BM plasma of MM patients has an increased concentration of IFN-γ, cytokine 

that, along with BTZ, modulates HLA-E expression. .................................................. 116 

Figure 27. Cytotoxic NK cells from MM patients at progression have increased NKG2A 

expression, but α-NKG2A Abs in monotherapy cannot restore NK cell activity against 

primary PC .................................................................................................................... 119 

Figure 29. NKAE cells are stable and efficiently transduced with lentivectors containing 

NKG2D- or BCMA-CAR. ............................................................................................ 124 

Figure 30. Cytokines used to expand NK or CAR NK cells ex vivo promote NKG2A 

expression ..................................................................................................................... 126 

Figure 31. HLA-E expression increases on MM cells after co-culture with CAR NKAE 

cells that release high concentrations of IFN-γ. ........................................................... 128 

Figure 32. NKG2D- and BCMA-CAR NKAE cells in combination with α-NKG2A 

blocking Abs exhibit higher cytotoxicity against MM cell lines, even after 24h-treatment 

of tumor cells with IFN-γ ............................................................................................. 130 

Figure 33. Higher degranulation capacity and IFN-γ release are exhibited by α-NKG2A-

pretreated NKG2D- and BCMA-CAR NKAE cells. .................................................... 133 

Figure 34. NKG2D-CAR NKAE cell immunophenotype is not markedly altered by 

NKG2A blockade but ZAP70/Syk phosphorylation is potentiated after tumor stimulation.

 ...................................................................................................................................... 134 

Figure 35. NKG2D- and BCMA-CAR NKAE cells pretreated with α-NKG2A Abs 

exhibit a higher antitumor activity against PC from MM patients, with low toxicity over 

CD34+ cells from the same patient and PBMCs from HD ........................................... 137 

Figure 36. Intraperitoneally infusion of BMS-986315 α-NKG2A Ab in mice decreases 

NKG2D- and BCMA-CAR NKAE cell percentages in vivo. ....................................... 141 

Figure 37. NKG2D- and BCMA-CAR NKAE cells persist for at least 29 days in NSG-

Tg(Hu-IL15) mice bearing U-266 ffLucGFP MM cells. ............................................. 143 

Figure 38. NKG2D-CAR NKAE cells pretreated with either the BMS-986315 Ab or its 

isotype significantly reduce tumor burden in NSG-Tg(Hu-IL15) mice bearing U-266 

ffLucGFP MM cells...................................................................................................... 145 

Figure 39. Lower doses of NKG2D- and BCMA-CAR NKAE cells can be also detected 

at least 29 days from infusion in NSG-Tg(Hu-IL15) mice bearing U-266 ffLucGFP MM.

 ...................................................................................................................................... 147 

Figure 40. BMS-986315 pretreatment on NKG2D-CAR NKAE cells significantly 

prolongs mouse survival in NSG-Tg(Hu-IL15) mice bearing U-266 ffLucGFP MM cells.

 ...................................................................................................................................... 149 

Figure 41. CIML-NKAE and NKAE cells are efficiently generated and stably transduced 

with lentivectors containing NKG2D-CAR. ................................................................ 151 



INDEX 

 
 

Figure 42. Anti-MM and degranulation activity is augmented in NKG2D-CAR CIML-

NKAE cells. .................................................................................................................. 153 

Figure 43. NKG2D-CAR CIML-NKAE cells have higher IFN-γ intracellular production 

and IFN-γ, perforin, sFasL and granzyme A release, compared to NKG2D-CAR NKAE 

cells ............................................................................................................................... 155 

Figure 44. Cytokine-induction of memory promotes proliferation activity and glycolytic 

state in CAR NKAE cells. ............................................................................................ 157 

Figure 45. NKG2D-CAR CIML-NKAE cells have a differential epigenetic and 

transcriptomic profile, compared to NKG2D-CAR NKAE cells. ................................ 159 

Figure 46. NKG2D-CAR CIML-NKAE cells have a higher expression of CD25, CD2 

and CD71, while display a reduced expression of NCRs, GLUT1, and CD57. ........... 161 

Figure 47. Cytokine-induction of memory in NKG2D-CAR NKAE cells significantly 

improves the lysis of PC from MM patients, without compromising CD34+ population 

from the same patients or isolated from CB samples. .................................................. 162 

Figure 48. The percentage of NKG2D-CAR CIML-NKAE cells in mouse PB is superior, 

compared to NKG2D-CAR NKAE cells at day 7 post-NK cell infusion. ................... 163 

Figure 49. NKG2D-CAR CIML-NKAE cells significantly prolong mice survival in NSG-

Tg(Hu-IL15) mice bearing U-266 ffLucGFP MM cells. ............................................. 164 

Figure 50. The blockade of NKG2A significantly augments NKG2D-CAR CIML-NKAE 

efficacy against MM cell lines...................................................................................... 166 

Figure 51. The blockade of NKG2A enhances NKG2D- and BCMA-CAR NKAE cells 

against MM cells. ......................................................................................................... 193 

Figure 52. The induction of memory properties improves NKG2D-CAR NKAE cells 

cytotoxicity against MM cells. ..................................................................................... 194 

 

 

 

 

 

 

 

 

 

 

 



INDEX 

 
 

LIST OF TABLES 

Table 1. R2-ISS characteristics ...................................................................................... 11 

Table 2. FDA-approved immunotherapies in clinical development for patients with 

RRMM. ........................................................................................................................... 40 

Table 3. Current clinical trials with CAR NK cells against hematologic tumors. ......... 54 

Table 4. Published results from oncohematologic CAR NK cell clinical trials. ............ 56 

Table 5. Current clinical trials using α-NKG2A antibodies for cancer treatment. ......... 68 

Table 6. Current clinical trials using CIML-NK cells against tumor malignancies. ...... 74 

Table 8. Primers used to amplify Psi and albumin sequences. ....................................... 87 

Table 9. Reagents and primary Abs used for flow cytometry. ....................................... 89 

Table 10. Description of Abs used for HLA-E expression on PC and NKG2C, NKG2A 

and PD-1 expression on CD8+ T cells and CD56+CD16+ NK cells. .............................. 91 

Table 11. Abs used for CyTOF experiment ................................................................. 103 

Table 12. Detailed description of all procedures carried out along RNA sequencing. 106 

Table 13. Characteristics of MM patients used in ex vivo experiments ....................... 136 

 

 

 

 

 

 

 

 

 



 

1 
 

 

 

 

 

 

 

 

 

 

 

1. INTRODUCTION 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

2 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



INTRODUCTION 

3 
 

1.1. Multiple myeloma 

1.1.1. Definition and general features 

Multiple myeloma (MM) is a clonal B cell malignancy, characterized by the accumulation 

of tumor plasma cells (PC), mainly in the bone marrow (BM) [1]. This tumor is called 

“myeloma” because it is generated in the BM, and “multiple” because it affects different 

parts of the axial skeleton, especially the vertebral column, ribs, pelvis and skull [2]. 

Malignant PC produce monoclonal immunoglobulins (Igs), also called M proteins. 

Monoclonal proteins can be a complete Ig or its κ or λ chains. The presence of these free 

light chains in urine is called Bence-Jones proteinuria. Around 15-20% of patients have 

light-chain MM, characterized by the secretion of κ or λ chains instead of a complete Ig 

in urine and/or blood [3]. 

MM can be developed from premalignant status, called monoclonal gammopathy of 

undetermined significance (MGUS), or asymptomatic MM, known as smoldering 

multiple myeloma (SMM). In late stages, MM can become independent from BM signals 

and exert to the bloodstream, although extramedullary disease can also appear at 

diagnosis. When 5% of PC are found in blood, it is called PC leukemia (PCL) [1, 4]. 

Clinical manifestations of this tumor are hypercalcemia, renal failure, bone lesions, 

anemia as well as recurrent infections, caused by the severe immunosuppression in the 

BM [1, 5].  

 

1.1.2 Epidemiology 

MM accounts for 10% of all hematologic tumors and it is the 2nd most common 

hematologic cancer, after non-Hodgkin lymphoma [1, 6].  

According to the American Cancer Society, during 2023, there is an estimation of 35,730 

new cases of MM (1.8% of all new cancer cases) and 12,590 deaths associated with this 

tumor (2.1% of all cancer deaths) in the United States. The 5-year relative survival of 

MM calculated between 2013-2019 is 59.8%, the median age at diagnosis is 69 years 

(31.9% of MM patients are diagnosed at 65-74 years) and the median age of death is 75 

[6, 7]. 



INTRODUCTION 

4 
 

In Spain, the Asociación Española Contra el Cáncer estimated 3,251 MM cases in 2021 

[8]. 

Generally, MM is slightly more common in men than women [9] and more than 64% of 

patients are diagnosed over the age of 65 (less than 4% of MM patients are below 44) [7]. 

In addition, African-American people have a 2-fold MM incidence and a 2- to 3-fold 

MGUS prevalence compared to Caucasians [10, 11].  

 

1.1.3. Etiology 

MM etiology is still unknown. However, environmental and hereditary factors have been 

described to influence MM development.  

Environmental factors such as radiation [12-14], products used in farming (herbicides, 

pesticides and organic solvents, like benzene) [15, 16] and hair dyes [17] have been 

described to promote MGUS or MM development, although some studies reported these 

data are not conclusive [18]. 

Obesity has also been associated with increased MM incidence (1.11-fold) and mortality. 

However, other studies have suggested that obesity can prevent cancer patient risk of 

death due to cachexia [19].  

Recently, our group has identified 4 MM responder patients to HCV antiviral treatment, 

suggesting that those MM which are developed by chronic antigen-stimulation can be 

controlled, or even cured, by eradicating the initiating viral infection [20]. In line with 

this, Igs from MGUS and MM patients have been found to specifically bind lysolipids 

from Gaucher’s disease, implying another antigenic stimulation role in MM [21, 22]. 

Hereditary genetic factors can also influence MM appearance [23] and a compilation of 

studies has determined a 2- to 4-fold familial risk for MM in an autosomal dominant 

pattern [24, 25]. 

 

1.1.4. Physiopathogenesis 

B cells are generated in the BM and migrate to germinal centers due to antigen 

recognition. Through somatic hypermutation and antigen selection, B cells undergo 



INTRODUCTION 

5 
 

affinity maturation to secrete antibodies (Abs). After class switch recombination, these 

post-germinal B cells, called plasmablasts, egress to the bloodstream. Mutations or 

failures during these processes result in an accumulation of malignant clonal PC in the 

BM, which originate MM [26].  

MM can be developed from a premalignant status, called MGUS, or an asymptomatic 

neoplastic stage, commonly known as SMM, with an average risk of progression to MM 

of 1% and 10% by year, respectively, [11]. Moreover, between 1-2% of patients have 

extramedullary disease since the first examination and around 8% of MM patients may 

progress to PCL [1]. 

Primary genetic events normally occur in these previous MM stages (MGUS and SMM) 

and are related to hyperdiploidy and translocations (Fig. 1) [27]. 

MM can be classified in hyperdiploid (h-MM) or non-hyperdiploid MM (nh-MM). In h-

MM (approximately 55-60% of patients), there are trisomies in the odd chromosomes (3, 

5, 7, 9, 11, 15, 19 and/or 21) and in nh-MM, 70% of cases have mutations in the Ig heavy 

chain locus [28, 29]. Most common IgH translocations involves cyclin D, followed by 

WHSC1 and FGFR3 dysregulation [30]. 

Along MM progression, second hits, including monosomies (chromosomes 13, 14 and 

17), deletions, duplications (chromosome 1q), secondary translocations, recurrent point 

mutations or epigenetic alterations, appear. Deletions in several chromosomes affect 

tumor suppressor genes, among others: 17p (TP53), 1p (FAM46C/CDKN2C), 13q 

(BRCA2/RB1), 12p (CDKN1B), 16q (WWOX) and 11q (BIRC2/3/MMP) [27]. 

Moreover, oncogenes (MYC and RAS), cell cycle regulators (eg. CDKN2A and 

CDKN2C) and transcription factors genes involved in the NF-κB pathway, emerge as 

second anomalies [28]. Other recurrent point mutations are DIS3, BRAF, TRAF and RB1 

[27]. In the last MM stage, PCL, there is an accumulation of abnormalities that correlates 

with a lower overall survival (OS) [31]. In addition, epigenetics alterations are involved 

in MM progression, such as DNA methylation changes on cyclin-dependent kinase 

inhibitors, WNT and JAK/STAT signaling pathway and tumor suppressor genes; 

alterations in histone methylation (upregulation of MMSET domain gene) and tumor 

suppressor miRNAs inactivation [28, 32].  

 



INTRODUCTION 

6 
 

 

Figure 1. Development of MM and genetic alteration events. MGUS: Monoclonal gammopathy of 

uncertain significance; SMM: smoldering multiple myeloma. Extracted from Pinto et al. 2020 [27].  

 

Not only do genetic abnormalities trigger MM proliferation, but also BM 

microenvironment (soluble factors and accessory cells) participate in the proliferation, 

sustenance, and drug resistances of this tumor (Fig. 2). 

Initially, the interaction of CXCR4 and α4β1, present in malignant PC, with the soluble 

factor CXCL12 and VCAM-1, respectively, expressed by BM stromal cells (BMSCs), 

favors tumor cell homing and adhesion to the BM. [33]. Apart from these relations, 

BMSCs express proangiogenic molecules (eg. vascular endothelial growth factor 

(VEGF)) and unleash IL-6 production, through NF-κB signaling pathway activation, 

which enhances MM development [34, 35].  

Moreover, MM cells promote a dysregulation between bone formation, generated by 

osteoblasts, and bone destruction, characteristic of this tumor and performed by 

osteoclasts. Principally, both BMSC and PC produce high levels of the receptor activator 

of NF-κB ligand (RANKL) that binds RANK on the surface of osteoclasts, enhancing 

their activity. Under normal conditions, osteoblasts and BMSC produce osteoprotegerin 

(OPG) to compete with RANKL, impeding its interaction with RANK in the osteoclast 

surface and regulating bone destruction. In MM, tumor cell interactions with BMSCs 



INTRODUCTION 

7 
 

reduce OPG secretion, favoring RANKL-RANK binding. In addition, MM cells release 

Dickkopf 1 (DKK1), which inhibits osteoblast function and therefore, impairs OPG 

production, fostering osteoclast activity. Other factors as macrophage inflammatory 

protein-1α (MIP-1α), IL-3 and IL-6, mostly secreted by MM cells, favor this disbalance 

between osteoblasts and osteoclasts, enhancing tumor growth and bone loss [34, 36, 37]. 

All these interactions produce bone resorption, hypercalcemia, and pathological fractures. 

Thus, these patients can be treated with an antibody (Ab) against RANKL (denosumab) 

or bisphosphonates (pamidronate and zoledronic acid) to decrease bone destruction [38]. 

In late stages, tumor microenvironment (TME) becomes hypoxic. In this context, there is 

a downregulation of CXCL12 levels [36] and MM cells reduce the expression of the 

CXCL12 receptor, CXCR4, losing tumor cell interactions with stromal cells and favoring 

their egression to the bloodstream [33]. 



INTRODUCTION 

8 
 

 

Figure 2. MM physiopathogenesis. MM: multiple myeloma; BMSC: bone marrow stromal cell; VEGF: 

vascular endothelial growth factor; CXCR4: CXC chemokine receptor 4; CXCL12: CXC chemokine ligand 

12; VCAM-1: vascular cell adhesion protein 1; DKK1: Dickkopf 1; RANK: receptor activator for nuclear 

factor κ B. RANKL: RANK ligand; OPG: osteoprotegerin; IL: interleukin; IL-6R: IL-6 receptor. Adapted 

from Palumbo and Anderson, 2011 [35]. Created with BioRender.com. 

 

1.1.5. Clinical development 

The typical course of MM is characterized by periods of remission and relapse with 

increasingly shorter response periods and an increasing number of rescue treatments [39]. 

Most clinical manifestations are due to the accumulation of PC in BM and other tissues 

and the high release of monoclonal proteins. The emergence of the well-known CRAB 

symptoms confirms MM diagnosis. CRAB stands for hypercalcemia, renal failure, 

anemia and bone lesions [1].  



INTRODUCTION 

9 
 

Bone lytic lesions, which can be detected by imaging, generate fractures, vertebral 

collapse, spinal cord compression and hypercalcemia. Osteoclast activity produces bone 

destruction, being the main cause of morbidity. This activity correlates with tumor burden 

and MM prognosis [1, 40]. 

Due to this high osteoclastic activity and bone resorption, enhanced by MM cells, there 

is an increase in the levels of calcium in the bloodstream. In addition, there is an 

association between hypercalcemia and altered renal function, detected by a decreased 

glomerular filtration rate (GFR) [37, 41].  

In 19-30% of the MM patients, renal failure is resulting from the precipitation and 

accumulation of light chains from the monoclonal Igs secreted by MM cells. This 

accumulation induces the formation of urinary casts in the kidney tubules, causing cast 

nephropathy. Normally, renal impairment can be detected by high levels of serum 

creatinine or a low GFR, and high light chains serum levels can be a prognostic factor of 

this damage [41-43]. 

Normocytic and normochromic anemia is also a common symptom in MM patients (70-

85% of cases), characterized by low levels of hemoglobin in the blood, which correlates 

with tumor burden and a worse prognosis [43, 44]. Lately, it has been discovered that 

MM BM microenvironment inhibits hematopoietic stem cell proliferation, affecting their 

differentiation to erythrocytes and diminishing the levels of hemoglobin [44]. Moreover, 

this anemia can be produced by a relative erythropoietin (EPO) deficiency, caused by 

renal impairment [45]. 

Infections are the main cause of death related to MM [42]. Seventy-five percent of 

patients at diagnosis have immunoparesis, caused by the lack of one or more Igs [46], and 

a reduction of immune cells, which lack of function originate a severe 

immunosuppression in MM patients, increasing the risk of infections. In addition, typical 

MM treatments such as bortezomib (BTZ), melphalan and stem cell transplant induce 

neutropenia, which increases the risk of viral infections [47, 48]. 

 

1.1.6. Diagnosis  

Since 2014, MM has been diagnosed if the BM contains more than 10% of malignant PC 

and one CRAB symptom or myeloma-defining biomarker (SLiM criteria) or 



INTRODUCTION 

10 
 

plasmacytoma. SLiM criteria can be ≥ 60% malignant PC in the BM, ≥ 100 serum free 

light chain ratio, and more than one focal lesion (≥ 5mm) detected by magnetic resonance 

imaging (MRI) [43].  

MM clinical symptoms are similar to other illnesses; therefore it is normally discovered 

by chance in a routine blood test. First diagnostic testing usually comprises blood and 

urine analysis, including hemogram, biochemistry, and protein studies [43]. High calcium 

levels (>11 mg/dL or >1 mg/dL above normal limit), detection of monoclonal Igs in blood 

and urine by electrophoresis with immunofixation, increased serum β2-microglobulin, 

high creatinine (>2 mg/dL) and serum free light chain ratio (≥ 100) may indicate MM. In 

addition, an imaging study is normally performed, comprising a low-dose whole-body 

computed tomography (CT), MRI, or PET-CT to identify bone lesions [1, 43, 49-51]. 

If the results of these determinations suggest MM disease, a BM aspirate with cytogenetic 

analysis is carried out to assure the MM diagnosis [43]. 

BM analysis includes cytogenetics, the detection of translocations, trisomies, and 

deletions through fluorescent in situ hybridization (FISH), as well as flow cytometry 

determination of malignant PC. Common immunophenotype of MM cells is defined by 

CD138+ CD38+ CD56+ CD45+ CD117+ CD19- [1, 52-54]. Moreover, CD81 has been 

described as a new prognostic factor that correlates with MM development [55] and other 

authors have suggested that MM clonogenic cells are CD138- [27]. However, these 

discoveries should be confirmed in further studies. 

 

1.1.7. Staging and prognosis 

In 2005, the International Staging System (ISS) was proposed to stratify MM patients at 

diagnosis based on serum β2-microglobulin and albumin levels. ISS-I was determined by 

< 3.5 mg/L of serum β2-microglobulin and ≥ 3.5 g/dL of albumin; ISS-III when the serum 

β2-microglobulin was higher than 5.5 mg/L and ISS-II, not ISS stage I or III. In 2015, a 

revision of the ISS (R-ISS) was proposed including LDH values and high-risk 

cytogenetics, del(17p) and translocations t(4;14) and t(14;16), to the prior classification. 

Patients were then stratified as R-ISS-I when serum albumin > 3.5 g/dL, serum β2-

microglobulin < 3.5 mg/dL and no high-risk cytogenetics or LDH abnormal values were 

detected. R-ISS-3 comprised serum β2-microglobulin > 5.5 mg/L and high cytogenetics 



INTRODUCTION 

11 
 

risk or high LDH levels. Neither R-ISS-I nor R-ISS-III patients were stratified as R-ISS-

II [56]. However, a new revision of this R-ISS staging, the R2-ISS, was done in 2022 to 

stratify better these intermediate-risk patients, including the risk feature 1q+ and deleting 

t(14;16) as a staging factor [57, 58]. 

Currently, depending on the following features (ISS-III 1.5 points, ISS-II 1 point, del(17p) 

1 point, high lactate dehydrogenase 1 point, t(4;14) 1 point, and 1q+ 0.5 points), values 

are summed to calculate a new score. Then, patients are stratified in these new R2-ISS 

risk groups: low, low-intermediate, intermediate-high, and high. Thus, median OS and 

progression-free survival (PFS) values from these new groups define better the prognosis 

and outcome of MM patients (Table 1) [57].  

 

 

 

Table 1. R2-ISS characteristics. ISS: international staging system; OS: overall survival; PFS: progression-

free survival; LDH: lactate dehydrogenase. Data extracted from D’Agostino et al. 2022 [57] 

 

1.1.8. Treatment for newly diagnosed MM (NDMM) patients 

From mid-50s, the combination of alkylating agents with corticosteroids achieved a 

median OS of 7.5 months [59]. Nowadays, without taking into account the outcomes with 

adoptive therapies, the 5-year survival of MM patients has increased up to 59.8% [7] and 

the median OS is 6 years [60]. 

Currently, treatments for NDMM patients are based on autologous stem cell transplant 

(ASCT) eligibility and risk-stratification (Fig. 3). Fit for transplant patients are chosen 

based on their adequate physical conditions and lack of comorbidities [1]. 

 

Risk factor Score 

ISS-III 1.5 

ISS-II 1 

Del(17p) 1 

High LDH 1 

t(4;14) 1 

1q+ 0.5 

R2-ISS (risk) Score OS (months) PFS (months) 

I (low) 0 Not reached 68 

II (intermediate-low) 0.5-1 109.2 45.5 

III intermediate-

high) 
1.5-2.5 68.5 30.2 

IV (high) 3-5 37.9 19.9 



INTRODUCTION 

12 
 

 

Figure 3. Schematic description of NDMM treatments. D: Daratumumab; V or BTZ: Bortezomib; R: 

lenalidomide; d: dexamethasone; T: thalidomide; M: melphalan; P: prednisone. GEM: Grupo Español de 

Mieloma. Data extracted from Grupo de Estudio de Gammapatías Monoclonales de Castilla y León clinical 

practice guideline 2023 [61]. Created with BioRender.com. 

 

Patients eligible for ASCT receive 3-4 cycles of induction therapy (DVTd, DVRd, or 

VRd) prior to stem cell collection and after ASCT, lenalidomide, or BTZ plus 

lenalidomide for high-risk cytogenetic patients, maintenance. If patients are not fit for 

transplant, a combination of 3 or 4 drugs containing immunomodulatory drugs (IMIDs), 

proteasome inhibitors (PIs), daratumumab, alkylating agents, and/or corticoids are 

indefinitely administered [1, 5]. At a median follow-up of 35.4 months, DVTd plus ASCT 

median PFS was not reached and 73% of complete responses (CR) were achieved in the 

CASSIOPEIA study [62]. Patients ineligible for transplant and treated with DVMP in the 

ALCYONE study obtained a median PFS of 32.9 months, after 40.1-month median 

follow-up [63]. 

 

 

 



INTRODUCTION 

13 
 

Autologous stem cell transplant (ASCT) 

Within this first scenario, some reports have suggested that, despite new agents being 

available for MM, the ASCT role in the first-line remains indispensable for achieving 

deep responses in NDMM patients [64-66]. Furthermore, the treatment with triplet or 

quadruplet regimens before ASCT, as induction therapies, is key for the impressive results 

obtained after this regimen [67, 68]. In addition, tandem ASCT, a second ASCT in less 

than 6 months, benefits the PFS and OS of high-risk NDMM, although this strategy is not 

routinely recommended outside clinical trials [69]. 

ASCT also impacts on immune cell recovery. In particular, CD56bright NKG2A+ NK cells 

are first reconstituted after 13 days from ASCT, while T cell recovery may take up to one 

year. At that moment, cytotoxic CD8+ T cells outnumber the inhibitory Tregs, although 

their PD-1 expression is high [70, 71]  

 

Immunomodulatory drugs (IMiDs) 

The first-line drugs that most affect immune effectors are IMiDs. Thalidomide, a synthetic 

derivative of glutamic acid, was the first IMiD used for MM treatment [72]. However, its 

neurological toxicities have arisen the discovery of new generation analogues with less 

adverse events, such as lenalidomide and pomalidomide, and recently avadomide, 

iberdomide and mezigdomide. IMiDs main target is cereblon, a member of the E3 

ubiquitin ligase complex. Activation of this complex causes the polyubiquitination of the 

transcription factors IKZF1 (Ikaros) and IKZF3 (Aiolos) and their degradation via 

proteasome. The lack of these factors, highly involved in PC development and 

differentiation, triggers MM cell death. Furthermore, IMiDs possess anti-inflammatory 

and anti-angiogenic effects in the TME together with immunomodulation properties [73-

76]. In particular, lenalidomide increases NK cell function via antibody-dependent 

cellular cytotoxicity (ADCC) [77] and upregulates TRAIL expression on NK cells [78] 

and the NK cell receptor ligand expression, PVR and MICA, on MM cells [75]. 

Remarkably, this drug is able to lower NK cell activation threshold, which means that the 

recognition of low levels of NK cell ligands on MM cells is able to trigger NK cell 

antitumor activity [79]. Moreover, lenalidomide inhibits TNF-α, IL-1β, and IL-6 

production and boosts clonal T cell proliferation [80, 81]. Nevertheless, toxicities as 



INTRODUCTION 

14 
 

thromboembolic events, teratogenicity, neutropenia, and infections have been described 

after IMiD treatment [82]. 

 

Proteasome inhibitors (PIs) 

Proteasome degrades cell polyubiquitin proteins to recover amino acids that can be reused 

for protein generation [83]. MM cells produce high levels of Igs and therefore the correct 

functioning of the proteasome is critical for maintaining the homeostasis of these cells. 

Proteasome inhibitors, which target the β5 subunit of the proteasome, produce an 

accumulation of ubiquitinated or misfolded proteins, leading to ER stress and apoptosis, 

preferentially on MM cells [84]. BTZ, a dipeptide derived from boronic acid, belongs to 

first class PI, followed by carfilzomib and ixazomib (second generation), and new PI 

(oprozomib, delanzomib and marizomib) are under research. BTZ, ixazomib and 

delanzomib are reversible PI, while carfilzomib, marizomib and oprozomib irreversibly 

bind the β5 subunit of the proteasome. [83, 85]. As well as IMiDs, BTZ also possess 

immunomodulation features. Not only BTZ inhibits the NF-κB canonical signaling 

pathway and decreases the expression of anti-apoptotic proteins, promoting DC and T 

cell apoptosis, but also diminishes proinflammatory cytokine secretion and costimulatory 

molecule expression on T cells [86, 87]. In addition, proteasome can also degrade signal 

peptides derived from HLA class I molecules, which are loaded on HLA-I and HLA-E 

molecules and expressed on the cell surface [88]. Therefore, the inhibition of proteasome 

caused by BTZ downregulates the expression of HLA-I and HLA-E, sensitizing MM cells 

to NK cell-mediated recognition and killing (Fig. 4)  [85, 89]. Of note, BTZ also increases 

NKG2D ligands (NKG2D-L), DR5, and PVR expression on the MM cell membrane, 

enhancing NK cell cytotoxicity through NKG2D, TRAIL, and DNAM-1 receptors, 

respectively [85, 90]. Moreover, this inhibitor enhances FasL-dependent NK cell 

cytotoxicity against tumor cells [91]. Toxicities as neuropathy, Herpes virus reactivation, 

thrombocytopenia, neutropenia, cardiovascular events, especially after carfilzomib 

treatment, and risk of infections have been described after PI administration [47, 83, 92]. 

 



INTRODUCTION 

15 
 

 

Figure 4. Bortezomib modulates HLA-E expression. ER: endoplasmic reticulum. HLA: human 

leukocyte antigen 

 

Alkylating agents 

Melphalan, a DNA alkylating agent, has been used for MM treatment since 1960s [9], 

although high doses of this drug have many side effects, such as prolonged BM 

suppression, cardiotoxicity [93] and increased mutational burden on surviving MM cells 

[94]. Currently, the ALCYONE phase III study results have considered the use of oral 

melphalan at low doses in combination with Daratumumab, BTZ, and prednisone for 

NDMM transplant ineligible patients, as this combination achieved 45.7 months of PFS 

and 83.6% OS at 3 years [63]. 

 

Daratumumab 

Recently, daratumumab, an α-CD38 Ab, has been established as a first-line treatment due 

to the longer PFS obtained in combination with VTd [62] or VMP  [63] with weak adverse 



INTRODUCTION 

16 
 

events (higher risk of infection and hematological toxicities) [95, 96]. Daratumumab 

characteristics and mechanisms of action will be explained below. 

 

1.1.9. Relapsed/refractory treatments in MM (RRMM) 

RRMM comprises patients who have not achieved at least minimal response to prior line 

treatments (primary refractory myeloma), have reached minimal response at some point 

but have not responded to salvage therapy, within 60 days of the last treatment (relapsed 

or refractory), or after 60 days from last treatment (relapsed) [97].  

Lenalidomide-resistant patients are normally treated with second-generation IMiDs, as 

pomalidomide, in combination with other drugs as daratumumab, PIs, cyclophosphamide, 

or dexamethasone. If they are not refractory to lenalidomide, but they have been treated 

with BTZ, other combinations with second-generation PI, as KRd, or DRd, are 

considered. However, the OS in triple- or quad-refractory patients is 9.2 months and in 

penta-refractory (refractory to anti-CD38 Abs, 2 PIs and 2 IMiDs) is 5.6 months [98]. 

Thus, there is an unmet medical need to find new strategies to target MM (Fig. 5). Of 

note, these measurements do not include last updates from chimeric antigen receptor 

(CAR) T cells and bispecific antibody (BiAB) responses.  



INTRODUCTION 

17 
 

 

Figure 5. Current FDA-approved treatments for NDMM and RRMM patients. NDMM: newly 

diagnosed multiple myeloma; RRMM: relapsed or refractory multiple myeloma; CAR: chimeric antigen 

receptor. Created with BioRender.com. 

 

Selinexor is a nuclear export inhibitor that blocks exportin 1 (XPO1), whose expression 

has been correlated with a shorter PFS and OS in MM patients. This inhibitor blocks 

XPO1-mediated transport through the nuclear membrane, retaining tumor suppressor 

proteins, as p53, RB, IκB or BRCA1/2, which maintain the cell cycle arrest, and 

oncogenes mRNA (eg. c-Myc, D1 cyclin and BCL-2), impeding their translation [99]. 

Selinexor activity has been tested in combination with the abovementioned conventional 

drugs for MM, showing promise and synergistic results. Nevertheless, its toxicities, 

gastrointestinal and side effects, asthenia, diarrhea, nauseas, hyporexia, weight loss, 

hyponatremia, neurological effects, neutropenia, and infections, are the main drawbacks 

[99, 100]. Furthermore, Selinexor also displays immunomodulatory effects on NK cells. 

This drug downregulates the expression of HLA-E [101], which preferentially binds the 

NKG2A inhibitory receptor, and upregulates the expression of the activating receptor NK 



INTRODUCTION 

18 
 

cell ligands PVR, ULBP-1, -2 and -3 on MM cells. This way, Selinexor favors MM 

susceptibility to NK cell lysis [102, 103]. 

Melphalan flufenamide (Melflufen) is another alkylating peptide. When it enters in the 

MM cells, it takes advantage of the high concentration of peptidases and esterases to 

quickly release the melphalan and trigger PC apoptosis. Compared to melphalan, 

Melflufen enhances melphalan intracellular concentration, being a more effective anti-

MM treatment [104]. This drug is approved for RRMM patients in combination with 

dexamethasone. A recent phase III, the LIGHTHOUSE study, has tested the efficacy of 

melflufen in combination with daratumumab and dexamethasone. Median PFS of this 

group has not been reached yet and overall response rate (ORR) was superior to the 

control group (59% melflufen-daratumumab-dexamethasone vs 30% daratumumab). 

Hematological toxicities, including neutropenia, thrombocytopenia and anemia, were 

described [105]. 

Remarkably, B-cell maturation antigen (BCMA) and G protein-coupled receptor class C 

group 5 member D (GPRC5D)-targeted immune therapies have been recently authorized 

for RRMM patients with ≥4 prior lines of therapy by the FDA and EMA [106, 107]. Two 

BCMA-CAR T therapies (ide-cel and cilta-cel) [108, 109], two BCMA-bispecific T cell 

engager (BiTE) (teclistamab and elranatamab) [110, 111], as well as a GPRC5D BiAb 

(talquetamab) [112] have been approved for patients with a poor prognosis, achieving 

impressive outcomes that will be reported below. 

Another salvage therapy for RRMM patients is a second autologous transplant. Stem 

cells, previously collected from NDMM transplant eligible patients or collected after 

relapse, can be administered after 24 months from the first ASCT [113], although patients 

who relapsed after 36 months achieved better results (94% OS) [114]. 

Other drugs used for RRMM patients, but not approved by the FDA, are BCL-2 inhibitors 

(venetoclax), histone deacetylase (HDAC) inhibitors (panobinostat, vorinostat) and DNA 

hypomethylating drugs (5-azacitidine and decitabine) [28]. 

Venetoclax is used for patients harboring t(11;14) or high BCL-2 expression levels [115, 

116], and it can be administered in combination with dexamethasone (CANOVA trial 

NCT03539744) or BTZ and dexamethasone (BELLINI trial) [117] in clinical trials. 

Although these treatments enhance PFS in RRMM patients, diverse cytopenias and 

gastrointestinal side effects have been reported. 



INTRODUCTION 

19 
 

Although Panobinostat was withdrawn in 2022 by the FDA, it has been used in clinical 

trials in combination with PIs, IMiDs and steroids, obtaining an OS ranging from 10-19 

months and most common adverse effects are hematological and gastrointestinal [118, 

119].  

Despite all this therapeutic armamentarium, MM remains an incurable disease and 

patients relapse with increasingly shorter response periods. In this regard, there is an 

urgent need to find new treatments with novel mechanisms of action to surmount this 

refractory setting. 

 

1.1.10. RRMM immunosuppression 

During previous MM stages, MGUS and SMM, there is an immune equilibrium that is 

completely impaired along progression to active MM, especially in the RRMM context. 

Due to the predominance of M protein in the BM, polyclonal Ig production and other Ig 

isotypes are reduced, causing immunoparesis. Simultaneously, pathologic PC evade from 

immunosurveillance due to the rise of immunosuppressor cells and the enhanced levels 

of soluble factors. Both foster MM cell proliferation, while diminishing NK and T cell 

function [46, 120]. Moreover, PC clones from triple-refractory MM patients become more 

proliferative, being less dependent on inflammation factors as IL-6 and TGF-β and, at the 

same time, downregulate critical pathways, IFN and TNF signaling, to avoid 

immunosurveillance [121]. 

Despite the percentage of NK and cytotoxic T cells in the BM from MM patients is 

superior compared to HD, there is a dysfunction in these immune effector cells that 

hampers MM proliferation control (Fig. 6) [122]. On the one hand, the efficacy of 

activating receptors (DNAM-1, NKG2D, NKp30) is downregulated on MM patient 

immune effectors. DNAM-1 expression has been found reduced on NK cells from active 

MM patients compared to patients in remission, diminishing NK cell DNAM-1-mediated 

cytotoxicity [123]. In addition, NKG2D activating receptor activity is impaired in NK 

and CD8+ T cells due to the decrease of its ligand expression, MICA, along MM 

progression to late extramedullary stages [124]. Contrary to most tumors, MM cells 

augment their HLA class I, classical and non-classical molecules, HLA-E. However, 

defects in the antigen processing-presenting machinery disrupt MM cell TCR-mediated 

recognition by T cells. In line with this, NK cells are also inhibited via KIR-HLA-I and 



INTRODUCTION 

20 
 

NKG2A-HLA-E interactions [125, 126]. Remarkably, KIR expression is upregulated on 

NK cells from MM patients [127]. Furthermore, immune effector efficacy is reduced via 

immune checkpoints. PD-1 and CTLA-4 expression is augmented on T cells from MM 

patients [128] and their binding to their corresponding ligands, PD-L1, highly expressed 

on MM cells, and CD80/86, respectively, impair their antitumor efficacy [129]. 

Additionally, BM T cells have an increased CD4/CD8 ratio, which may justify their lower 

degranulation and decreased TNF-α and IFN-γ signaling [121]. 

Moreover, different interleukins, IL-6 and IL-10, cytokines, TGF-β and IFN-γ, together 

with other inhibitory soluble factors, soluble MICA, IDO, and adenosine (ADO), are 

highly expressed in the BM of MM patients, impairing immune effector function [33, 

130, 131]. MM cell as well as myeloid-derived suppressor cell (MDSC) and BMSC 

production of IL-6 not only enhances tumor proliferation, but also diminishes NK cell 

function [33, 132] and inhibits dendritic cell (DC) generation from CD34+ progenitors 

and therefore, T cell stimulation [133]. Similarly, IL-10 hampers effector T cell 

proliferation [33] and blocks the production of a potent IFN-γ inducer on NK cells, 

reducing one of the major mechanisms of these cells to exert their antitumor activity 

[134]. TGF-β strongly suppresses NK cell activity, as it diminishes IFN-γ production as 

well as NKG2D and NKp30 activating receptor and T-bet transcription factor expression 

[135]. Moreover, this cytokine is able to decrease T cell activation and proliferation, since 

it inhibits the presence of T cell costimulatory molecules on DCs, which allow antigen 

presentation [136]. IFN-γ is able to increase PD-L1 expression on tumor cells through 

JAK/STAT signaling, inducing T and NK cell exhaustion via PD-1/PD-L1 interaction 

[137]. In addition, an enhanced concentration of IDO generates a lower availability of 

tryptophan in the TME, as it is converted to kynurenine. The lack of tryptophan, an 

indispensable amino acid for T cell survival, causes the cell cycle arrest and apoptosis of 

these immune cells. At the same time, the obtained metabolite, kynurenine, impairs 

interactions among DCs, T, and NK cells [138, 139]. Furthermore, ADO release by MM 

cells, mediated by CD38, inhibits T and NK cell cytotoxicity [140]. Apart from these 

molecules, shedding of the NKG2D-L, MICA, from the surface of MM cells acts as 

decoy, neutralizing NKG2D function, and also fosters its internalization, impairing both 

NK and cytotoxic T cell function [131]. Of note, high soluble MICA levels correlate with 

a poor OS and PFS in MM patients [141].  



INTRODUCTION 

21 
 

Additionally, there is a recruitment of immunosuppressive cells in the BM from MM 

patients that fosters tumor cell proliferation, while hampers T and NK cell function. M2 

macrophages produce VEGF, which enhances tumor angiogenesis and other suppressor 

cytokines, such as IL-10 and ADO. Moreover, high infiltration of these type of 

macrophages has been correlated with a lower prognosis as well as with an augmented 

MDSCs proportion [33, 142]. MDSCs diminish T cell proliferation, together with DCs, 

and release different soluble factors that induce Tregs differentiation. Tregs are highly 

activated in this tumor due to the soluble factors released to the TME by MM cells. 

Moreover, cytokines produced by Tregs, such as IL-10 and TGF-β, attenuate effector T 

and NK cell function [33, 143]. Altogether, direct interaction with immune suppressive 

cells or the release of soluble factors by the latter or MM cells, enhances tumor 

proliferation and reduces endogenous T and NK cell function. This persistent 

immunosuppressive status suggests the need of new adoptive treatments with ex vivo 

stimulated immune effectors for MM patients. 

 

 

 



INTRODUCTION 

22 
 

 

Figure 6. Immunosupression in MM. MICA: major histocompatibility complex class I chain-related 

protein A; HLA: human leukocyte antigen; KIR: killer-cell immunoglobulin-like receptors; MM: multiple 

myeloma; ADAM10/17: a disintegrin and metallopeptidase 10/17; VEGF: vascular endothelial growth 

factor; IDO: indoleamine-pyrrole 2,3-dioxygenase; ADO: adenosine; TGF-β: transforming growth factor-

β; CTLA-4: cytotoxic T-lymphocyte antigen 4; PD-1: programmed cell death 1; TNF-α: tumor necrosis 

factor α; IFN-γ: interferon-γ; Treg: regulatory T cells; MDSC: myeloid-derived suppressor cells; DC: 

dendritic cells. Created with BioRender.com. 

 

1.2. Immunotherapy in MM 

Over the last decades, several approaches to enhance immune effector cell activity against 

tumors have been developed. In MM, monoclonal Abs (mAbs) targeting CD38 [96, 144] 

as well as BiAbs [110, 112] and BCMA-CAR T therapies [108, 109], have been already 

approved by the FDA, as it has been previously described. The most relevant feature of 

these therapies is choosing an appropriate target antigen, with high expression on 

malignant cells and low expression or absence on healthy tissues, to increase antitumor 

responses, avoiding on-target off-tumor toxicities (Fig. 7). 



INTRODUCTION 

23 
 

 

Figure 7. MM targets and associated immunotherapies. ADC: antibody-drug conjugate; BiAbs: 

bispecific antibodies; CAR: chimeric antigen receptor; mAb: monoclonal antibody. BCMA: B-cell 

maturation antigen; SLAM: signaling lymphocyte activation molecule family member 7; CD44v6: CD44 

variant domain 6; GPRC5D: G protein-coupled receptor class C group 5 member D; NKG2D-L: natural 

killer group 2 member D ligands; FcRH5: Fc receptor-homolog 5; ide-cel: idecabtagene vicleucel; cilta-

cel: ciltacabtagene autoleucel. Created with BioRender.com. 

 

1.2.1. MM immunotherapeutic targets  

• B-cell maturation antigen (BCMA; TNFRSF17; CD269) 

BCMA, also called CD269 or TNFRSF17, is a type III transmembrane glycoprotein and 

a member of the TNF-receptor superfamily. The expression profile of this protein is very 

limited. BCMA is present in mature B cells, plasmablasts and PC, and its expression is 

augmented on malignant PC along MM progression [145]. However, neurons and 

astrocytes placed in basal ganglia also have BCMA expression and few Parkinsonism 

cases have been reported after BCMA-CAR T therapies [146]. BCMA recognizes two 

ligands: B-cell activation factor (BAFF), which enhances MM cell adhesion to stromal 

cells, and, with higher affinity, a proliferation-inducing ligand (APRIL), which promotes 

MM cells survival and proliferation (Fig. 8). In addition, BAFF, normally expressed on 

monocytes, neutrophils and DCs, can bind BCMA, BAFF-R and the transmembrane 

activator and calcium modulator and cyclophilin ligand interactor (TACI) [147, 148]. 

However, APRIL, found on monocytes and tumors cells, only recognize TACI and, with 



INTRODUCTION 

24 
 

higher affinity, BCMA [147, 149]. Of note, APRIL and BAFF serum levels have been 

found increased on MM patients compared to HD [150]. 

BCMA binding to its ligands triggers pathways (PI3K/AKT and MAPK/ERK)  involved 

in the proliferation and survival of malignant PC and the sustenance of the 

immunosuppressive TME, through the induction of anti-apoptotic and 

immunosuppressive proteins in PC (PD-L1 and TGF-β) [147, 151, 152]. 

 

 

Figure 8. BCMA signaling. BAFF: B-cell activating factor; APRIL: a proliferation-inducing ligand; TACI: 

transmembrane activator and calcium modulator and cyclophilin ligand interactor; BCMA: B-cell 

maturation antigen; BAFF-R: BAFF receptor; PI3K: phosphoinositide 3-kinase; MAPK: mitogen-activated 

protein kinase; AKT: Ak strain transforming; ERK: extracellular signal-regulated kinase; NF-κB: nuclear 

factor κ B. Created with BioRender.com. 

 

Nevertheless, BCMA-targeted therapy efficacy can be reduced by the excessive levels of 

the soluble form of BCMA, found in the BM from MM patients, blocking their interaction 

with malignant PC [153]. In line with this, an augmented shedding of BCMA diminishes 



INTRODUCTION 

25 
 

the presence of this glycoprotein on the surface of MM cells, and the soluble BCMA acts 

as a decoy, avoiding tumor cell recognition by BCMA-targeted treatments. BCMA 

shedding is caused by γ-secretases, a multisubunit protease complex that mediates protein 

cleavage. Inhibition of these proteases recovers α-BCMA therapy activity [154]. 

 

• NKG2D-L 

NKG2D, a type II homodimer, belongs to the C-type lectin superfamily. NKG2D is a 

potent activating receptor of NK cells, but it is also expressed on the surface of CD8+ T 

cells, γδ T cells and NKT cells [155]. Under pathological conditions, as Crohn's disease 

[156] or cytomegalovirus infection [157], this receptor can be found on CD4+ T cells. 

NKG2D does not contain ITAM motifs to promote activating signals, but it forms a 

hexamer complex with DAP10, which contains the sequence YINM. This complex 

recruits PI3K/AKT and Grb2/Vav1 to trigger cytotoxicity when binding to NKG2D-L 

(Fig. 9) [158, 159]. 



INTRODUCTION 

26 
 

 

Figure 9. NKG2D-L/NKG2D axis. NF-κB: nuclear factor κ B; PIs: proteasome inhibitors; iHDACs: 

histone deacetylase inhibitors; NKG2D-L: NKG2D ligands; ULBPs: unique long 16 binding proteins; 

MICA/B: major histocompatibility complex class I chain-related protein A/B; ADAM10/17: a disintegrin 

and metallopeptidase 10/17; DAP10: DNAX-activating protein 10; PI3K: phosphoinositide 3-kinase; Grb2: 

growth factor receptor-bound protein 2; AKT: Ak strain transforming. Created with BioRender.com. 

 

NKG2D-L comprise 8 molecules: the MHC class I chain-related (MIC) proteins A and B, 

and the unique long 16 binding proteins (ULBP) 1 to 6. These ligands are normally 

induced by stress, including oxidative or thermal stress, heat shock or inflammation [160], 

DNA-damage [161], viral infection [162] or malignant transformation. Remarkably, at 

least one of these ligands is expressed on more than 85% of tumor cells, including stem-

like MM cells [124, 155, 163]. However, NKG2D-L expression has been also detected in 



INTRODUCTION 

27 
 

healthy tissues as gastrointestinal epithelium [164]. NKG2D-L expression is 

transcriptionally and post-transcriptionally regulated. The transcription factors E2F, NK-

κB, Sp family and p53 are involved in NKG2D-L expression. For example, p53 is able to 

amplify some NKG2D-L transcription and the drug-mediated induction of p53 has been 

associated with a higher expression of ULBP1 and 2. In addition, NF-κB induction 

enhances MICA expression [165]. At a protein level, some genotoxic drugs and 

chemotherapeutic agents, as PI and HDAC inhibitors, can increase NKG2D-L presence 

on the cell membrane [166]. However, the matrix metalloproteases (MMPs) ADAM10 

and ADAM17 participate in MICA, MICB, and ULBP2 shedding, augmenting the levels 

of these soluble forms in the TME, as we have described previously in RRMM 

immunosuppression, and impairing NKG2D-mediated cytotoxicity of NK and cytotoxic 

T cells [165]. Of note, supraphysiological levels of sMICA do not hamper NKG2D-CAR 

T cell efficacy [167]. 

 

• GPR5CD 

GPRC5D is an orphan receptor, which still does not have a recognized ligand, with seven 

transmembrane segments that belongs to the G protein-coupled receptor family. GPRC5D 

is highly expressed on MM cells and it is unlikely to be shed, being a suitable target for 

targeted therapies, although GPRC5D expression has been also detected in hard 

keratinized tissues and hair follicles [168, 169]. In addition, on-target off-tumor toxicities 

as mucosal toxicity [170] and nail- and skin-related events have been reported in more 

than 60% of MM patients, after treatment with GPRC5D BiAbs [112]. However, 

GPRC5D-CAR T cells have not induced alopecia in mice and cynomolgus monkeys 

[171]. Although in MM GPRC5D expression correlates with high-risk myeloma markers 

and could be related to tumor PC proliferation, its function is still unknown [168]. 

 

• Fc receptor-homolog 5 (FcRH5) 

FcRH5, another orphan receptor, also called FcRL5, IRTA2 or CD307, is encoded by one 

of the last 6 new genes that have been discovered from the Ig superfamily. FcRH5 is a 

differentiation antigen expressed on B cells, from pre-B cells and especially on their late 

stages, memory B cells and PC, after the development of antigen-primed B cells [172]. 



INTRODUCTION 

28 
 

Considering MM, this receptor is highly found in malignant PC, especially in patients 

with the gain 1q21, being associated to high-risk MM [173]. Although its ligands are still 

unknown, FcRH5 has been found to increase proliferation and possible toxicities are 

suggested to affect skin, eyes, and gastrointestinal tract [172, 174]. 

 

• CD38 

The ectoenzyme CD38 is a type II transmembrane protein that belongs to the ADP-ribosyl 

cyclase family. It is expressed on almost all hematopoietic cells: B, NK, and T cells and 

myeloid cells [175]. CD38 is high and uniformly expressed on MM cells compared to 

healthy tissues. However, this expression, positively controlled by STAT1 and negatively 

controlled by STAT3, can be downregulated due to selective pressure after CD38-targeted 

treatments [176, 177]. CD38 catalyzes two reactions: the cleavage of NAD+ into ADP-

ribose and the transglycosylation of NADP and nicotinic acid to obtain NAADP. 

Substrates from these reactions are involved in calcium release and ADO secretion, which 

participates in MM immunosuppression [177]. Moreover, CD38 has been described to 

also mediate cell migration and adhesion when binding CD31 or hyaluronic acid [178]. 

On-target off-tumor toxicities involve the reduction of B and T cell number [179] but 

remarkably, NK cell fratricide after α-CD38 Ab therapies [180]. 

 

• CD138 

CD138, also known as syndecan-1, as it belongs to the syndecan family, is a heparan 

sulfate proteoglycan. Although it can be found on squamous epithelial cells from liver, 

bladder, and gastrointestinal tract [181], this molecule is a lineage marker of PC and it is 

highly expressed on MM cells [182]. Of note, it is used to identify tumor PC by flow 

cytometry. However, in late stages, CD138 can be shed from MM cells. Both a high 

number of CD138low cells and high levels of soluble CD138 correlate with a worse 

prognosis [183-185]. CD138 can bind a multitude of proteins as integrins and 

extracellular matrix molecules, playing a key role in adhesion. Moreover, this 

proteoglycan binds growth factors [186], cytokines [187], and chemokines [188], and 

mediates wound healing, endocytosis and micropinocytosis [182]. In addition, it can bind 



INTRODUCTION 

29 
 

APRIL and TACI, promoting tumor growth and survival [189]. Possible on-target off-

tumor may affect the bladder, liver, or gastrointestinal tract tissues [181]. 

 

• CD44 variant domain 6 (CD44v6) 

CD44v6 is a spliced isoform of the CD44 protein, a group of type I transmembrane 

proteins. CD44 gene alternative splicing can generate different variants with different 

extracellular domains [190]. Ten different proteins can be generated between exon 5 and 

16 from CD44, and CD44v6 corresponds to alternative splicing in exon 11. This aberrant 

protein is commonly found in tumor cells, including MM, acute myeloid leukemia (AML) 

and pancreatic, colon, breast, and head/neck cancers, with low expression on activated T 

cells, monocytes, and keratinocytes and no expression on hematopoietic stem cells. 

Remarkably, it has been associated with a poor prognosis and metastasis [191], probably 

due to its expression on tumor stem cells [192]. Standard CD44 binds hyaluronic acid 

[193], collagen, matrix metalloproteinases [194] and E-selectin [195] to promote cell 

migration and tumor cell survival. Moreover, the addition of the exon 11-codified peptide 

to CD44 protein allows CD44v6 binding to osteopontin [196], hepatocyte growth factor 

(HGF) [197] and VEGF [198]. Due to its expression on keratinocytes, skin toxicities have 

been reported after CD44v6-targeted therapies [199], but not with CD44v6-CAR T cells 

[200]. Moreover, these treatments might reduce the number of CD44v6+ monocytes and 

macrophages [201], cells involved in cytokine release syndrome (CRS), diminishing the 

grade of this side effect [202]. 

 

• SLAMF7 

SLAMF7, CD319, CRACC or CS1 is a glycoprotein receptor, member of the signaling 

lymphocytic activation molecule and of the CD2 subset of the immunoglobulin 

superfamily [203]. SLAMF7 can be found in lymphocytes and mature DCs and it is 

expressed on healthy and pathologic PC [204]. SLAMF7 is a homophilic receptor, it 

interacts with itself when it is expressed on the cell membrane of other cells or with its 

soluble form, promoting PC proliferation [205]. Moreover, IMiD downregulation of 

Ikaros reduces both SLAMF7 forms, soluble and on the membrane, diminishing MM 



INTRODUCTION 

30 
 

progression [147]. As it has been described for CD38, SLAMF7 expression in NK and T 

cells promotes fratricide after SLAMF7-targeted immunotherapies [206, 207]. 

 

• CD19 

CD19 is a type I glycoprotein transmembrane protein, a member of the Ig superfamily, 

and a BCR co-receptor molecule. Similarly to FcRH5, it is expressed on all the B cell 

lineage, but it has a low expression in late stages, as PC. CD19 is involved in B cell 

maturation, differentiation and activation [208]. Although CD19-targeted therapies have 

shown great results against acute lymphocytic leukemia (ALL) [209] and B-cell 

lymphomas [210], CD19 expression on tumor PC is restricted to MM stem-like cells [208, 

211], characterized by a higher proliferation and drug resistance [212]. CD19-CAR T cell 

combinations with other BCMA therapies [213, 214] and ASCT, together with melphalan, 

have shown synergistic effects [212]. 

Other promising targets for MM treatment are CD229 [215], integrin β7 [216], TACI 

[217], and SEMA4a [218]. 

 

1.2.2. Monoclonal antibodies (mAbs) 

mAbs are an alternative approach for cancer patients, engineered to specifically target 

antigens that are overexpressed on tumor cells. Most mAbs can exert their antitumor 

activity through four mechanisms of action: ADCC, antibody-dependent cellular 

phagocytosis (ADCP), complement-dependent cytotoxicity (CDC), and direct tumor 

killing (Fig. 10) [219]. 

 



INTRODUCTION 

31 
 

 

Figure 10. Mechanisms of action of mAbs. CDC: complement-dependent cytotoxicity; ADCC: antibody-

dependent cellular cytotoxicity; ADCP: antibody-dependent cellular phagocytosis; mAb: monoclonal 

antibody; MAC: membrane attack complex. Created with BioRender.com. 

 

Daratumumab 

Daratumumab is a first-in-class α-CD38 human naked IgG1κ mAb [177]. This Ab was 

approved by the FDA in 2015 and is currently used as a first-line treatment in NDMM 

patients together with PI, IMiDs, and corticosteroids [1]. Apart from the mAb properties 

described above, Daratumumab also modulates the TME through the decrease of ADO 

production and the reduction of CD38+ Tregs, Bregs and MDSCs, while augmenting T 

cell proliferation [177, 220]. 

In RRMM patients, the last results of phase III trial POLLUX reported a higher OS of 

Daratumumab in combination with lenalidomide and dexamethasone, 67.6 months, 

compared to 51.8 months obtained after lenalidomide and dexamethasone, and PFS, 45 

vs 17.5 months, respectively [221]. Another phase III (CASTOR) has compared 

Daratumumab together with BTZ and dexamethasone versus BTZ and dexamethasone. 

Last update has described an enhanced OS (49.6 vs 38.5 months, respectively) and PFS 

(16.7 vs 7.1 months, respectively) [222]. 

However, acquired resistance to this treatment has been described. Firstly, one of the 

events that leads MM progression after Daratumumab is the loss of CD38 in MM cells. 

It has been described that CD38- clones persist after treatment or that there is a 



INTRODUCTION 

32 
 

downregulation within the first weeks and CD38 expression is recovered after 3-6 months 

[223]. Another challenge is that NK cells also express CD38 on their surface, being a 

Daratumumab target. Thus, NK cell number is significantly decreased, affecting ADCC 

function, and NK cells that remain are exhausted and lack CD16 and granzyme B 

production. Some groups have reported the use of CD38 KO exogenous NK cells in 

combination with Daratumumab to recover NK cell-mediated tumor killing [224, 225] 

Toxicities associated with these treatments are infusion reactions, neutropenia, anemia, 

lymphopenia, pneumonia, thrombocytopenia, and diarrhea [221, 222]. 

 

Isatuximab 

Another α-CD38 Ab commonly used in RRMM treatment is isatuximab. It is a naked 

chimeric IgG1 mAb approved by the FDA and EMA in 2020 [226, 227]. IKEMA phase 

III trial has reported a PFS of 35.7 months in the isatuximab, carfilzomib, and 

dexamethasone group, compared to the 19.2 months from the carfilzomib and 

dexamethasone arm [228]. Moreover, in the ICARIA study, the PFS of the isatuximab, 

pomalidomide and dexamethasone group was 17.5 months versus 12.9 months in the 

control group (pomalidomide and dexamethasone) [144].  

Most common adverse effects are related to infusion reactions along with respiratory 

infections [144, 229]. Moreover, reactivation of hepatitis B has been described in patients 

after daratumumab or isatuximab treatments [230]. 

New mAbs targeting CD38, as MOR202 and TAK-079, are still under research [231]. In 

addition, new approaches include the HexaBody-CD38 described by Hiemstra et al, 

which highly fosters CDC compared to Daratumumab [232] and its efficacy is being 

studied in a clinical trial (NCT04824794). 

 

Elotuzumab 

The α-SLAMF7 humanized naked IgG1κ mAb elotuzumab, was the first-in-class mAb 

approved for MM treatment in 2015 by the FDA and in 2016 by the EMA [226, 233]. 

Mechanisms of action of elotuzumab are principally ADCC and NK cell-mediated 

activation, but it is also capable of inhibiting MM cell adhesion to BMSCs [234, 235]. 



INTRODUCTION 

33 
 

However, as it has been described for other mAbs, elotuzumab treatment also reduced 

NK cell count due to the expression of CS1 in this immune subset [236]. 

ELOQUENT-3 phase II trial in RRMM patients demonstrated a significantly higher OS 

(29.8 months) in the elotuzumab plus pomalidomide and dexamethasone arm compared 

to the 13.8 months obtained by the pomalidomide and dexamethasone group [237]. 

Few adverse events have been reported after elotuzumab treatment: anemia, neutropenia, 

and respiratory infections [237]. 

Anti-MM efficacy of other mAbs, as siltuximab (α-IL-6) and SEA-BCMA (α-BCMA), is 

being studied in clinical trials [219].  

 

1.2.3. Antibody-drug conjugate (ADC) 

ADCs are recombinant mAbs frequently conjugated to small cytotoxic molecules, as a 

payload. When ADCs specifically target the antigens expressed on the cell surface, they 

are endocytosed and release the toxin after lysosomal cleavage, eliminating the tumor 

cells [219]. The payload can be tubulin inhibitors, DNA damaging agents, topoisomerase 

I inhibitors or RNA polymerase II inhibitors [238]. 

 

Belantamab mafodotin 

Belantamab Mafodotin is an afucosylated humanized α-BCMA Ab conjugated to 

monomethyl auristatin F (mafodotin), a microtubule polymerization blocker. It was the 

first-in-class ADC approved by the FDA in 2020 for RRMM treatment [239], after 

DREAMM-2 results. In this phase II study including 97 patients in the 2.5 mg/kg cohort, 

31% of them achieved OR and 2.9 months of PFS. Most common grade 3-4 toxicities 

were keratopathies, including reduced visual acuity, blurred vision, dry eye, photophobia, 

thrombocytopenia, and anemia [240]. However, the subsequent phase III trial 

(DREAMM-3) showed no improvement in PFS, when patients were treated with this 

ADC, compared to the pomalidomide and dexamethasone arm. For this reason, the FDA 

provisionally halted its approval in 2022 [241, 242]. In Spain this drug remains approved 

and there are still clinical trials with this ADC. In 2023, a national compilation of results 



INTRODUCTION 

34 
 

of its compassionate use has revealed a median OS rate of 41.8% and a median PFS of 

3.61 months [243]. 

Other promising BCMA ADCs as AMG 224 (NCT02561962) and CC-99712 

(NCT04036461) as well as α-CD138 (indatuximab) and an α-CD38 conjugated to IFN-

α2 (TAK-573) are being studied in clinical trials. However, ADCs targeting BCMA and 

other specificities are being discarded due to their low efficacy and severe adverse events, 

like lorvotuzumab mertansine (α-CD56), MEDI2228 (α-BCMA), DFRF4539A (α-

FcRH5), and azintuxizumab vedotin (α-SLAMF7) [244]. 

 

1.2.4. Bispecific T-cell engagers (BiTEs) and bispecific antibodies (BiAbs)  

BiTEs and BiAbs are designed to concomitantly bind immune cells and tumor cells. 

BiTEs are made by linking two scFv and BiAbs are recombinant Abs with a Fc region 

and two epitope-binding sites. One specific domain normally binds CD3 in T cells and 

the other, a tumor antigen on the surface of the malignant cells. In this way, these 

molecules redirect T cells to tumor cells and enhance the formation of immune synapses, 

favoring cancer cell elimination and T cell activation and expansion, independently of 

TCR-MHC recognition, antigen-presenting cells and costimulation [245]. Thus, these 

molecules represent an off-the-self therapy. 

 

Teclistamab 

Teclistamab (JNJ-64007957) is a humanized IgG4 BiAb, recently approved for RRMM 

patients. This BCMAxCD3 BiAb fosters T cell activation after tumor cell killing, 

exhibiting incredible efficacy against MM cells in vitro, even in whole blood cultures, 

and in vivo. Of note, soluble BCMA does not impair Teclistamab-mediated T cell 

cytotoxicity [246]. In the first clinical trial evaluating this molecule, MajesTEC-1, 

RRMM patients achieved an ORR of 63%, 39.4% of these responses were CR, and 

median PFS was 11.3 months. Most common toxicities were CRS, neutropenia, 

thrombocytopenia, neurotoxicity, and infections, in the 76% of patients [110]. In addition, 

new phase Ib and III clinical trials, MajesTEC-2 (NCT04722146) and MajesTEC-7 

(NCT05552222), respectively, are testing teclistamab efficacy compared to current 

triplets in RRMM patients.  



INTRODUCTION 

35 
 

Elranatamab 

Elranatamab (PF-06863135) is a humanized IgG2a BCMAxCD3 BiAb [245]. Last update 

of the phase II MagnetisMM-3 (NCT04649359) clinical trial shown 61% of ORR and 

35% CR. PFS has still not been reached. Neutropenia, anemia, CRS and infections were 

the most common adverse events described, slightly reduced in the biweekly dosing arm 

[111]. These outcomes gave the green light for its approval in 2023. 

The activity of other BCMAxCD3 BiAbs is being currently tested in clinical trials: 

• Linvoseltamab (REGN5458): an IgG4κ-based CD3xBCMA bispecific 

(NCT03761108). 

• Alnuctamab (CC-93269; EM901): a BiAb that recognizes two different epitopes 

of BCMA (NCT03486067). 

• TNB-383B (ABBV-383): a BiAb with 2 BCMA binding domains and low affinity 

for CD3 (NCT03933735). 

• Pacanalotamab (AMG 420, Bi 836909; NCT02514239) and Pavurutamab (AMG 

701; NCT03287908): BCMAxCD3 BiTEs. 

 

Talquetamab 

Talquetamab (JNJ-64407564) is a humanized BiAb that targets GPRC5D on PC and CD3 

on T cells. Preclinical studies highlighted the increased talquetamab efficacy over MM 

cells in vitro and in vivo, along with T cell proliferation [169], although BMSCs and Tregs 

impaired talquetamab-mediated T cell killing [247]. Clinical experience with talquetamab 

within MonumenTAL-1 study resulted in 73% ORR and 11.9 months of PFS with the 

higher dose. Although 79% of patients experienced CRS and 75% developed neutropenia, 

together with dysgeusia and skin-related adverse events, great efficacy outcomes 

promoted its approval in 2023 by the FDA and EMA. Moreover, a phase Ib/II clinical 

trial (RedirecTT-1) is studying the antitumor activity of talquetamab in combination with 

teclistamab (NCT04586426) and a phase Ib (TRIMM-2), its efficacy combined with 

daratumumab. First results demonstrated 84% of ORR (34% CR) and 78% of ORR (45% 

CR), respectively [112, 248]. 

Another relevant GPRC5DxCD3 BiAb is Fortintamig (RG6234; NCT04557150). 



INTRODUCTION 

36 
 

Cevostamab 

Cevostamab (BFCR4350A) is an IgG FcRH5xCD3 BiAb [245]. Li et al. have 

demonstrated specific and effective killing of Cevostamab against primary MM cells as 

well as against MM cell lines, showing no toxicity against FcRH5 negative cells. In 

addition, they showed that the BiAb specifically binds the membrane-proximal epitope 

(gD) of the FcRH5 protein [249]. Currently, there are 6 clinical trials ongoing. The phase 

I clinical trial (NCT03275103) reported a 54.5% ORR at the higher dose, although CRS 

within the first 24 hours occurred in more than 80% of the patients. Due to this quick and 

common CRS in almost all patients, tocilizumab, an α-IL-6R Ab, is administered prior to 

the BiAb infusion [250].  

Another promising BiAb is ISB 1342 (GBR 1342), the first CD38xCD3 BiAb engineered 

(NCT03309111). 

 

1.2.5. Oncolytic viruses, vaccine peptides and marrow-infiltrating lymphocytes  

Other immunotherapeutic strategies designed to target MM are oncolytic viruses, 

vaccines peptides and marrow-infiltrating lymphocytes (MILs). 

Oncolytic virotherapy consists of engineered self-replicating viruses that selectively 

infect tumor cells, causing cancer cell killing and antigen spreading, which favors the 

patient immune system reaction to the tumor. Oncolytic virus can recognize adhesion 

antigens or receptors, overexpressed on tumor cells, and, alterations in intracellular 

signaling pathways and metabolism favor their infection on malignant cells [251]. 

Moreover, the viral genome can be modified to increase tumor cell recognition (eg. 

adenovirus targeting E2F/Rb pathway), while decreasing pathogenicity, and to express 

anti-oncogenic or suicide genes. Different types of oncolytic viruses with distinct surface 

attachment receptors have been edited for this purpose for MM treatment. These oncolytic 

viruses can be intratumorally or systemically administered. Of note, the efficacy of 

reovirus, measles virus and VSV is currently being tested in clinical trials and first results 

have been modest [252, 253]. 

Peptide-based vaccines are used to enhance antigen recognition by the immune system. 

In MM, due to the immune cell exhaustion generated by the TME, most approaches have 

combined immunogenic peptides with DCs to potentiate antitumor responses. Some of 



INTRODUCTION 

37 
 

the current clinical trial target selected for this cancer are MUC1, RHAMM-R3, Bcl-2 

family, XBP1, CD138 and CS1, as well as idiotype Igs, suggesting promising results [219, 

254]. 

In addition, adoptive immunotherapy is being studied, including MILs. These are tumor-

specific T cells that reside in the BM [254]. Due to the high proportion of T cells that 

selectively target cancer cells, a clinical trial combining ex vivo activated autologous 

MILs with ASCT, and an allogenic myeloma vaccine has been carried out. This triplet 

combination achieved a median of 81.3% ORR, 18.6 months PFS and 31.3% CR 

(NCT01045460). 

Moreover, other adoptive approaches are activated and expanded NK cells [255], DCs 

[256], and engineered T and NK cells to express a CAR against tumor antigens, as it is 

described below. 

 

1.2.6. CAR adoptive immunotherapy 

Although the abovementioned treatments have obtained substantial results in clinical 

trials, CAR adoptive therapy has revolutionized the treatment of cancer patients, 

especially with hematologic tumors. Effector cells are frequently extracted through 

apheresis, activated, and transduced with CAR molecules. Then, these CAR cells are 

expanded and infused into the patient, to specifically kill target-positive cells, after 

lymphodepletion to avoid patient immune system rejection [254]. First studies were 

developed using T cells as CAR hosts, but there are other immune cells, as NK, iNKT, γδ 

T cells, macrophages and DCs that have been efficiently modified to express CAR 

sequences (Fig. 11) [257].  



INTRODUCTION 

38 
 

 

Figure 11. Track record of CAR immune effectors in preclinical and clinical studies. GVHD: graft 

versus host disease; CRS: cytokine release syndrome; ICANS: immune effector cell-associated 

neurotoxicity syndrome; DC: dendritic cell; CAR: chimeric antigen receptor; MHC: major 

histocompatibility complex; TCR: T cell receptor. Extracted from Daher and Rezvani, 2021 [257] 

 

CAR molecules contain an extracellular domain that recognizes an antigen on the surface 

of the tumor cells. This domain can be a fragment of a natural receptor, whose ligands are 

overexpressed on malignant cells, or the single chain variable fragment (scFv) from a 

mAb that specifically binds a tumor antigen. Then, it contains a hinge region, a 

transmembrane domain, and a signaling endodomain. Depending on the latter, CAR 

structures can be classified into 5 groups. First generation CAR molecules only contain 

the CD3ζ chain from the TCR, second generation also have a costimulatory domain, and 

third generation include two costimulatory domains and the CD3ζ. In addition, fourth 

generation CARs, also known as armored CARs, have been developed to incorporate a 

cytokine expression inductor into second generation structures or to express interleukins 

from a bicistronic transgene. Recently, next or fifth generation CARs have been designed. 

These molecules incorporate a signaling domain of an interleukin receptor (normally IL-

2) to foster JAK/STAT activation pathway. Most common costimulatory domains used 

are 4-1BB, CD28, OX40, CD27, CD40, and ICOS, among others [258]. 

 

 

 



INTRODUCTION 

39 
 

1.2.7. CAR T immunotherapy in MM 

CAR T cells are an innovative therapy that has reported impressive results against 

hematological malignancies as CD19+ lymphomas and B-cell lymphoblastic leukemias. 

CAR T cell therapy first approval by the FDA, Tisagenlecleucel, was in 2017 against 

CD19 for children and young adults with relapse or refractory B-ALL, obtaining 76% OS 

and 50% PFS at 12 months [259, 260] and to date, a total of 6 CAR T products have been 

globally approved for BCMA or CD19 positive tumors [261]. In addition, other two 

therapies have been approved by the Chinese National Medical Products Administration 

(NMPA), Relmacabtagene autoleucel, an α-CD19 CAR T, and Equecabtagene 

Autoleucel, an α-BCMA CAR T. Recently, the EMA has granted the α-CD19 

varnimcabtagene autoleucel the PRIME designation. In Spain, this CD19-CAR T therapy, 

also called ARI-0001, is administered under hospital exemption. These groundbreaking 

achievements have paved the way for the development of new CAR T treatments, not 

only for hematological malignancies, but also for solid tumors. The two CAR T therapies 

that have been recently authorized for clinical use in MM against BCMA are idecabtagene 

vicleucel (ide-cel) and ciltacabtagene autoleucel (cilta-cel). Ide-cel was approved in early 

2021 by the FDA and later by the EMA, due to KarMMa1 trial outcomes. Recently, the 

KarMMA-3 phase III clinical trial reported 71% of responses and a median PFS of 13.3 

months in the ide-cel arm compared to the 4.4 obtained by the control group, which 

included 5 different standard regimens combining Daratumumab, PI, IMiDs and 

corticosteroids for the treatment of RRMM patients. Toxicities grade 3 or 4 were observed 

in 93% of the patients, above all CRS and immune effector cell-associated neurotoxicity 

syndrome (ICANS) [108]. A year later, the FDA and EMA gave the green light for cilta-

cel, based on 89% OS and 77% 12-month progression free rate outcomes obtained in the 

CARTITUDE-1 clinical trial [262]. Lately, a total of 208 patients were enrolled in the 

CAR T therapy arm within the phase III CARTITUDE-4 clinical trial. OS was 84.6% in 

this group with 73.1% of CR and 60.6% MRD-. Nevertheless, 76.1% of patients suffered 

CRS, but only a small percentage developed neurotoxic adverse events (4.5% of patients) 

or severe CRS (1.1% of patients) [109]. Results from FDA-approved immunotherapies 

are summarized in Table 2. 

 



INTRODUCTION 

40 
 

Treatment 
Trial 

phase 

No. of 

patient

s 

ORR 

Median 

duration of 

response 

Median 

PFS 

Median 

OS 
CRS 

Neuroto

xicity 

Nonrelapse 

death 

  nº nº (%) months months months 
% of any grade 

(≥grade 3) 
nº (%) 

FDA-approved therapies 

Selinexor 2 122 32 (26) 4,4 3,7 8,6 NA NA 12 (10) 

Belantamab 

mafodotin 
2 97 31 (32) 11 2,8 13,7 NA NA 7 (7) 

Idecabtagene 

vicleucel 
2 128 94 (73) 10,7 8,8 24,8 

84 

(5) 
18 (3) 17 (14) 

Ciltacabtagene 

autoleucel 
1B-2 97 95 (98) NR NR NR 

95 

(5) 
22 (12) 16 (17) 

Teclistamab 1-2 165 
104 

(63) 
18,4 11,3 18,3 

72 

(1) 
14 (1) 19 (12) 

Talquetamab 

(GPRC5D-

CD3) 

1-2 74 49 (66) 
10.2-13 at 

two doses 
NA NA 

78 

(1) 
NA NA 

Elranatamab 2 94 57 (61) NA NA NA 
61 

(0) 
2 (0) 1 (1) 

BCMA-CD3 bispecific antibody in clinical development 

ABBV-383 2 94 57 (61) NA NA NA 
61 

(0) 
2 (0) 1 (1) 

Linvoseltamab 

(REGN5458) 
1 118 NA NA NA NA 

54 

(3) 
5 (NA) 6 (5) 

Pavurutamab 

(AMG701) 
1 73 37 (51) NA NA NA 

38 

(0) 
4 (0) 5 (7) 

Alnuctamab 

(CC-93269) 
1 30 13 (43) NA NA NA 

77 

(3) 
NA 3 (10) 

Non-BCMA-CD3 bispecific antibody in clinical development 

Cevostamab 

(FCRH5-CD3) 
1-2 161 64 (45) 11,5 NA NA 

81 

(1) 
14 (1) 6 (4) 

 

Table 2. FDA-approved immunotherapies in clinical development for patients with RRMM.  ORR: 

overall response rate; PFS: progression-free survival; OS: overall survival; CRS: cytokine release 

syndrome; FDA: Food and Drug Administration; GPRC5D: G protein-coupled receptor class C group 5 

member D; FCRH5: Fc receptor-homolog 5; BCMA: B-cell maturation antigen; NR: not reached. NA: not 

available. Adapted from Mailankody et al. 2022 [263]. Point-of-care these outcomes have to be carefully 

compared due to the differences in basal populations. 

 

Currently, there are a total of 195 clinical trials testing CAR T therapies in MM patients, 

most of them targeting BCMA. Moreover, GPRC5D, SLAMF7, CD38, FcRH5, NKG2D-

L, and CD138 specificities are also being studied, among others. In particular, two clinical 

trials (NCT03018405 and NCT02203825) have examined NKG2D-CAR T cell efficacy 



INTRODUCTION 

41 
 

against MM. THINK phase I study tested the activity of NKR-2, also known as CYAD-

01, which demonstrated great cytotoxicity and cytokine production against MM cell lines 

in vitro. Apart from tumor cells, these NKG2D-CAR T cells have unique properties, as 

they also target the tumor neovasculature, MDSCs, and Tregs present in the TME [264, 

265]. However, only 3 MM patients were enrolled in the THINK study and just one could 

be evaluated, who did not respond to the treatment, but treatment-related toxicities were 

low-grade, including CRS in 71% of patients [160]. The second study (NCT02203825) 

enrolled 5 MM patients and NKG2D-CAR T cells could be detected in 2 of them within 

the first two weeks. Nevertheless, efficacy was limited, no severe adverse events were 

reported, and all patients had to be treated with other therapies after relapse [266]. Of 

note, both treatments were first generation CAR (NKG2D ectodomain and CD3ζ 

signaling), no lymphodepletion was realized prior to NKG2D-CAR T therapy infusions 

and these MM patients expressed low levels of NKG2D-L. All these characteristics may 

have impacted on the poor NKG2D-CAR T cell efficacy reported. 

Contrary to CD19-CAR T therapy, BCMA-CAR T cell responses did not achieve plateaus 

[267]. This lack of stability in the responses might be caused by the following limitations. 

On the one hand, effector cell-dependent resistances, referring to T cells, include scFv 

immunogenicity and CAR T cell exhaustion. Most CAR scFv are derived from mice or 

are not completely human and the immune system can secrete human anti-mouse Abs 

(HAMAs). This fact limits CAR T cell expansion and persistence, and also annuls the 

effect of following doses. Some strategies such as changing scFv origin for fully-human 

sequences, applying a conditioning regimen before CAR T therapy to deplete endogenous 

lymphocytes, or using natural receptor- or ligand-based CARs instead of scFv, have been 

proposed [268]. 

Exhaustion is a reversible state that can be defined as lack of function, high expression of 

inhibitory receptors (PD-1, TIM-3, LAG-3, TIGIT, and CTLA-4), and T cell 

differentiation. This phenotype has been reported in most non-responders CAR T cell 

patients. CAR T cell exhaustion can be caused by the CAR structure. CD28 costimulatory 

domain within CAR molecule has been reported to enhance exhaustion compared to other 

signaling domains, such as 4-1BB or ICOS. In addition, accumulation of scFv on the cell 

membrane can result in tonic signaling and persistence antigen exposure also causes T 

cell exhaustion [269, 270]. 



INTRODUCTION 

42 
 

Both immunogenicity and exhaustion directly impact on CAR T cell persistence, being 

partially responsible for low stability in the responses. Moreover, TME, as it has been 

described previously, not only affects patients NK and T cells, but also CAR T cells, 

through the expression of inhibitory ligands, the release of cytokines (TGF-β, IL-10), or 

the interaction with immunosuppressive cells [269-271]. These factors inhibit CAR T cell 

activity and persistence, diminishing therapy efficacy. 

On the other hand, there are also target cell-dependent resistances that impair CAR T cell 

cytotoxicity. Principally, antigen loss or negative clone appearance are the main obstacles. 

Many tumors downregulate the CAR-target antigen, therefore they cannot be killed by 

the infused therapy. One solution is the use of Dual-CARs, carrying two different CARs 

in the same cell [272], or changing the specificity of the CAR T cell depending on the 

antigens expressed on the tumor cell, as Ruffo et al. have proposed [273]. This self-

labeling protein tag (SNAP)-synthetic Notch (synNotch) technology allows the binding 

of different Abs with diverse specificities to the CAR T cell, avoiding tumor cell escape. 

Moreover, target antigens such as BCMA or FcRH5 can be cleaved from the cell 

membrane and bind CAR T cells, impairing their cytotoxicity [154, 173]. The 

combination of γ-secretase inhibitors and BCMA-CAR T cells has been studied to avoid 

BCMA shedding and to augment this therapy efficacy [154]. Of note, NKG2D-CAR T 

therapy has been demonstrated in several reports to overcome this resistance, as soluble 

NKG2D ligands do not affect NKG2D-CAR T cell antitumor activity [167]. Furthermore, 

extramedullary disease and high-risk cytogenetics have been reported as risk factors for 

CAR T therapy failure [274]. 

In addition, new strategies are being developed to enhance CAR T cell trafficking to the 

tumor (especially in solid tumors), increase expansion and survival in vivo, augment 

cancer recognition (dual targeting, switch CAR specificity), control therapy growth 

(suicide genes) or target the immunosuppressive microenvironment [258]. Moreover, 

different studies have suggested the efficacy of a second CAR T treatment or BiAbs after 

BCMA-CAR T relapses, even with the same target-specificity, therefore sequential 

administration of immunotherapies enhances patient outcomes [275, 276]. 

One of the major limitations of CAR T therapies is treatment-related toxicity. Main 

adverse events are CRS, ICANS and on-target off-tumor toxicities. CRS, developed in 

almost all patients, cause fever and, in severe cases, hypotension, hypoxia and/or organ 

dysfunction, although these symptoms are normally controlled by α-IL-6R Ab infusions. 



INTRODUCTION 

43 
 

In most cases, ICANS is developed in patients who have already experienced CRS. This 

neurotoxicity mainly causes confusion, impaired motor skills, apraxia, dysphasia, 

aphasia, and, in severe cases, seizures, cerebral oedema, motor weakness or coma [277]. 

Other relevant toxicities are hemophagocytic lymphohistiocytosis, a second 

inflammatory state after CRS [278] and prolonged cytopenias that can persist for more 

than 90 days [279]. 

Related to manufacturing, autologous individualized CAR T therapies have a high cost, 

vein-to-vein time is quite long and there is a high rate of production failure, due to tumor 

contaminations or lack of expansion, as most patients have a high tumor burden or have 

been treated with multiple lines of treatment, respectively [280]. 

This scenario highlights the need to find alternative immune effectors to enhance CAR 

therapy persistence and efficacy, avoiding toxicities (CRS and ICANS), TME 

immunosuppression or antigen loss consequences, as the use of NK cells. 

 

1.3. CAR NK immunotherapy in MM 

NK cells have emerged as safer and universal effectors for CAR treatment. Most relevant 

advantage regarding NK cells as CAR effectors is the low probability of graft versus host 

disease (GvHD). Contrary to T cells [281], NK cells lack TCR, which means that its 

function is not restricted to HLA recognition, facilitating its use in an allogeneic context 

without engineering. This fact guarantees the accessibility of every patient to this therapy, 

as CAR-NK cells can be already prepared from different sources (eg. peripheral blood 

(PB), cord blood (CB)) and rapidly infused into the patient, reducing costs, time, and 

manufacturing failures [282, 283]. In addition, multiple infusions of CAR NK cells can 

be safely administered, as NK cells release a different cytokine profile (IFN-γ, TNF-α, 

and GM-CSF) and have a limited expansion capacity, diminishing CRS probability. 

Moreover, neurotoxicity has not been described in CAR NK therapies [284, 285]. 

Additionally, on-target off-tumor toxicities disappear when using CAR NK cells instead 

of CAR T cells, redirected to the same target [286]. This effect could be attributed to KIR 

restriction or the short half-life of mature NK cells in circulation (around 2 weeks). For 

this reason, suicide switch is not needed to control this therapy proliferation [287-289] 

but the addition of cytokines as IL-2 and IL-15 can improve CAR NK cell lifespan [290]. 



INTRODUCTION 

44 
 

NK cells as CAR effectors can overcome some CAR-T cell resistances, regarding CAR 

target antigen loss. Not only CAR NK cells can exert their antitumor activity through 

CAR signaling, but also through their native activating receptors, such as natural 

cytotoxicity receptors (NCRs), DNAM-1, and NKG2D, guaranteeing NK cell efficacy, 

even if the CAR target is downregulated [282]. 

 

1.3.1. Unique biological properties of NK cells over T cells for CAR engineering 

NK cells are innate cytotoxic lymphocytes that represent 5-15% of lymphocytes in PB. 

Their main function, as the first line of defense, is to identify and quickly eliminate 

abnormal, stressed, virally infected, senescent, and tumor cells, including metastatic cells 

[291-294]. NK cells are differentiated along 6 stages from lymphoid progenitor cells in 

the BM. After maturation in primary lymphoid organs, NK cells recirculate in the PB and 

tissues. Along differentiation, NK cells acquire activating receptor expression, as CD16 

and NKG2D, as well as perforin and granzymes to foster their cytolytic functions. 

Furthermore, NK cells increase KIR expression while downregulating CD62L, NKG2A, 

CCR7 and CXCR3 expression (Fig. 12) [295] 

Although the NK population is highly heterogeneous, it has been traditionally classified 

into two main groups based on their CD56 expression. Cytotoxic CD56dim NK cells, 

comprising 90-95% in PB, represent a more mature subset and express CD16, while 

regulatory CD56bright NK cells, around 5% in PB, are more immature and normally CD16-

. When NK cells are activated, CD56 expression increases, but this CD56bright population 

does not resemble the regulatory characteristics of this traditional subset [296]. 

Remarkably, NK cells lack CD3 and TCR expression, not being restricted to antigen 

presentation, as T cells [284, 285].  

 



INTRODUCTION 

45 
 

 

Figure 12. NK cell differentiation and maturation. Adapted from Cichocki et al. 2013 [295]. Created 

with BioRender.com. 

 

The action of these effector cells depends on a balance between germline-encoded 

activating and inhibiting receptors (Fig. 13). NK cells are educated to identify and tolerate 

the own cells due to the recognition of HLA class I molecules, through a process called 

licensing. NK cells are able to eliminate tumor cells that do not express HLA-I molecules 

(“non-self”) or have downregulated HLA-I expression (“missing-self”), as well as 

damaged or stressed cells through the recognition of activating-receptor ligands, 

overcoming the KIR-HLA inhibitory signal (“induced self”) [284, 292]. This NK cell 

tolerance is regulated by HLA receptors, KIRs, expressed on the NK cell surface. NK 

cells need the recognition of at least one self-inhibitory KIR binding HLA-I molecules to 

acquire functional education or licensing. When these effector cells lose KIR expression, 

also known as unlicensed NK cells, they become hyporesponsive [297]. 

KIRs are high polymorphic receptors, expressed on NK, NKT, and CD8+ T cells, which 

comprise 15 gene loci, located on chromosome 19, and their allelic variants [298, 299]. 

Their nomenclature depends on their structure and function. KIRs can possess a long 

cytoplasmic tail, normally containing ITIM motifs, which display an inhibiting response 

and a few KIRs have a short cytoplasmic tail, which normally includes ITAM motifs 

(activating responses). Of note, KIR2DL4 can conduct both responses, although it has 

ITIM motifs. To increase even more KIR complexity, KIRs can contain 2 or 3 Ig-like 



INTRODUCTION 

46 
 

domains, called KIR2D or KIR3D, respectively [285, 298]. Apart from KIRs, other 

inhibitory receptors have been identified on NK cells, including TIGIT, PD-1, CD96, 

KLRG1, and TIM-3.  

PD-1 is an inhibitory receptor with an extracellular IgV domain, expressed on T, B and 

NK cells, as well as on DCs under inflammatory conditions [282]. PD-1 binds to their 

ligands PD-L1, highly expressed on tumor cells, including MM [129], and PD-L2, also 

expressed on malignant cells and DCs, macrophages, and B cells [300, 301]. Both PD-L1 

and PD-L2 are transmembrane glycoproteins which contain IgC and IgV domains [282]. 

When binding to its ligands, PD-1 recruits SHP2 and SHP3 to exert their inhibitory 

activity, inducing effector cell exhaustion. However, although PD-1 is one of the main 

inhibitory signals in T cells, its role in NK cells has not been dilucidated yet [302]. 

TIGIT belongs to the PVR-like proteins family and contains an extracellular Ig variable 

domain and an intracellular ITIM and Ig tyrosine tail (ITT)-like motif. It shares homology 

with another inhibitory receptor, CD96 (TACTILE), both expressed on T and NK cells. 

These receptors recognize their ligands, PVR and Nectin-2, expressed on tumor cells and 

involved in cell adhesion and polarization, although TIGIT has more affinity for PVR 

than CD96 [303, 304]. Remarkably, TIGIT expression on NK cells that infiltrate the 

tumor has been found higher compared to those outside the tumor [285]. 

TIM-3, located in chromosome 5, belongs to the TIM family of immunoregulatory 

proteins, encoded by three genes, HAVCR1, HAVCR2 and TIMD4. When binding to 

galectin 9, it interacts with BAT3 through its cytoplasmic tail containing five tyrosines. 

TIM-3 can be found on T, NK, myeloid and mast cells and it is involved in the regulation 

of immune responses in cancer and autoimmunity [305]. Of note, its expression on NK 

cells is a prognostic factor in solid tumors [306] and it has been reported that PD-1+ and 

TIM-3+ NK cells have impaired IFN-γ and granzyme B production [285].  

KLRG1 is a transmembrane glycoprotein containing an ITIM domain. It is expressed on 

NK and T cells and their ligands, E-, N and R-cadherin, are involved in cell adhesion. 

This receptor inhibits AKT phosphorylation, impairing NK and T cell proliferation [307] 

and it has been associated with effector cell senescence [308]. 

NKG2 family members play a key role in the NK cell function balance. These receptors 

are members of the C-type lectin-like receptor superfamily and are expressed on NK cells 

and cytotoxic T cells [309]. Depending on the ITIM or ITAM motifs of each member, 



INTRODUCTION 

47 
 

they can be classified into activating (NKG2C, NKG2D, and NKG2E) or inhibiting 

(NKG2A and NKG2B) receptors. Moreover, NKG2A, NKG2C, and NKG2E form 

disulfide-linked heterodimers with the unfunctional CD94 protein, while NKG2D is a 

homodimer. NKG2A, NKG2C and NKG2E bind the HLA class-I non-classical molecule, 

HLA-E, although NKG2A has 6 times more affinity for this molecule than NKG2C [310]. 

The relevance of this inhibitory interaction will be later described. The activating NKG2D 

receptor specifically binds their ligands (MICA/B and ULBP1-6), as previously 

mentioned. 

NCRs comprise the activating receptors NKp30, NKp44, and NKp46, type I 

transmembrane proteins that belong to the Ig superfamily and are expressed on NK cells. 

They bind several ligands, such as BAG6, B7-H6, PCNA, and 21spe-MLL5, among 

others, as the complete panel of ligands has not been discovered yet. NKp30, NKp44 and 

NKp46 contain ITAM domains and besides, NKp30 and NKp46 are associated with 

FcεRIγ and/or CD3ζ intracellular domains, while NKp44 is associated with DAP12 to 

trigger NK cell killing function through Syk and ZAP70 phosphorylation [311]. Soluble 

NCR ligands, as well as the soluble factors IDO, TGF-β, and PGE2 can downregulate 

these receptors surface expression, diminishing NK cell activity [282]. 

The well-known CD16 receptor (FcγRIII), encoded by the FCGR3A gene, is associated 

with CD3ζ and FcRγ, forming homodimers or heterodimers. It has two isoforms, CD16A, 

highly expressed on NK cells, which is a transmembrane glycoprotein that recognizes Fc 

domains of IgG1 and IgG3, and CD16B mainly found on neutrophils. Binding of CD16A 

to Ab-coated cells principally fosters ADCC, although it is also involved in NK cell 

proliferation. However, this receptor can be cleaved and shed, or downregulated by 

ADAM17 metalloproteases or soluble factors, respectively, within TME, impairing NK 

cell ADCC function [291, 311]. 

Another activating and adhesion receptor is DNAM-1. It is a transmembrane glycoprotein 

that contains two Ig-like extracellular domains and three tyrosine residues in the 

cytoplasmic tail. DNAM-1, normally expressed on NK, T cells and other immune subsets 

such as monocytes, competes with TIGIT and CD96 for binding PVR and Nectin-2, 

although the TIGIT receptor has more affinity for these ligands. DNAM-1 binding to their 

ligands triggers Fyn phosphorylation and participates in IFN-γ secretion and NK 

antitumor activity [282, 312, 313]. 



INTRODUCTION 

48 
 

 

Figure 13. Main NK cell activating and inhibiting receptors. PD-L1: programmed cell death ligand 1; 

HLA: human leukocyte antigen; CLEC2D: C-type lectin domain family 2 member D; NCR: natural 

cytotoxicity receptors; NKG2D-L: NKG2D ligands; MICA/B: major histocompatibility complex class I 

chain-related protein A/B; ULBP: unique long 16 binding protein; 4-1BBL: 4-1BB ligand; KIR: killer-cell 

immunoglobulin-like receptor; DAP10: DNAX-activating protein 10; DNAM-1: DNAX accessory 

molecule-1; SIRPα: signal regulatory protein α; TIM-3: T cell immunoglobulin domain and mucin domain-

3; TIGIT: T cell immunoreceptor with Ig and ITIM domain; PD-1: programmed cell death 1. Adapted from 

Valeri et al. 2022 [280]. Created with BioRender.com. 

 

Based on this activating/inhibiting balance, NK cells exert their cytotoxicity through 4 

mechanisms: direct killing, mediated by perforin and granzyme release, cytokine 

production, which also implies an immunomodulatory function (eg. IFN-γ and TNF-α), 

ADCC, through CD16, and death-receptor pathways, mediated by TRAIL and FasL 

apoptosis induction [314, 315]. Therefore, compared to CAR T cells, which can only 

display their antitumor efficacy through CAR signaling and mainly release IL-6, IL-8 and 

IL-1β causing CRS, NK cells possess a variety of mechanisms and activating receptors 

to target tumor cells and secrete a different cytokine profile including IFN-γ, GM-CSF, 

TNF-α, and IL-3, which produces a low CRS probability. Therefore, the high expression 

of chemokine receptors and chemokine release, as CCL2, CCL3, CCL4, CCL5 and XCL1 

favor their trafficking to tumor niches. All these advantages enhance NK cell suitability 

as a CAR effector [287, 316, 317].  

 

 



INTRODUCTION 

49 
 

1.3.2. NK cell sources for CAR NK therapy 

Adoptive therapy using CAR NK cells can be generated from multiple sources: PB NK 

cells, CB NK cells, NK cell lines (as NK-92), decidual NK cells extracted from placenta 

[318] and NK cells derived from undifferentiated progenitors as hematopoietic 

stem/progenitor cells (HSPCs) from CB [319] or placenta [320], human embryonic stem 

cells (hESC), or induced pluripotent stem cells (iPSC) (Fig. 14). 

 

Peripheral blood (PB) NK cells 

Most CAR NK cell preclinical and clinical studies have been developed using NK cells 

derived from PB in an allogeneic context [321], where they represent 5-15% of all 

lymphocytes [294]. Although it is a low percentage, large volumes of PB samples can be 

obtained from apheresis and buffy coat samples and the same donor can be used for 

further product generations [296]. Then, PB mononuclear cells (PBMCs) are expanded 

with feeder cells along with cytokines (eg. IL-2 and IL-15) to promote NK cell activation 

and expansion, reaching 90% of NK cells in around 7 days [306, 322]. Of note, PB NK 

cells are very heterogeneous and display a more mature profile. These cells display a 

higher expression of activating receptors, as DNAM-1, NKG2D, and NCRs as well as 

CD16. In addition, PB NK cells highly express KIRs, adhesion molecules, as CD2 and 

CD62L, and the maturation antigen CD57 [293, 323]. Final balance results in an increased 

direct killing, ADCC and apoptosis-induction, through TRAIL [322]. However, NK cells 

from PB are refractory to engineering with lentivectors, due to several intrinsic antiviral 

mechanisms that hamper DNA insertion [293, 324]. In addition, it has been also reported 

that long NK cell cultures drive telomerase shortening, meaning that CAR NK therapy 

persistence could be decreased [293]. 

 

Umbilical CB NK cells 

NK cells derived from umbilical CB have been also used as CAR effectors in preclinical 

and clinical studies. Although the percentage of these heterogeneous NK cells is superior 

(around 20-30%) compared to PB NK cells, the total volume is limited to one unit of CB, 

much lower than apheresis, and it is not possible to obtain repeated samples for the same 

donor [296]. However, these cells have several advantages, as the immediate availability 



INTRODUCTION 

50 
 

of CB samples, better transduction efficiency with viral vectors, lower risk of GvHD, due 

to the reduced percentage of T cells in these samples, less KIR and NKG2A expression 

and a higher expression of CXCR4 and expansion capacity with feeder cells and 

cytokines. Moreover, CB NK cells can be cryopreserved, facilitating the infusion of 

multiple doses from the same donor. Contrary to PB NK cells, CB NK cells display a 

more immature phenotype and express less natural activating receptors as DNAM-1, 

NKG2C, and CD16, therefore they display a lower ADCC capacity [282, 293, 306, 321]. 

Nevertheless, when CB CAR NK cells are activated and expanded, they acquire a more 

mature phenotype and demonstrate similar antitumor activity, compared to PB CAR NK 

cells [325]. Of note, CB CAR NK cells engineered to express IL-15 have been detected 

in patients after 12 months from infusion [326].  

 

NK cell lines 

Described NK cell lines are NK-92, NKT, HANK-1, YT, KHYG-1, SNK-6, NK-YS and 

NK3.3. Except from NK3.3, which has been established from healthy donor (HD) blood, 

these cell lines have been generated from NK cell lymphomas and leukemias. To date, 

the only NK cell line approved by the FDA for clinical use is the NK-92 [321, 327]. The 

use of NK cell lines represents a universal and unlimited source of NK cells. Advantages 

of using NK-92 cells as CAR effector are their rapid expansion, easy genetic 

modifications, and homogeneity, although they do not express CD16, being incapable of 

exerting ADCC. On the contrary, NK-92 cell line main disadvantage is the necessity of 

irradiation prior to infusion, which reduces its cytotoxicity and persistence in vivo. Thus, 

the efficacy of these NK cell effectors in clinical studies is limited [293].  

 

Stem cell-derived NK cells 

Other sources are NK cells derived from CD34+ hematopoietic stem/progenitor cells 

(HSPCs), human embryonic stem cells (hESC), or induced pluripotent stem cells (iPSC). 

These stem cell-derived cells are homogenous, unlimited, due to the possibility of 

differentiating NK cells from the same donor multiple times, and have a low expression 

of KIRs and CD16. Since these cells display a more immature profile and require multiple 

differentiation steps, their manufacturing is longer than NK cells from PB or CB (around 



INTRODUCTION 

51 
 

3-5 weeks) and they do not recapitulate all the NK cell receptor repertoire [293, 296, 328, 

329]  

Preclinical and clinical studies developed with iPSC-derived NK cells have obtained 

clinical infusion numbers and have been easily genetically engineered [327]. 

 

Figure 14. NK cell sources for immunotherapy. CAR: chimeric antigen receptor; HPCs: hematopoietic 

progenitors; PBMCs: peripheral blood mononuclear cells; iPSCs: induced pluripotent stem cells. Created 

with BioRender.com. 

 

1.3.3. CAR NK preclinical and clinical studies in MM 

During the last decade, several groups have been studying CAR NK cell efficacy against 

hematological tumors, demonstrating the feasibility of this therapy as a new cancer 

treatment. This fact has fostered their use in clinical trials, where preliminary data have 

shown a higher safety and similar responses, compared to CAR T cells. 

Regarding MM, different CAR specificities, as well as NK cell sources, have been tested 

in preclinical studies. First strategies were conducted using the NK-92 cell line targeting 

CS1 [206] and CD138 [330]. Both groups demonstrated high efficacy in vitro and in vivo, 

paving the way for further studies using CAR NK cells in MM. Recently, NK-92 cells 

have been also engineered to express two specificities within the same cell, dual-CAR 



INTRODUCTION 

52 
 

NK cells: BCMA and CD19 [331], CD138 and CD19 [332] as well as BCMA and 

NKG2D, using different CAR costimulatory domains (unpublished results from our 

group). This strategy assures CAR NK cell efficacy even if one CAR-target antigen is 

downregulated. Moreover, Motais et al. have modified BCMA-CAR NK-92 cells to 

express soluble TRAIL, enhancing TRAIL-R expressing tumor cells apoptosis [333]. 



INTRODUCTION 

53 
 

Nº Status 
Study 

start 

Treatme

nt 

CAR 

construction 

NK cell 

source 
Phase Pat Location NCT 

1 Recruiting 2021 RRMM BCMA CAR CB I 27 China NCT05008536 

2 Unknown 2019 MM BCMA CAR NK-92 I/II 20 China NCT03940833 

3 Recruiting 2022 MM BCMA CAR Allogeneic I 19 China NCT05652530 

4 Recruiting 2023 
RRMM 

and PCL 
BCMA CAR Allogeneic I 18 China NCT06045091 

5 
Not yet 

recruiting 
2022 MM 

LUCAR-B68 

(BCMA CAR) 
Allogeneic I 34 China NCT05498545 

6 Recruiting 2021 MM 

FT576 (BCMA 

CAR) + 

Daratumumab 

iPSCs I 168 USA NCT05182073 

7 Terminated 2022 AML NKG2D CAR CB I 9 China NCT05247957 

8 Recruiting 2023 AML NKG2D CAR Unknown 
Unkno

wn 
30 China NCT05734898 

9 Recruiting 2020 

RR 

MDS, 

RR AML 

NKX101 

(NKG2D 

CAR/mbIL-15) 

PB I 90 USA NCT04623944 

10 Recruiting 2021 NHL 
CAR-NK019 

(CD19 CAR) 
PB I 25 China NCT04887012 

11 
Not yet 

recruiting 
2020 NHL CD19 CAR Unknown I 9 China NCT04639739 

12 Unknown 2019 

Refractor

y B-cell 

lymphom

a 

CD19 CAR Unknown I 9 Unknown NCT03690310 

13 Recruiting 2021 

ALL, 

CLL, 

NHL 

CD19 CAR CB I 27 China NCT04796675 

14 Unknown 2016 

B-cell 

leukemia 

and 

lymphom

a 

CD19 CAR NK-92 I/II 10 China NCT02892695 

15 Completed 2009 AML CD19 CAR PB I 14 USA NCT00995137 

16 Withdrawn 2019 DLBCL CD19 CAR 

Modified 

NK-92 

(haNK) 

I 0 USA NCT04052061 

17 Recruiting 2022 

ALL, 

CLL, 

NHL 

CD19 CAR PB I 15 China NCT05410041 

18 Completed 2022 ALL CD19 CAR Unknown I 2 China NCT05563545 

19 Withdrawn 2022 

B-cell 

leukemia 

and 

lymphom

a 

CD19 CAR Allogeneic I/II 0 China NCT05570188 

20 Recruiting 2023 DLBCL CD19 CAR Unknown I 12 China NCT05673447 

21 Recruiting 2022 

B-cell 

lymphobl

astic 

leukemia/

lymphom

a 

CD19 CAR Allogeneic I 30 China NCT05654038 

22 
Active, not 

recruiting 
2020 

B-cell 

lymphom

a 

FT596 (CD19 

CAR) 
iPSCs I 3 USA NCT04555811 

23 Recruiting 2022 

Relapsed 

or 

Refractor

y B-cell 

malignan

cies 

JD010 (CD19 

CAR) 
Allogeneic I 12 China NCT05645601 

24 Recruiting 2022 
RR B-

cell NHL 
CAR-NK019 
(CD19 CAR) 

CB I 48 China NCT05472558 

25 Recruiting 2023 

ALL, 

CLL, B-

cell 

lymphom

a 

JD001 (CD19 

CAR) 
Allogeneic I 12 China NCT05739227 

26 Recruiting 2021 

B-cell 

malignan

cies 

NKX019 (CD19 

CAR/mbIL-15) 
PB I 150 

USA and 

Australia 
NCT05020678 

27 Suspended 2013 ALL 
CD19 CAR + IL-

2 
PB I 20 

Singapor

e 
NCT01974479 



INTRODUCTION 

54 
 

 

Table 3. Current clinical trials with CAR NK cells against hematologic tumors. NCT: national clinical 

trial; RRMM: relapsed or refractory multiple myeloma; PCL: plasma cell leukemia; AML: acute myeloid 

leukemia; MDS: myelodysplastic syndrome; NHL: non-Hodgkin lymphoma; CLL: chronic lymphocytic 

leukemia; DLBCL: diffuse large B cell lymphoma; ALL: acute lymphocytic leukemia; ANLL: acute 

28 Completed 2017 

B-cell 

leukemia 

and 

lymphom

a 

iCasp9/CAR 

CD19/IL-15 + 

AP1903 

CB I/II 44 USA NCT03056339 

29 Withdrawn 2019 

B-cell 

lymphom

a 

CAR 

CD19/iCasp9/IL-

15 + Rituximab 

CB I/II 0 USA NCT03579927 

30 
Active, not 

recruiting 
2020 

CLL, B-

cell 

lymphom

a 

FT596 (CD19 

CAR) + 

Rituximab 

iPSCs I 98 USA NCT04245722 

31 Recruiting 2021 

B-ALL, 

B-cell 

lymphom

a 

QN-019a (CD19 

CAR) +/- 

Rituximab 

Allogeneic I 24 USA NCT05379647 

32 
Not yet 

recruiting 
2023 RR NHL 

CD19 t-haNK ± 

N-803 and 

rituximab 

NK-92 I 20 USA NCT05618925 

33 Recruiting 2021 
B-cell 

NHL 
CD19/IL-15 CAR CB II 242 USA NCT05020015 

34 Recruiting 2023 
CD19+ 

tumors 

CNTY-101 

(CD19/EGFR/ 

IL-15 CAR) + IL-

2 

iPSCs I 75 Unknown NCT05336409 

35 Unknown 2019 

Refractor

y B-cell 

lymphom

a 

CD19/CD22 CAR Unknown I 10 Unknown NCT03824964 

36 Recruiting 2023 
RR B-

cell NHL 
CD19/CD70 CAR CB I/II 48 China NCT05842707 

37 Recruiting 2022 
B-cell 

NHL 
CD19/CD70 CAR CB I 48 China NCT05667155 

38 Unknown 2016 

Lympho

mas or 

CD7+ 

leukemia

s 

CD7 CAR NK-92 I/II 10 China NCT02742727 

39 Unknown 2019 

Refractor

y B-cell 

lymphom

a 

CD22 CAR Unknown I 9 Unknown NCT03692767 

40 Recruiting 2021 AML CD33 CAR Unknown I 27 China NCT05008575 

41 Unknown 2016 
AML, 

ANLL 
CD33 CAR NK-92 I/II 10 China NCT02944162 

42 Recruiting 2022 AML 
QN-023a (CD33 

CAR) 
iPSC I 18 China NCT05601466 

43 Recruiting 2022 AML 
QN-023a (CD33 

CAR) 
iPSC I 19 China NCT05665075 

44 Recruiting 2020 AML CD33/CLL1 CAR Unknown I 18 China NCT05215015 

45 
Not yet 

recruiting 
2023 AML CD33/CLL1 CAR iPSC I 102 China NCT05987696 

46 Recruiting 2023 AML CLL1 CAR iPSC I 24 China NCT06027853 

47 
Not yet 

recruiting 
2023 

Hematolo

gical 

malignan

cies 

CD5/IL-15 CAR CB I/II 48 USA NCT05110742 

48 Recruiting 2022 

B-cell 

lymphom

a, MDS, 

AML 

CD70/IL-15 CAR CB I/II 94 USA NCT05092451 

49 Recruiting 2023 

RR 

leukemia, 

AML 

BPDCN 

CD123 CAR Unknown I/II 36 China NCT06006403 

50 Recruiting 2022 RR AML 
JD023 (CD123 

CAR) 
Allogeneic I 12 China NCT05574608 



INTRODUCTION 

55 
 

nonlymphocytic leukemia; BPDCN: blastic plasmacytoid dendritic cell neoplasm; BCMA: B-cell 

maturation antigen; CB: cord blood; iPSC: induced pluripotent stem cell; PB: peripheral blood; CAR: 

chimeric antigen receptor; Modified from Valeri et al. 2022 [280]. 

  

CAR NK cells have been also generated from primary cells obtained from PB, CB, and 

iPSCs for MM treatment. Our laboratory has engineered PB NKG2D-CAR NK cells that 

efficiently target MM cells in vitro and in vivo [155] and Golubovskaya et al. have 

developed PB BCMA-CAR NK cells, which have shown enhanced cytotoxicity and 

cytolytic cytokine production against MM cell lines [334]. Moreover, different 

approaches have been conducted to enhance PB CAR NK cell efficacy. Ng et al. have 

modified BCMA-CAR NK cells including CXCR4 to increase their migration to the BM 

[335]. Moreover, Wang et al. have incorporated the inducible MyD88/CD40 protein 

switch coupled with ectopic IL-15 to foster BCMA-CAR NK cell proliferation and anti-

MM killing [336]. Karvouni et al., in order to avoid CD38-CAR NK cells fratricide, as 

NK cells also express CD38, have carried out the knock out of CD38 in these cells using 

CRISPR-Cas9 technology [337]. The same strategy was conducted by Brophy et al. using 

CD38-CAR NK cells from CB [338]. In addition, our group has expertise in CB CAR 

NK cells, generating NKG2D- or BCMA-CAR NK cells that have shown efficacy in 

killing MM cells (unpublished results). Regarding NK cells derived from iPSC, FT555 

therapy, which co-targets GPRC5D and CD38, has been used against MM cells in 

combination with daratumumab [339] and Cichocki et al. have introduced 4 modifications 

into iNK cells to augment their efficacy and proliferation, avoiding fratricide: a BCMA-

CAR, a non-cleavable version of CD16, a membrane-bound IL-15/IL-15R fusion knock-

in, and a CD38 knock out. These CAR NK cells have been called FT576 [340]. 

Nowadays, there a total of 69 clinical trials testing CAR NK cell efficacy and safety 

against cancer cells, 50 in hematologic tumors (Table 3) and 6 of them in MM: BCMA-

CAR NK-92 cells (NCT03940833), CB BCMA-CAR NK cells (NCT05008536), iPSC-

derived FT576 therapy (described by Cichocki et al; NCT05182073) [340], and BCMA-

CAR NK cells (unknown NK cell origin; NCT05652530, NCT06045091 and 

NCT05498545). However, only the multiengineered FT576 product has reported its 

preclinical activity [340]. Thus, there is a lack of preclinical studies supporting the use of 

CAR NK cells in MM treatment. 



INTRODUCTION 

56 
 

Regarding hematological malignancies, most of these studies are phase I clinical trials 

and only 6 have updated their preliminary results (Table 4), one of them including MM 

patients. 

First-in-human CB CAR NK cells were reported by Rezvani’s group for the treatment of 

CD19+ non-Hodking’s lymphomas or CLL. CB NK cells were expanded ex vivo using 

K562-mb21-41BBL and engineered to express a CD19 CAR, IL-15 and an iCas9 safety 

switch. The phase I trial reported objective responses in 8 out of 11 patients (73%), 7 of 

them achieved CR. Of note, CAR NK cells persist at low levels for 12 months. 

CD19-CAR NK cells did not produce severe adverse events as CRS, neurotoxicity, 

hemophagocytic lympho-histiocytosis, or GvHD, despite the HLA mismatch. Moreover, 

the use of rimiducid, to activate the suicide switch, was not needed [326].  

Regarding MM, preliminary results of FT576 reported a 22% ORR in combination with 

daratumumab. Notably, this combination did not produce CRS, ICANs or GvHD. Due to 

the low efficacy, dose escalation is ongoing in both monotherapy and in combination with 

daratumumab [341].  

 

Product NCT Source 
Type of 

tumor 
Results Toxicities 

CD19 -CAR NK NCT03056339 CB NHL or CLL 

72.7% ORR 

(8/11) 

63,6% CR 

(7/11) 

Post-remission therapy was 

administered to five of the 

responding patients. No patients 

experienced CRS or neurotoxicity. 

CD33-CAR NK NCT05008575 CB AML 60% CR (6/10) No ICANS, 1 patients CRS grade 2 

BCMA-CAR NK 

(FT576) ± 

daratumumab 

NCT05182073 iPSCs MM 

1 VGPR in 

monotherapy 

1 PR and 1 MR 

in combination 

22% ORR 

There were no events of any grade 

of CRS, ICANS, or GvHD 

CD19-CAR NK 

(NKX019) 
NCT05020678 PB 

B-cell 

malignancies 
70% CR (7/10) 

Early safety profile supports 

outpatient administration and shows 

no neurotoxicity / ICANS, GvHD, 

or >Gr3 CRS 

NKG2D-CAR NK 

(NKX101) 
NCT04623944 PB 

AML and 

MDS 

67% CR (4/6; 2 

of them CRs are 

MRD-) 

No CRS, no ICANs or GvHD. 50% 

hematologic events (anemia and 

neutropenia) 

CD19-CAR NK 

(FT596) ± 

rituximab: 

Discontinued Jan 

2023 after 

promising initial 

data Aug 2021 

NCT04555811 iPSCs 
B-cell 

lymphoma 

71.4% OR 

(10/14; 7 of 

them CR) 

No Dose-limiting Toxicities, and 

No Adverse Events of Any Grade of 

ICANS or GVHD, were Observed; 

Two Low-grade Adverse Events of 

CRS were Reported 

 

Table 4. Published results from oncohematologic CAR NK cell clinical trials. BCMA: B-cell maturation 

antigen; NCT: national clinical trial; CB: cord blood; iPSCs: induced pluripotent stem cells; PB: peripheral 

blood; NHL: non-Hodgkin lymphoma; CLL: chronic lymphocytic leukemia; AML: acute myeloid 



INTRODUCTION 

57 
 

leukemia; MM: multiple myeloma: MDS: myelodysplastic syndrome; ORR: overall response rate; CR: 

complete response; VGPR: very good partial response; PR: partial response; MR: minimal response; MRD: 

minimal residual disease; CRS: cytokine release syndrome; ICANS: immune effector cell-associated 

neurotoxicity syndrome; GvHD: graft versus host disease. 

 

These data suggest that CAR NK cells are highly safe, they do not exert CAR T-related 

toxicities as CRS and ICANs, but their efficacy outcomes can be improved overcoming 

some of the following limitations. First, although there are multiple sources of NK cells 

available for CAR engineering, expansion protocols and viral transduction are 

challenging. Cell-cytokine priming using soluble cytokines or feeder cells engineered to 

express membrane-bound cytokines and/or T cell costimulators, generates a great amount 

of NK cells but its GMP-implementation for clinical use is still being developed. In 

addition, NK cells are refractory to viral transduction and low percentages of CAR+ NK 

cells may be also limiting their efficacy. Related to this low efficacy, NK cells can be 

killed through fratricide via Fas/FasL or due to self-recognition CAR target-ligand, when 

CAR NK cells are engineered to express a CAR, whose ligands are also expressed on 

CAR NK cells (eg. CD38 and SLAMF7). Similarly, CAR NK cells can acquire the 

expression of the CAR-target ligand from the target cell through trogocytosis, being 

susceptible to CAR NK cell killing [342]. 

Loss of efficacy can also occur when homing into the tumor bed is reduced due to the 

downregulation of chemokine receptors. In addition, once CAR NK cells reach the tumor, 

their function can be impaired via cell-cell contact with immunosuppressive cells or the 

interaction with soluble factors released by both, tumor cells and immunosuppressive 

cells in the TME. Furthermore, CAR NK cells can become senescent or even exhausted, 

by the inhibitory interaction with tumor cells via immune checkpoints. In terms of 

persistence, CAR NK cells can be rejected by the host T cells when they recognize non-

self HLA molecules on CAR NK cell surface, diminishing CAR NK cell number or 

affected by the TME, limiting their in vivo sustenance [280]. However, some studies have 

been focused on overcoming these limitations with different approaches. 

 

 

 



INTRODUCTION 

58 
 

1.4. Combination of CAR NK cells with other strategies in 

MM  

1.4.1. Strategies to enhance CAR NK cell efficacy 

As it has been mentioned above, most CAR NK cell limitations affect CAR NK cell 

efficacy. Small inhibitory molecules (eg. GSK3) [343] and activating receptor Ab agonists 

(eg. 2B4, 4-1BB, and SLAMF7) [296] can enhance their cytolytic function as well as 

IMiDs and HDAC inhibitors have been described to modulate their activity, increasing 

activating receptor expression on CAR NK cells and target ligands on the surface of tumor 

cells [344]. Another strategy is the combination of CAR NK cells with mAbs to enhance 

ADCC function (eg. Daratumumab: NCT05182073), oncolytic viruses, as oHSV-1, 

which enhances EGFR-CAR NK cell efficacy against breast cancer brain metastases 

[345], or with bispecific killer cell engagers (BiKEs), facilitating CAR NK cell 

redirecting to the tumor. In this regard, NKG2DxBCMA and CD16xBCMA BiKEs have 

been designed to simultaneously bind NK cells and MM cells, which has demonstrated 

efficacy both in vitro and in vivo [213, 346]. In addition, taking advantage of the CAR 

specificity, NKG2D-CAR NK cell activity has been tested in combination with a 

NKG2DxErbB2 BiAb against solid tumors [347]. Furthermore, difficulties in trafficking 

to tumor sites or inhibition caused by the TME can be overcome by engineering CAR NK 

cells to express chemokines receptors (CXCR4) [335] to redirect them to the tumor or 

knocking out the expression of inhibitory soluble factors receptors, as A2AR and TGFβR, 

as well as by designing CAR molecules, to directly target immunosuppressive cells, 

respectively (eg. CD73-CAR NK cells) [296]. Additionally, CAR NK cell exhaustion can 

be overcome by using blocking Abs [348] or deleting inhibitory receptors [349] to avoid 

immune checkpoint inhibition. Moreover, knocking out negative NK cell regulators as 

CIS, which impairs IL-15 signaling, also improves CAR NK cell function [350] (Fig. 15). 

 



INTRODUCTION 

59 
 

 

Figure 15. Main strategies to enhance CAR NK cells efficacy and persistence. HLA: human leukocyte 

antigen; TCR: T cell receptor; IL-15R: IL-15 receptor; CAR: chimeric antigen receptor; CIML: cytokine-

induced memory-like; BiKE: bispecific killer cell engager; HDAC: histone deacetylase; IMiDs: 

immunomodulatory agents. TGFβR: transforming growth factor-β receptor; A2AR: adenosine A2a 

receptor. Created with BioRender.com. 

 

1.4.1.1. Immune checkpoint inhibitors 

NK cells display inhibitory receptors on the cell surface to avoid toxicity against healthy 

cells. However, the high expression of these molecules on CAR NK cells also decreases 

their antitumor efficacy. In this way, several groups are studying the combination of these 

modified effector cells with blocking Abs or the deletion of the genes encoding inhibitory 

receptors to disrupt immune checkpoint suppression. 

The inhibition of KIRs, one of the main brakes of the NK cell, has been recently object 

of study. The mAb IPH2101 (1-7F9), also called lirilumab, which targets KIR2D, has 

exhibited enhanced preclinical NK cell activity against MM cells, without affecting 

healthy tissues, as well as in combination with lenalidomide [351] and daratumumab 

[352]. Clinical trials did not observe objective responses as a single agent in MM [353]. 

NK cells isolated from these patients had less degranulation and cytokine production, 

probably due to the removal of KIR2D surface molecules through target cell trogocytosis 



INTRODUCTION 

60 
 

[354] or lack of NK cell licensing, as maintained KIR blockade may impair NK cell 

activity [355]. However, a phase I trial administering IPH2101 in combination with 

lenalidomide (NCT01217203) described 5 objective responses and no autoimmunity 

effects [356]. 

Although PD-1 expression is low on NK cells [357], some groups have studied the 

disruption of the PD-1/PD-L1 axis in the cancer field. Yang et al. reported very 

preliminary data about an increased tumor control of BCMA/GPRC5D-CAR NK cells 

against MM cells when they are combined with an α-PD-L1 mAb [358]. Another 

preclinical analysis studying pidilizumab (CT-011), an α-PD-1 mAb, enhanced NK cell 

killing, in an autologous context, against MM cells [359]. However, clinical trial results 

using pembrolizumab, another α-PD-1 Ab, were controversial. Some studies showed 

activity as monotherapy or in combination with lenalidomide, but lack of long-term 

responses and a low benefit:risk ratio, due to the high immune-mediated adverse events, 

dismissed the use of this Ab for MM patients [360-362]. 

In MM, TIGIT ligands (PVR and Nectin-2) have been found overexpressed after BTZ 

treatment [90], inhibiting NK and T cell efficacy. A preclinical study has exhibited 

enhanced anti-MM killing after blocking TIGIT on NK cells [363]. The same strategy has 

been carried out in other tumors, deleting TIGIT expression on CAR NK cells [364] or 

blocking TIGIT through mAbs on NK cells [365-367]. Both approaches have shown 

increased NK cell activity against tumors and could be implemented for MM treatment 

with CAR NK cells. 

Another inhibitory receptor that could hamper CAR NK cell efficacy is TIM-3. Its 

blockade has been described to increase cytokine production and enhance NK cell anti-

MM activity [368]. However, in AML setting, TIM-3+ NK cells have been defined to be 

more mature and TIM-3 blockade diminished IFN-γ secretion [369, 370]. More studies 

are needed to fully understand TIM-3 role in NK cell inhibition. 

Apart from these inhibitory receptors, a main inhibitory immune checkpoint in cancer is 

the HLA-E/NKG2A axis. 

 

 

 



INTRODUCTION 

61 
 

1.4.1.1.1. Description and modulation of the HLA-E/NKG2A axis 

HLA class I can be classified into classical (Ia) including HLA-A, -B, and -C, and 

nonclassical (Ib), comprising HLA-E, -G, and -F, CD1 antigens and MICA and MICB 

[88, 371]. HLA-E, Qa-1b in mice, is located in the MHC complex on chromosome 6, 

between HLA-A and HLA-C genes, and contains 8 exons. Although HLA-E is known for 

displaying a low polymorphism, with two dominant protein variants expressed in 99% of 

cases (HLA-E*01:01 and HLA-E*01:03), a recent study has updated HLA-E number of 

proteins to 110, from 256 different alleles [372]. HLA-E is composed of three 

extracellular α domains (α1-3), encoded by exons 2-4, and it is displayed on the cell 

surface together with β2-microglobulin. Moreover, it has a leader peptide (exon 1) before 

the extracellular domains and, after them, a transmembrane region (exon 5) and the 

intracellular domains (exons 6-7) [372]. In order to stabilize its expression, HLA-E needs 

to bind peptides. Most common peptides are derived from intracellular proteins or 

pathogens, amino acid residues 3-11 of signal peptides of HLA-A, -B, -C, and -G 

molecules, and Hsp60 [88, 373]. These peptides, containing 8 to 10 amino acids, are 

normally generated in the cytosol, after protein degradation by the proteasome, and 

transported into the ER by a transporter associated with antigen processing (TAP) [374]. 

Therefore, a functional TAP, the availability of these peptides and their binding affinity 

to HLA-E, determine HLA-E expression on the cell surface (Fig. 4) [373, 375]. This 

nonclassical molecule is expressed in all nucleated healthy cells, at low levels [88], except 

from trophoblasts cells in the placenta and ductal epithelial cells in the testis and 

epididymis [376], which indicates a key role in immune tolerance [375]. Moreover, HLA-

E can be overexpressed in neoplastic cells. 

HLA-E expression is elevated in several tumors as Ewing sarcoma [377], MM [378], 

cervical cancer [379], AML [380, 381], melanoma [382], and pancreatic ductal 

adenocarcinoma (PDAC) metastatic cells [383], correlating with a poor prognosis in non-

small cell lung carcinoma [384], breast [385], ovarian [379, 386], MM [125, 387], 

glioblastoma [388], and colon cancer [389]. Regarding MM studies, Lagana et al. inferred 

HLA-E expression on MM patients from HLA-A data included in the CoMMpass study 

[387] and Yang et al. demonstrated that high expression of HLA-E correlates with an 

advances ISS stage and high-risk cytogenetics, proposing HLA-E as biomarker for high-

risk MM [125]. Thus, an appropriate correlation between HLA-E expression and worse 

prognosis in MM patients has not been analyzed yet.  



INTRODUCTION 

62 
 

Furthermore, HLA-E expression can be modulated by different cytokines or drugs. BTZ 

treatment reduces HLA-E surface expression on MM cells, due to the ER stress generated 

by this drug and the reduction of signal peptides from HLA molecules degradation, caused 

by the proteasome impaired function [85]. At the same time, BTZ increases NKG2D-L 

expression, being more susceptible to NK cell killing [390]. Another MM drug, Selinexor, 

also downregulates HLA-E expression on lymphoma cells [101], as well as dinaciclib, a 

CDK inhibitor, on AML cells [391]. In addition, selenite, an inorganic selenium 

compound, has been found to abrogate HLA-E protein expression due to oxidative stress 

induction [392]. 

On the contrary, many studies have reported that IFN-γ upregulates HLA-E expression 

on MM cells [349, 393], AML cells [394], chronic lymphocytic leukemia (CLL) B-cells 

[348], and even on solid tumor cells [395]. However, it has been also reported that this 

cytokine can also induce HLA-E shedding by MMP [382]. Regarding MM, Pérez-Andrés 

et al. discovered that this cytokine, mainly produced by NK cells and T cells, is expressed 

at high levels in patients with MM [130]. 

As it has been previously described, NKG2A is a type II transmembrane glycoprotein that 

belongs to the C-type lectin-like receptor superfamily, which includes the NKG2 family. 

NKG2C and NKG2E (activating receptors) and NKG2A (inhibiting receptor) bind HLA-

E, although HLA-E affinity binding to NKG2A is 6 times higher. Therefore, NK cell 

common response to HLA-E-bearing cells is inhibitory [375, 396]. 

NKG2A protein is encoded by the KLRC1 gene, consists of 7 exons, and it is located in 

chromosome 12, as well as the gene that encodes CD94, another C-lectin type II receptor. 

NKG2A is displayed in the cell membrane forming a heterodimer with CD94, a receptor 

that lacks intracellular signaling [397]. NKG2A possesses a cytoplasmic tail containing 

ITIM domains, which recruits SHP-1/2 and triggers VAV1, ZAP70 and src-family 

kinases, as Fyn and Lck, dephosphorylation, reducing NK cell activity. This inhibitory 

complex, NKG2A/CD94, is expressed on NK cells, invariant NKT cells, activated CD8+ 

T cells and γδ T cells [103].  

NKG2A, as well as KIRs, is indispensable for NK cell education and autoimmune control 

[396]. However, under certain circumstances, as the cytokine-mediated activation of NK 

cells, typical education rules are not needed, thus unlicensed NK cells can acquire 

complete functions and mediate NK cell response [398].  



INTRODUCTION 

63 
 

Fifty percent of unstimulated PB circulating NK cells and around 60% of tumor-

infiltrating NK cells express NKG2A [399]. Of note, NKG2A expression correlates with 

a poor prognosis in AML, liver, and head and neck cancer patients [103]. 

NKG2A expression can be also modulated by cytokines and drugs. Remarkably, 

dasatinib, a tyrosine kinase inhibitor, used for chronic myeloid leukemia (CML) 

treatment, reduces NKG2A expression, while augmenting NK cell antitumor efficacy. 

However, other tyrosine kinase inhibitors as imatinib and nilotinib do not downregulate 

NKG2A expression. Cytokines IL-21, IL-15, IL-12, IL-10, and TGF-β are known to 

overexpress NKG2A presence on the immune effectors cell membrane [396]. 

The interaction of HLA-E and NKG2A disrupts the actin network, impairing effector-

target immune synapse [400]. NKG2A binding to HLA-E is tuned depending on the 

peptides that HLA-E is presenting and therefore NKG2A inhibitory signaling can be 

modulated or attenuated. For example, HLA-G peptides presenting on HLA-E molecules 

induce the strongest inhibition reaction [401], while Hsp60 loading of HLA-E molecules 

impairs NKG2A binding to HLA-E [402]. 

Due to the high expression of HLA-E in diverse tumors and the increased expression of 

NKG2A, above all on NK cells, as well as in activated or CAR-modified NK cells, diverse 

alternatives have been proposed to impair this immune checkpoint inhibitory response 

(Fig. 16). 



INTRODUCTION 

64 
 

 

Figure 16. Modulation and signaling of HLA-E/NKG2A axis. HLA: human leukocyte antigen; mAb: 

monoclonal antibody; PEBL: protein expression blocker; ITIM: immunoreceptor tyrosine-based inhibitory 

motif; SHP-1: Src homology region 2 domain-containing phosphatase-1; LCK: lymphocyte-specific 

protein tyrosine kinase; SYK: spleen tyrosine kinase; ZAP-70: Zeta-chain-associated protein kinase 70. 

Adapted from Fisher et al. 2022 [103]. Created with BioRender.com. 

 

Direct targeting of HLA-E has been proposed, using the 3H4 and TFL-033 mAbs. Despite 

these Abs have demonstrated efficacy in vitro [103], most studies have been focused on 

targeting the NKG2A receptor instead.  

 

1.4.1.1.2. Preclinical studies targeting NKG2A 

The elimination of NKG2A to improve NK cell responses has been also carried out 

against different types of cancers. A siRNA has been used to permanently reduce NKG2A 



INTRODUCTION 

65 
 

mRNA levels in NK cells, which has enhanced NK cell efficacy against B lymphoblastoid 

cells [403]. A new strategy to retain NKG2A receptor by the ER was developed by 

Kamiya et al. Fragments from the mouse anti-human NKG2A blocking Ab (Z199) were 

linked to ER-retention domains to generate NKG2A protein expression blockers 

(PEBLs). These proteins hampered NKG2A expression on NK cell surface, enhancing 

their cytotoxic activity against different HLA-E+ tumors, even more than blocking NK 

cells with the mouse α-NKG2A Ab Z199 [394]. Moreover, Ruggeri et al. demonstrated 

the efficacy of monalizumab, a humanized α-NKG2A IgG4 blocking Ab, on NK cells 

against engrafted leukemia mice and proposed the use of this Ab in combination with 

adoptive NK cell therapy or after stem cell transplant recovery, to foster tumor reduction 

[404]. In CLL and EBV+ lymphomas, the improvement of NK cell cytotoxicity due to 

NKG2A blockade with monalizumab, has been demonstrated, suggesting its use in 

clinical trials [348]. In solid tumors, monalizumab treatment significantly enhances 

antitumor NK cell killing against HLA-E+ tumor cells [383, 395], although its 

combination with an α-PDL1 Ab (durvalumab) was needed to reduce tumor burden in 

vivo [399]. In addition, intratumoral co-injection of human PB NK cells with 

monalizumab was able to reduce tumor growth in immunodeficient mice bearing head 

and neck squamous cell carcinoma (HNSCC) [405]. Remarkably, Battin et al. have 

discovered that monalizumab blocking efficacy on NK cells depends on the peptides 

presented by HLA-E, as monalizumab was more effective when HLA-E loaded HLA-A, 

-B, and -C peptides, instead of HLA-G peptide [402].  

Regarding MM and out of the CAR setting, first approaches isolating NKG2A+ and 

NKG2A- NK cells reported that NKG2A- degranulation was not impaired by HLA-E+ 

MM cells, while NKG2A+ NK cells degranulation was affected [378]. Although 

Mahaweni et al. reported that NKG2A blockade with the Z199 Ab did not enhance 

NKG2A+ NK cells degranulation against primary MM cells, other studies have 

demonstrated the relevance of targeting this inhibitory receptor in this tumor [406]. 

Tognarelli et al. shown that the blockade of NKG2A with the same Ab, the Z199, 

consistently increases the killing activity of activated NK cells, expanded from diagnosed 

MM patients or HD PB [393]. In 2022, during the development of this thesis, Bexte et al. 

deleted the NKG2A-encoding KLRC1 locus in NK cells by CRISPR-Cas9, demonstrating 

their enhanced efficacy against MM primary cells [349]. 



INTRODUCTION 

66 
 

Concerning CAR NK cells, combination of this adoptive therapy with complete or partial 

NKG2A elimination or with NKG2A neutralization has not been reported yet. 

 

1.4.1.1.3. Clinical trials with α-NKG2A neutralizing antibodies 

Despite these impressive preclinical results interfering the HLA-E/NKG2A axis, clinical 

study results are controversial. Most trials are currently testing monalizumab, in 

monotherapy or in combination with other blocking Abs or chemotherapy (Table 5).  

However, monalizumab efficacy as a single agent against solid tumors did not meet the 

expected results, only achieving stable disease responses against gynecologic 

malignancies [407]. In addition, first HNSCC patients from the phase II UPSTREAM 

trial were treated only with monalizumab, obtaining an OS of 6.7 months and 1.7 months 

of median PFS. Due to these poor results, they decided to combine this α-NKG2A Ab 

with durvalumab, an α-PD-L1 Ab, in the following cohort of patients [408]. Similar and 

other combinatorial strategies were carried out by other groups. Regarding HNSCC 

patients (NCT02643550), Cohen et al. tested the efficacy and safety of the combination 

of monalizumab and cetuximab, an α-EGFR Ab. In a cohort that included patients who 

had relapsed after platinum-based therapies, an 8.5-month median OS, a 27.5% ORR and 

a 4.5-month median PFS were reported. Moreover, a second cohort including patients, 

who not only had relapsed to platinum therapies, but also to PD-L1 inhibitors, an ORR of 

20% and a median duration of response of 5.2 months were obtained. More than 10% of 

patients suffered from dermatitis acneiform, dry skin, pruritus, fatigue, hypomagnesemia, 

skin fissures, infusion-related reaction, mucosal inflammation, nausea, paronychia, rash, 

asthenia and diarrhea after treatment [409]. Additionally, a new triplet regimen including 

monalizumab, cetuximab, and durvalumab has been proposed for HNSCC patients within 

the same clinical trial (NCT02643550). Last results reported an ORR of 32.5% (3 patients 

achieved CR) and the median PFS was 6.9 months. Most common adverse events were 

increased lipase or amylase and dermatitis acneiform [410]. The phase II study COAST 

for non-small cell lung cancer (NSCLC) patients has combined monalizumab and 

durvalumab, obtaining a 72.7% 12-month PFS and 35.5% ORR, better results than its 

control (only durvalumab). Median PFS was 15.1 months. Most frequent adverse events 

found after durvalumab plus monalizumab were also described for durvalumab 

monotherapy (cough, dyspnea, pneumonitis, asthenia, and pruritus) [411]. Moreover, 



INTRODUCTION 

67 
 

MIMOSA phase II trial was designed for HER2+ breast cancer patients, combining 

trastuzumab and monalizumab. Unfortunately, recently published results reported that, 

although both treatments were well tolerated, they did not achieve objective responses 

[412]. 

Regarding hematologic tumors, a phase I trial with a short follow-up has shown promising 

results. Monalizumab efficacy and safety has been determined after allogeneic stem cell 

transplant in patients with AML, myelodysplastic syndrome (MDS), lymphoma, CLL, 

and myelofibrosis. This drug was highly tolerated, with no side effects, and the short-term 

efficacy analysis revealed that 13 out of 15 patients had CR. Of note, the blockade of 

NKG2A on circulating NK cells after 14 weeks from infusion was detected [413]. 

These results indicate that monalizumab monotherapy is not able to restore NK and T cell 

cytotoxicity against tumor cells within the patient, especially in solid tumors, probably 

due to their exhausted phenotype induced by tumor cells and the TME. However, 

fostering these immune cells efficacy with other blocking Abs or new approaches, such 

as CAR NK cell adoptive therapy, may result in better clinical outcomes. 

Another human α-NKG2A Ab that is being clinically studied is BMS-986315. This α-

NKG2A is an inert IgG1.3, which contains 3 mutations in the heavy chain, L234A, L235E 

and G237A, to prevent Fc-mediated activity, through CD16 recognition, and C1q binding. 

Estimated half-life of this Ab in circulation is approximately 16.5 days [414]. 

Remarkably, the efficacy and toxicity of this Ab has been never compared to other α-

NKG2A Abs as monalizumab or KSQ. Although preclinical data with BMS-986315 Ab 

has not been found, its efficacy is being tested against solid tumors, in combination with 

nivolumab or cetuximab (NCT04349267), and NSCLC, together with nivolumab and 

chemotherapy (NCT06094296). Of note, neither this Ab has been used against 

hematologic tumors, including MM, nor in combination with CAR NK cells. 

 

 

 

 

 



INTRODUCTION 

68 
 

Nº Status 
Study 

start 
Treatment Condition Phase Pat Location NCT 

1 Recruiting 2022 
Monalizumab plus 

Durvalumab 
NSCLC III 999 Global NCT05221840 

2 Recruiting 2022 

Monalizumab plus 

durvalumab plus 

platinum doublet 

chemotherapy 

NSCLC II 350 Global NCT05061550 

3 
Active, not 

recruiting 
2020 

Monalizumab plus 

Cetuximab 
HNSCC III 370 Global NCT04590963 

4 
Active, not 

recruiting 
2021 

Monalizumab plus 

Trastuzumab 
Breast cancer II 38 Netherlands NCT04307329 

5 Recruiting 2019 
Monalizumab plus 

Durvalumab 
NSCLC II 120 France NCT03833440 

6 
Active, not 

recruiting 
2019 

Monalizumab plus 

Durvalumab 
NSCLC II 188 Global NCT03822351 

7 Recruiting 2017 Monalizumab HNSCC II 340 Europe NCT03088059 

8 Unknown 2016 Monalizumab 
Hematologic 

malignancies 
I 18 France NCT02921685 

9 Completed 2015 

Monalizumab plus 

Cetuximab ± anti-

PD-L1 

HNSCC I/II 143 Global NCT02643550 

10 Terminated 2015 
Monalizumab plus 

Ibrutinib 
CLL I/II 22 USA NCT02557516 

11 
Active, not 

recruiting 
2016 

Monalizumab plus 

Durvalumab 
Solid tumors I/II 383 Global NCT02671435 

12 Recruiting 2023 

Monalizumab plus 

Durvalumab plus 

Chemotherapy 

SCLC II 38 USA NCT05903092 

13 Withdrawn 2017 

Monalizumab plus 

Chemotherapy plus 

radiotherapy 

Esophagus and 

gastroesophageal 

junction 

carcinomas 

I/II 0 Belgium NCT03307941 

14 Completed 2019 
Monalizumab plus 

Durvalumab 
NSCLC II 84 Global NCT03794544 

15 Completed 2020 Monalizumab 
Hematological 

and solid tumors 
I 19 France NCT04333914 

16 Withdrawn 2020 
Monalizumab plus 

Durvalumab 

Colorectal 

cancer 
II 0 Unknown NCT04145193 

17 Recruiting 2020 

BMS-986315 ± 

nivolumab or 

cetuximab 

Solid tumors I/II 308 America NCT04349267 

18 
Not yet 

recruiting 
2023 

BMS-986315 plus 

nivolumab and 

chemotherapy 

NSCLC II 196 Global NCT06094296 

 

Table 5. Current clinical trials using α-NKG2A antibodies for cancer treatment. NCT: national clinical 

trial; NSCLC: non-small cell lung cancer; HNSCC: head and neck squamous cell carcinoma; PD-L1: 

programmed cell death ligand 1; CLL: chronic lymphocytic leukemia; SCLC: small-cell lung cancer; 

NSCLC: non-small cell lung cancer. 

 

1.4.2. Strategies to enhance CAR NK cell persistence 

Within efficacy challenges, one of the relevant limitations of CAR NK cells is their low 

lifespan. In vivo persistence after adoptive transfer has been previously reported to 

correlate with clinical responses [415]. Mature NK cell persistence in circulation is 

around 1-2 weeks and activated and expanded human NK cells can survive from 4 to 5 

weeks without cytokine stimulation and up to 68 days when they are engineered to 

express IL-15, in xenografted immunodeficient mice [280]. First approaches tested 

multiple CAR NK cell infusions to enhance their sustenance in circulation, but the 



INTRODUCTION 

69 
 

persistence of CAR NK cells after subsequent infusion doses was even shorter [416]. 

After lymphodepletion recovery, T cells from the host generate allorejection of CAR NK 

cells via HLA-TCR mismatch, reducing their number in the bloodstream. For this reason, 

administration of haploidentical CAR NK cells has been carried out by many groups [417, 

418]. In line with this, multiengineered CAR NK cells can avoid recognition through 

knocking out HLA-I molecules or expressing the HLA-E ones, reducing T and NK cell 

allorejection, respectively, by the host [296]. Alternatively, CAR sequences can be 

modified to induce interleukin secretion (armoured CAR) [419], enhancing NK cell 

cytotoxicity and persistence in vivo. This time, second infusions of CD19-CAR/IL-15 NK 

cells achieved a longer CAR NK cell persistence [420]. Other strategies are the 

administration of exogenous IL-15, using IL-15 superagonists as NKTR-255 [421] or 

ALT-803 [422]. In line with this, oncolytic virus carrying IL-15/IL-15R complexes have 

enhanced EGFR-CAR NK cell proliferation (Fig. 14) [296].  

It has been reported that, although NK cells belong to the innate immunity, they can 

acquire memory properties after their interaction with haptens, viruses or tumor cells as 

well as after an activation with a combination of cytokines (cytokine-induced memory-

like; CIML). One of the main properties of these cells is their improved persistence 

compared to conventional NK cells [423, 424]. 

 

1.4.2.1 Cytokine-induced memory-like NK cells 

First studies reported memory-like properties in mouse NK cells after their exposure to 

haptens or viruses (MCMV, Zika) [423, 425]. Mouse Ly49H+ NK cells were found to 

proliferate after MCMV interaction and remain for months at low frequency, as happens 

with T cells. The same reaction was found in human NK cells regarding CMV, hantavirus, 

and chikunguya virus [423]. In the case of CMV, NKG2C+ NK cells proliferated and then 

contracted, persisting for up to a year. These cells displayed a mature phenotype (CD56dim 

CD57+), reduced NKG2A expression and produced high amounts of IFN-γ [426]. A gene 

expression array comparing these memory-like NK cells (NKG2C+) with NKG2C- 

revealed 89 upregulated and 102 downregulated genes, suggesting that the response to 

virus exposure triggers several gene expression modifications that lead to memory 

properties in NK cells [427]. However, chronic CMV stimulation fostered PD-1 and 

LAG-3 expression and consequently, NK cell exhaustion [428]. 



INTRODUCTION 

70 
 

Alternatively, tumor-priming is also able to induce a memory-like phenotype in NK cells 

(TIML-NK cells). These cells have been described to be tumor-specific, show a higher 

cytotoxicity and perforin release and display maturity markers (CD57+ KLRG1+ KIR+). 

Remarkably, exposure to tumor cells rapidly induces S-phase entry [424]. TIML-NK 

cells, primed with the tumor product INKmuneTM, have been tested in AML and MDS 

patients, proving safety and quick peripheral NK cell activation [429]. 

In 1999, Fehniger et al. tested the effect of IL-18, IL-15, or IL-12 on human NK cells, 

where IL-18 and IL-12 induced more IFN-γ, the combination of IL-15 and IL-12 

significantly increased IL-10, TNF-α, and MIP-1β, while IL-15 and IL-18 augmented 

GM-CSF production [430]. In 2009, Cooper et al. cultured mouse NK cells overnight 

with the combination of the three cytokines, IL-18, IL-15, and IL-12. These preactivated 

NK cells had enhanced IFN-γ secretion and persisted over 3 weeks in the mice. Moreover, 

these cells strongly responded upon stimulation compared to conventional NK cells, 

suggesting that the combination of these three cytokines induces a memory-like 

phenotype, calling them CIML-NK cells (Fig. 17). Notably, all these properties passed on 

to their daughter cells, therefore this preactivation promotes DNA changes that can be 

inherited [431]. In 2012, Romee et al. corroborated these memory properties in human 

NK cells. These CIML-NK cells had improved IFN-γ production, enhanced response after 

stimulation, higher proliferation, and their daughter cells also inherited all these 

characteristics [432]. 

 



INTRODUCTION 

71 
 

 

Figure 17. CIML-NK cells display a different phenotype compared to conventional NK cells. KIR: 

killer-cell immunoglobulin-like receptor; TGFβ: transforming growth factor-β; CIML: cytokine-induced 

memory-like. Adapted from Tarannum et al., 2021 and Berrien-Elliott et al. 2023 [296, 433]. Created with 

BioRender.com. 

 

Since then, several studies have been focused on this new population, mainly in 

describing its phenotype. CIML-NK cells have been described to overexpress activation 

receptors such as CD25 [434-436] and CD69, NKG2A, CD56 [437], and Sca-1 while 

decreasing CD62L, CD16 [435, 438], and CXCR4 expression [439]. Moreover, they had 

enhanced SEMA7A (a potent cytokine stimulator) [440], TNF-α [441], IFN-γ [398, 435, 

442], FasL, and perforin production, and the transcription factors T-bet and Blimp-1 were 

upregulated [438]. In contrast, CIML-NK cells possess a lower expression of TGFβR, 

being less inhibited by the presence of TGFβ in the TME, and KIRs [443]. Of note, the 

exposure of NK cells to IL-12, IL-15, and IL-18 has been reported to restore the functional 

activity of unlicensed NK cells [444]. 

Recently, mass cytometry analysis revealed that CIML-NK cells have high CD56, Ki-67, 

NKG2A, NKG2D, NKp30, and NKp44 expression and low CD16 and CD11b expression 

in vitro, as well as an increased frequency of TRAIL, CD69, NKG2A, NKp30, and 

CD62L positive NK cells and a reduced CD27 and CD127 NK cell percentage in vivo, 



INTRODUCTION 

72 
 

compared to baseline NK cells [445]. Another CyTOF report described that CIML-NK 

cells exhibit reduced CD16, CD57, CD8, NKp30, NKp46, and NKG2D expression and 

increased CD56 and CD69 expression [446]. Additionally, a transcriptomic study shown 

an enrichment in KLRC1 mRNA and GATA3 target genes compared to conventional NK 

cells [445]. 

CIML-NK cells have shown in vitro efficacy against hematological malignancies as ALL 

[437, 447], AML [437, 442, 445], as well as against solid tumors such as ovarian [441], 

hepatocellular carcinoma [448], melanoma [449, 450], sarcoma, and renal cell carcinoma 

[437].  

Additionally, CIML-NK cells proliferated more compared to conventional NK cells in 

vitro and in vivo [436, 439], although a higher apoptosis rate has been also associated 

[441]. 

Furthermore, some groups have focused their studies on unveiling CIML-NK cell 

metabolism. These cells have an increased metabolic activity, demonstrated by an 

enhanced glycolytic capacity and oxidative phosphorylation (OXPHOS), above all during 

the initial days after preactivation with IL-12, IL-15, and IL-18, compared to IL-15 

control NK cells [451, 452]. Of note, limiting OXPHOS flux during preactivation, 

strongly impaired CIML-NK cell IFN-γ production and cytotoxicity, suggesting the 

importance of nutrients for CIML-NK correct differentiation [452]. Furthermore, CIML-

NK cells overexpress nutrient transporter expression, including the glucose transporters 

GLUT1 and GLUT3, the amino acid transporter CD98, and the transferrin receptor CD71 

[453]. 

Although CIML-NK cells have shown efficacy in preclinical studies, there are several 

modifications or combinations that can be carried out to optimize these effector cells 

responses. Since CIML-NK cells overexpress NKG2A, and based on previously 

described results blocking or knocking out NKG2A expression on conventional NK cells, 

Berrien-Elliot et al. impaired NKG2A inhibitory signaling in CIML-NK cells with an α-

NKG2A Ab, enhancing transfectant HLA-E+ leukemia cell killing [445]. Moreover, this 

group has also combined CIML-NK cells with cetuximab, resulting in increased NK cell 

activity against HNSCC [454]. In addition, Kerbauy et al. have combined CIML-NK cells 

with a bispecific CD30xCD16 Ab (AFM13), enhancing CD30+ lymphoma tumor killing 

[455]. 



INTRODUCTION 

73 
 

Recently, CIML-NK cells have been modified to enhance their efficacy through CAR 

redirection. First study engineered CIML-NK cells to express a CD19-CAR, achieving 

improved in vivo responses against NK-resistant lymphomas, [456]. In line with this, He 

et al. also modified CIML-NK cells to express a CD19-CAR and demonstrated enhanced 

antitumor efficacy against B cell malignancies [457]. Dong et al. incorporated a TCR-like 

CAR structure that recognizes the intracellular neoepitope NPM1 presented by HLA-A2 

on the cell membrane, targeting AML cells with NPM1c mutations [458]. CIML-NK cells 

have been also redirected to solid tumors through CAR expression. Jacobs et al. 

incorporated an α-EphA2 CAR in CIML-NK cells, resulting in a higher IFN-γ production 

and cytotoxicity against HNSCC [454]. 

Due to the impressive preclinical results, CIML-NK cell safety and antitumor efficacy are 

being tested in several clinical trials, most of them in AML patients (Table 6). First-in-

human treatment with CIML-NK cells (NCT01898793) started in 2014. First results 

considering 9 AML patients described 55% of ORR and 45% CR [442]. Moreover, the 

NCT02782546 study revealed that 87% of the AML patients had achieved CRs after one 

month from CIML-NK cells infusion. Of note, these cells persisted for over 2 months in 

AML patients [417]. Another AML trial in pediatric and young adult patients, achieved 

CRs in 4 out of 8 evaluable patients, with no significant toxicities. NK cells were the 

dominant lymphocyte subset in BM of these patients until day 100 after CIML-NK cells 

infusion and ex vivo analysis demonstrated that these NK cells retained their increased 

function for at least 4 weeks after administration [418]. Additionally, the NCT04024761 

study described the rapid expansion of CIML-NK cells after infusion, although 4 out of 

6 myeloid malignancies patients developed pancytopenia and all of them had fever, due 

to IL-2 administration during the first 12 days. Remarkably, 66.67% of patients achieved 

CR [459]. 

Recently, a CIML-NK cell product, WU-NK-101, has been generated by the company 

Wugen. These allogeneic PB CIML-NK cells have shown the same phenotypic 

characteristics, as previously described by other groups. They have improved glycolytic 

and OXPHOS capacity and exhibited an enhanced cytotoxic capacity that is not impaired 

by cytokines or inhibitory cells present in the TME [460]. Nowadays, this product is being 

tested as monotherapy (NCT05470140) and in combination with cetuximab 

(NCT05674526). 



INTRODUCTION 

74 
 

Of note, a clinical trial is testing autologous CIML-NK cell efficacy in NDMM patients 

who are MRD+ and eligible for ASCT, in combination with low dose IL-2, intravenous 

Igs (IVIGs) and the molecule BHV-1100 (also known as KP1237). It is a CD38 Ab 

recruiting molecule (ARM) that binds CD38 on the target cell and also the exogenous 

IVIGs [461]. However, preliminary results have not been published yet and no preclinical 

data from these trials is available, indicating the lack of knowledge of CIML-NK cell 

efficacy against MM. Moreover, to date, no clinical data showing CAR CIML-NK cells 

against hematologic tumors have been published. 

 

Nº Status 
Study 

start 
Treatment Condition Phase Pat Location NCT 

1 Recruiting 2023 CIML NK AML I/II 20 USA NCT05580601 

2 
Not yet 

recruiting 
2023 

Autologous or 

allogeneic ML NK 

cells plus 

nivolumab and 

relatlimab 

Melanoma I 33 USA NCT05629546 

3 Withdrawn 2022 
CIML NK plus 

IL-2 
AML and MDS II 0 USA NCT04893915 

4 Recruiting 2020 

CIML NK plus N-

803 ± ipilimumab 

or cetuximab 

HNSCC I 25 USA NCT04290546 

5 Withdrawn 2023 CIML NK AML II 0 USA NCT04354025 

6 Recruiting 2017 CIML NK AML I/II 110 USA NCT03068819 

7 Terminated 2014 
CIML NK plus 

IL-2 
AML and MDS I/II 89 USA NCT01898793 

8 Recruiting 2017 
CIML NK plus 

ALT-803 
AML II 60 USA NCT02782546 

9 
Active, not 

recruiting 
2019 CIML-NK 

AML, MDS, 

myeloproliferative 

neoplasm and 

juvenile 

myelomonocytic 

leukemia 

I 50 USA NCT04024761 

10 Recruiting 2021 

Autologous ML 

NK plus BHV-

1100 

MM I 25 USA NCT04634435 

11 Recruiting 2023 WU-NK-101 AML I 24 
USA and 

Australia 
NCT05470140 

12 
Not yet 

recruiting 
2023 

WU-NK-101 plus 

Cetuximab 

Colorectal and 

HNSCC 
I 30 - NCT05674526 

13 Terminated 2021 
CT101a (CIML-

NK) 
AML I 3 China NCT05256277 

 

Table 6. Current clinical trials using CIML-NK cells against tumor malignancies. NCT: national 

clinical trial; CIML: cytokine-induced memory-like; AML: acute myeloid leukemia; MDS: 

myelodysplastic syndrome; HNSCC: head and neck squamous cell carcinoma; MM: multiple myeloma. 

 

 

 

 

 



 

75 
 

 

 

 

 

 

 

 

 

 

2. HYPOTHESIS AND OBJECTIVES 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

76 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



HYPOTHESIS AND OBJECTIVES 

77 
 

Despite all the therapeutic strategies approved for the treatment of MM, most patients 

progress, with increasingly shorter response periods, and MM is still an incurable disease. 

MM microenvironment is characterized by a high immunosuppression, mainly in late 

stages of the disease. The discovery of new targeted therapies, as CAR T cells, has 

revolutionized the treatment of hematologic malignancies, although severe toxicities, 

such as CRS, neurotoxicity and prolonged cytopenias, together with CAR T cell 

exhaustion, responsible for suboptimal responses, have arisen the need to find new CAR 

effectors. Allogeneic CAR NK cells have emerged as a safer, off-the-self, cost-effective 

approach, compared to CAR T. However, CAR NK cell reduced activity via inhibitory 

immune checkpoints and low persistence in vivo, mainly caused by the 

immunosuppressive TME, are limiting CAR NK antitumor responses in preclinical 

studies and clinical trials. Published studies have described the HLA-E/NKG2A axis as a 

main inhibitory signal in other tumors, correlating with a poor overall survival, and 

combinatorial strategies to disrupt this interaction have reported promising results in 

cancer patients. In addition, cytokine-induction of a memory-like phenotype on NK cells 

is able to enhance NK cell lifespan and proliferative capacity, which have an impact on 

NK cell efficacy.  

 

2.1. Hypothesis 

Combined strategies of CAR activated and expanded NK (NKAE) cells with α-NKG2A 

Abs or the induction of a memory phenotype can enhance CAR NK immunotherapy anti-

MM efficacy and persistence. 

 

 

 

 

 

 

 



HYPOTHESIS AND OBJECTIVES 

78 
 

2.2. Objectives 

Considering the current limitations of CAR NK cells in the treatment of hematologic 

tumors, the main objective of this thesis is the generation of optimized therapeutic 

combinations for CAR NK cells to enhance RRMM treatment. Specific objectives of this 

thesis are: 

1. To analyze the relevance of the HLA-E/NKG2A axis in MM.  

2. To potentiate the anti-MM activity of NKG2D- and BCMA-CAR NKAE cells by 

disrupting the HLA-E/NKG2A axis with a novel α-NKG2A antibody.  

3. To increase the immunotherapy persistence and efficacy by reprogramming 

NKG2D-CAR NKAE cells towards a cytokine-induced memory-like phenotype 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

79 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

3. MATERIALS AND METHODS 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

80 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



MATERIALS AND METHODS 

81 
 

3.1. Cell lines and primary samples 

3.1.1. Cell lines 

MM cell lines NCI-H929 (ACC 163), L-363 (ACC 49) and RPMI-8226 (ACC 402) were 

purchased from DSMZ (Leibniz Institute, Germany) and U-266B1 (TIB-196) from 

ATCC, XG-1 cell line was provided by Dr. Shelly Lawson (Sheffield University, UK), 

ARP-1 cell line (CVCL_D523) was provided by Dr. Joshua Epstein (Arkansas Cancer 

Research Center, Arkansas, USA) and BTZ-resistant cell lines NCI-H929 R20 and 

RPMI-8226 R7 were kindly provided by Dr. Joaquín Teixidó (Centro de Investigaciones 

Biológicas Margarita Salas, Madrid) [462].  

K562-mb21-41BBL cell line (Clone 9.mbIL21 or CSTX002), provided by Dr. Dean A. 

Lee (Nation Wide Children, Ohio, USA), has been used to promote NK cell in vitro 

activation and expansion. For this purpose, K562, an erythroleukemic cell line with 

positive Philadelphia chromosome, was modified to constitutively express membrane-

bound IL-21 (mb21), 4-1BB ligand (4-1BBL or CD137-L), CD64, CD86 and truncated 

CD19. Both IL-21 and 4-1BBL promote NK cell activation and proliferation [463]. 

HEK293T, an embryonic cell line modified to express SV40 T antigen, was required for 

lentiviral packaging. This cell line was obtained from DSMZ. 

All cell lines, except HEK293T, were cultured in Roswell Park Memorial Institute 

(RPMI)-1640 medium (Biowest) supplemented with 10% fetal bovine serum (FBS; 

Hyclone), 2 mM L-glutamine and 100 IU/mL penicillin/streptomycin (P/S; Lonza) at 

37ºC in a humidified atmosphere with 5% CO2. Additionally, XG-1 cell line were 

cultured with 1 ng/ml recombinant human IL-6 (Miltenyi). HEK293T cell line was 

cultured in Iscove´s Modified Dulbecco´s Medium (IMDM; Lonza) supplemented with 

10% FBS and 100 IU/ml P/S. The medium was changed every 2-3 day and cell count was 

assessed in a Neubauer chamber (VWR) using 0.4% Trypan Blue (Gibco) exclusion 

method. The absence of mycoplasma was regularly tested by Polymerase Chain Reaction 

(PCR) and cell line integrity was tested through short tandem repeat (STR) analysis in the 

Genomic Unit from Centro Nacional de Investigaciones Oncológicas (CNIO), Madrid. 

 

 



MATERIALS AND METHODS 

82 
 

3.1.1.1. Cell lines modified in the laboratory 

Additionally, some cell lines were modified due to experiment requirements. 

For BTZ-resistant RPMI-8226 R20 cell line generation, RPMI-8226 R7 cells were 

gradually cultured with 2 nM BTZ (MedChemExpress) up to 20 nM BTZ, until culture 

viability was up to 80%.  

For mouse models, U-266 and RPMI-8226 cell lines were lentivirally transduced with 

ffLucGFP vector to constitutively express luciferase and GFP for tumor burden detection. 

All these cells were cultured as indicated in 3.1.1. 

 

3.1.2. Primary samples 

HD PB samples (n=52), as well as MGUS and MM patients BM aspirates (n=76), were 

obtained from Hospital 12 de Octubre (H12O). Healthy BM samples from orthopedic 

patients (n=20) were obtained from Hospital de parapléjicos de Toledo, hereinafter 

referred to as HD BM samples, and CB samples (n=4) from Centro de Transfusiones de 

la Comunidad de Madrid. Both HD and MM patients gave their written informed consent 

to collaborate in the study (nº20/326), previously approved by the H12O ethics 

committee, in accordance with the Declaration of Helsinki protocol. All primary samples 

were analyzed, according to the arrival order of patients at hospital. 

 

3.2. Expansion, culture and CAR transduction of activated 

and expanded NK cells (NKAE) 

3.2.1. Generation of CAR NKAE and CAR CIML-NKAE cells from PB samples 

NKAE and CIML-NKAE cells were generated from PB mononuclear cells (PBMCs). 

Two 9ml-EDTA PB tubes, collected from HDs, were diluted 1:4 with Phosphate-

Buffered Saline (PBS; Gibco) and added 3:1 over Ficoll-Paque (Sigma-Aldrich). Samples 

were centrifuged at 1,800 rpm for 20 minutes (acceleration 6; deceleration 3). PBMCs 

layer was collected, washed with PBS, and centrifuged at 1,600 rpm for 10 minutes. 

PBMCs were then co-cultured at a ratio of 1:1.5 with K562-mb21-41BBL cells, 

previously irradiated under 100 Gy, at 1.2 x 106 cells/ml in RPMI-1640 medium 



MATERIALS AND METHODS 

83 
 

supplemented with 10% human AB serum (Sigma-Aldrich), 2 mM L-glutamine, 100 

IU/mL P/S and 50 IU/ml IL-2 (Miltenyi). After 3 days, the medium was changed, 

maintaining the culture at 1.2 x 106 cells/ml. On day 6 of culture, NKAE cells were 

isolated by immunomagnetic negative selection following the manufacturer´s instructions 

(NK cell Isolation Kit, Human, Miltenyi). NKAE cell purification was analyzed using 

flow cytometry. On day 8, purified NKAE cells were lentivirally transduced with 

NKG2D- or BCMA-CAR vectors (as described in 3.2.5) for 16 hours. After washing with 

PBS, NKG2D- and BCMA-CAR NKAE cells were preferentially used between days 14-

18 of culture and maintained until a maximum time of 23 days, changing the medium 

every 3 days (Fig. 18A).  

To obtain CAR CIML-NKAE cells, after removing lentivector-containing medium, CAR 

NKAE cells were exposed to 100 ng/ml IL-18 (MBL), 20 ng/ml IL-15 (Miltenyi), 10 

ng/ml IL-12 (Miltenyi) and 50 IU/ml IL-2 and cultured at 3.7 x 106 cells/ml using the 

medium described above. Control NKG2D-CAR NKAE cells were also incubated at 3.7 

x 106 cells/ml with 50 IU/ml IL-2. After 16h, cells were washed three times with PBS, to 

remove residual cytokines, and resuspended in the culture medium with a maintenance 

dose of 50 IU/ml of IL-2 at 1.5 x 106 cells/ml until day 23 of culture, changing the medium 

every 3 days, but experiments were mainly performed from day 14 to 18 of culture (Fig. 

18B). 

Considering that transgene stable expression using lentivectors is achieved between 5-7 

days [464], at day 6 after transduction, CAR expression was confirmed. 

 

 



MATERIALS AND METHODS 

84 
 

 

Figure 18. CAR NKAE cell generation protocol. Schematic representation of A) NKG2D- and BCMA-

CAR NKAE cell and B) NKG2D-CAR CIML-NKAE and NKAE cell generation and expansion from HD 

PB samples. PBMCs: peripheral blood mononuclear cells; IL: interleukin; mb21: membrane-bound IL-21; 

NKAE cells: activated and expanded NK cells; CIML: cytokine-induced memory-like. Created with 

BioRender.com. 

 

3.2.2. CAR molecule design and transfer plasmid amplification 

Second generation NKG2D- and BCMA-CAR molecules as well as ffLucGFP reporter 

gene were needed for the development of this thesis. CAR molecule design and 

lentivector production were performed in collaboration with the Division of 

Hematopoietic Innovative Therapies from CIEMAT, Madrid.  

NKG2D-CAR transfer plasmid (pTRPE_NKG2D-ECD_4-1BBz) was kindly provided by 

Daniel J. Powell Jr. (Perelman School of Medicine, Pennsylvania, USA). The 

extracellular domain contains the 82-216 amino acid sequence from the human NKG2D 

receptor ectodomain [465], and the intracellular region contains 4-1BB costimulatory 

domain, bound to the CD3ζ chain. 

For the BCMA-CAR transfer plasmid (pTRPE_BCMA_4-1BBz) design, the NKG2D-

CAR lentiviral backbone was used. This sequence was designed using SnapGeneTM 

software and synthesized at GeneArt (Thermofisher). NKG2D ectodomain sequence was 

replaced by the kappa light chain signal peptide (κLC) followed by clone 4ZFO α-BCMA 

single-chain variable fragment (scFv) (mouse/human chimera J22.9-xi mAb) and the 

hinge and transmembrane domain of the CD8a receptor. The sequence was flanked by 

the restriction sites of the Nhe I and Nde I enzymes (New England Biolabs) and inserted 

in NKG2D-CAR backbone using a one-step cloning with T4 DNA ligase (Biolabs). The 

product was transformed into One Shot Stbl3 chemically competent E.coli by heat shock 

(Thermofisher). 



MATERIALS AND METHODS 

85 
 

PRRL_Luc_EGFP transfer plasmid was used for the luciferase and GFP expression in U-

266 and RPMI-8226. 

For transfer plasmid DNA amplification, transformed bacteria were cultured over 

ampicillin selective plates containing 2% (p/v) Luria Bertani (LB) agar (Condalab) and 

50 ug/ml ampicillin and then, plates were incubated overnight at 37ºC. The following 

day, bacteria colonies were picked and cultured in ampicillin LB liquid medium 

(Condalab) overnight at 37ºC in an orbital shaker. Plasmid DNA was then extracted using 

QIAprep Spin Miniprep Kit (Qiagen) and evaluated by restriction digestion pattern using 

gel electrophoresis. After this step, 4-5 confirmed bacteria clones were selected and 

cultured in antibiotic-treated LB liquid medium overnight at 37ºC with constant shaking. 

Plasmid DNA was then extracted with QIAGEN Plasmid Maxi Kit (Qiagen) and 

restriction digestion pattern was corroborated. Plasmid DNA quantity and quality were 

measured using a Nanodrop ND-1000 Spectrophotometer (Thermo Fisher Scientific) and 

Sanger sequencing was carried out by the company Stab Vida to confirm the transfer 

plasmid sequence. 

 

3.2.3. Production of 3rd generation lentivector supernatants 

Once DNA transfer plasmids are amplified, HEK293T cells were used to produce 3rd 

generation self-inactivating lentivector supernatants containing NKG2D-CAR, BCMA-

CAR or ffLucGFP vectors. Nine million HEK293T cells were cultured in a treated-P150 

plate (ThermoFisher Scientific) and the following reagents were added to carry out the 

transfection: 4 plasmids, including pMD2.VSV.G (viral envelope; PlasmidFactory), 

pMDLg/pRRE (VIH-1 Gag/Pol plasmid; PlasmidFactory), pRSV-Rev 10 (transactivator; 

PlasmidFactory), pAdvantage™ (to enhance translation initiation, Promega Biotech) and 

the corresponding transfer plasmid; and 2.5M CaCl2 (Sigma), 2X HBS (containing 281 

mM NaCl, 100 mM HEPES, 1.5 mM Na2HPO4, at pH 7.12 and 0.22µm-filtered) and 

0.1X buffer TE (10 mM Tris (pH 8), 1 mM EDTA (pH 8); diluted 1:10 in sterile distilled 

H2O (dH2O) and 0.22 µm filtered). After 2 days, supernatants were collected and 0.45 

µm filtered (Fig. 19). Ultracentrifugation at 20,000 rpm during 2 hours at 4ºC was done 

to concentrate and purify the lentiviral particles. Later, the supernatant was removed, and 

lentiviral particles were resuspended in RPMI-1640 medium. After 30 minutes of 

shaking, lentivector supernatants were aliquoted and stored at -80ºC.  



MATERIALS AND METHODS 

86 
 

 

Figure 19. Schematic representation of lentivector production. Created with BioRender.com. 

 

3.2.4. Titration of lentivector supernatants 

To calculate the lentivector particles contained in the supernatants, 1 x 105 HEK293T/well 

were cultured in 24-well treated-plates (Thermo Fisher Scientific) using serial dilutions 

of the lentiviral particles. After 7 days, cells were collected and the transduction 

percentage in each dilution was tested by flow cytometry using recombinant human 

BCMA protein (rhBCMA), α-NKG2D Ab or GFP fluorescence (see 3.3.1.1 and table 9). 

Lentivector titer was determined, taking into account the number of cells initially 

cultured, and assuming one copy of vector per cell, by the following formula: 

Titer (viral particles/ml) = number of cells x % CAR+/GFP+ cells x Dilution/100. 

 

3.2.5. Lentiviral transduction 

Non-treated plates were coated with retronectin at 2 μg/cm2 24 hours before their use and 

stored at 4ºC or 3-4 hours before and stored at RT. Due to cell size variability, cells were 

plated differently: 1 x 106 NKAE cells were cultured at 1.33 x 106 cells/ml per well in 

non-treated 12-well plates and 5 x 105 MM cells were cultured at 1 x 106 cells/ml per well 

in non-treated 24-well plates. Moreover, a distinct multiplicity of infection (MOI) was 

also used for each lentivector: MOI 5 for all NKG2D-CAR NKAE cell transductions, 

MOI 8 BCMA-CAR and MOI 10 ffLucGFP. 



MATERIALS AND METHODS 

87 
 

Cells were then spinoculated at 2,000 rpm for 20 minutes at 24ºC and incubated at 37ºC. 

After 16 hours, cells were washed with PBS and resuspended in their corresponding 

medium (see 3.2.1 and 3.1.1). Only in the case of BCMA-CAR transduction, a second 

round of transduction was performed at MOI 8, following the conditions described above, 

thus, total MOI for this CAR is 16. 

After 6 days from transduction, stable CAR or ffLucGFP expression was determined by 

flow cytometry. 

 

3.2.6. Study of vector copy number (VCN) per cell in transduced NK cells   

In addition to flow cytometry transduction determination, VCN per cell was carried out. 

One week after transduction, DNA and RNA from CAR effector cells were first extracted 

using AllPrep DNA/RNA/Protein Mini Kit (Qiagen) following the manufacturer´s 

instructions. DNA and RNA quantity and quality were measured using the Nanodrop ND-

1000 (Thermo Fisher Scientific). RNA was stored at -80ºC for RNA-seq (described in 

3.11) and DNA at -20ºC for VCN. 

To determine VCN, the viral packaging signal sequence, Psi, was amplified by 

quantitative-PCR (q-PCR) using the Applied 7500 Fast Real-Time PCR system (Thermo 

Fisher Scientific) and the copy number was normalized to albumin. The primers used are 

included in Table 8. Albumin amplification was detected by a VIC fluorescent probe, and 

Psi sequence by a FAM probe. Duplicates were carried out per sample and VCN was 

calculated by interpolating the cycle threshold values (Ct) from the DNA sample to the 

Ct of a standard curve of a known concentration of dsDNA that contains albumin and Psi 

sequences.  

 

Psi forward (Psi.F) 5’ CAGGACTCGGCTTGCTGAAG 3’ 

Psi reverse (Psi.R) 5’ TCCCCCGCTTAATACTGACG 3’ 

Albumin forward (Alb.F) 5’ GCTGTCATCTCTTGTGGGCTG 3’ 

Albumin reverse (Alb.R) 5’ ACTCATGGGAGCTGCTGGTTC 3’ 

 

Table 8. Primers used to amplify Psi and albumin sequences. 

 

 



MATERIALS AND METHODS 

88 
 

3.3. Multiparametric flow cytometry and cell sorting 

3.3.1. Multiparametric flow cytometry  

Cells were collected (~1-2 x 105), washed with PBS, centrifuged at 1,400 rpm for 5 

minutes and resuspended in 100 μl of the Ab cocktail, prepared in PBS, in a 5 ml round 

bottom test tube (Falcon). Abs used are compiled in table 9. Cells were stained during 30 

minutes at 4ºC in darkness, then washed with PBS and resuspended in 100 μl of PBS with 

0.2 μg/ml 4',6-diamidino-2-phenylindole (DAPI). 

The flow cytometer FACSCanto™ II (BD Biosciences) contains 3 laser lines: 405 nm 

(violet), 488 nm (blue) and 633 nm (red), and detects 8 different fluorescence channels.  

Antigen Dilution Clone Fluorochrome Source 
Catalog 

number 

CD2 1:100 RPA-2.10 PE Biolegend 300208 

CD3 1:100 UCHT1 PE/Cy7 Biolegend 300420 

CD3 1:100 UCHT1 APC/Cy7 Biolegend 300426 

CD8 1:100 HIT8a PerCP/Cy5.5 Biolegend 300924 

CD16 1:100 3G8 APC/Cy7 Biolegend 302018 

CD19 1:100 SJ25C1 APC/Cy7 Biolegend 363010 

CD25 1:100 BC96 APC Biolegend 302610 

CD34 1:100 4H11 PE/Cy7 
Life 

Technologies 
25-0349-42 

CD34 1:100 4H11 APC 
Life 

Technologies 
17-0349-42 

CD38 1:100 LD38 FITC Cytognos CYT-38F 

CD45 1:100 2D1 PerCP/Cy5.5 Biolegend 368504 

CD56 (NCAM) 1:100 HCD56 APC Biolegend 318310 

CD56 (NCAM) 1:100 HCD56 PerCP/Cy5.5 Biolegend 318322 

CD56 (NCAM) 1:200 HCD56 PE Biolegend 318306 

CD57 1:100 HNK-1 PE Biolegend 359612 

CD69 1:100 FN50 PE Biolegend 310906 

CD62L 1:100 DREG-56 APC Biolegend 304810 

CD71 1:100 M-A712 APC-H7 
BD 

Biosciences 
655408 

Fas (CD95) 1:100 DX2 
Alexa Fluor 

488 
Biolegend 305616 

CD107a 1:100 H4A3 PE 
BD 

Biosciences 
555801 

CD138 1:100 MI15 PE/Cy7 Biolegend 356514 

CD138 1:100 MI15 BV 421 Biolegend 356516 

NKG2A (CD159a) 1:100 Z199 PE 
Beckman 

Coulter 
IM3291U 

NKG2C (CD159c) 1:100 134591 
Alexa Fluor 

488 
R&D Systems FAB138G 

KIR2DL2/L3 

(CD158b) 
1:100 DX27 PerCP/Cy5.5 Biolegend 312614 



MATERIALS AND METHODS 

89 
 

KIR3DL1 

(CD158e1) 
1:100 DX9 PE/Cy7 Biolegend 312720 

KIR3DL2/CD158k 1:100 539304 PE R&D Systems FAB-2878P 

KIR2DL1/KIR2DS5 1:100 143211 
Alexa Fluor 

488 
R&D Systems FAB1844G 

FasL (CD178) 1:100 NOK-1 APC 
Life 

Technologies 
17-9919-42 

CCR7 (CD197) 1:100 G043H7 PE Biolegend 353204 

IL-15Rα (CD215) 1:100 JM7A4 PE Biolegend 330208 

DNAM-1 (CD226) 1:100 TX25 FITC Biolegend 337104 

TRAIL (CD253) 1:100 RIK-2 PE Biolegend 308206 

PD-1 (CD279) 1:100 EH12.2H7 PE Biolegend 329906 

NKG2D (CD314) 1:100 149810 PE R&D Systems FAB139P 

NKp46 (CD335) 1:100 29A1.4 PE Biolegend 137604 

NKp44 (CD336) 1:100 P44-8 PE Biolegend 325108 

NKp30 (CD337) 1:100 P30-15 PE Biolegend 325208 

KLRG1 1:100 14C2A07 PE Biolegend 368610 

TIGIT 1:100 A15153G PE Biolegend 372704 

GLUT1 1:100 202915 
Alexa Fluor 

488 
R&D Systems FAB1418G 

Ki67 1:100 Ki-67 
Brilliant Violet 

421™ 
Biolegend 350506 

BCMA (CD269) 1:100 19F2 PE Biolegend 357504 

HLA-E 1:100 
3D12HLA-

E 
APC ThermoFisher 17-9953-42 

ULBP2/5/6 1:100 165903 PE R&D Systems FAB1298P 

ULBP1 1:100 170818 PE R&D Systems FAB1380P 

ULBP3 1:100 166510 PE R&D Systems FAB1517P 

MICA/B 1:100 6D4 PE Biolegend 320906 

IFN-γ 1:100 45-15 PE Miltenyi Biotec 
130-113-

493 

 

Other reagents Dilution Clone Fluorochrome Source 
Catalog 

number 

Biotinylated 

rhBCMA (CD269) 
1:200 - - Adipogen 

ANC-519-

030 

IgG1, κ 1:100 MOPC-21 PE BD Biosciences 555749 

IgG1, κ 1:100 P3.6.2.8.1 APC 
Life 

Technologies 
17-4714-42 

IgG2a, κ 1:100 
MOPC-

173 
PE Biolegend 400211 

IgG2b, κ 1:100 MPC-11 PE Biolegend 400314 

Streptavidin 1:300 - PE BD Biosciences 554061 

Propidium Iodide - - - Sigma Aldrich P4864 

DAPI - - - Sigma Aldrich D9542 

 

Table 9. Reagents and primary Abs used for flow cytometry. FITC: Fluorescein isothiocyanate; PE: 

Phycoerythrin; PerCP/Cy5.5: Peridinin chlorophyll protein-Cyanine5.5; PE/Cy7: PE-Cyanine7; APC: 

Allophycocyanin; APC/Cy7: APC-Cyanine7; DAPI: 4',6-diamidino-2-phenylindole. 

 



MATERIALS AND METHODS 

90 
 

Using FACSDiva v.8.0.1 software, debris was first excluded by Forward Side 

Channel-Area (FSC-A) vs Side Scatter Channel-Area (SSC-A). Once cells are gated, 

single cells were selected by FSC-A and FSC-Height (FSC-H), avoiding aggregates, 

which show a deviation from the linearity. To analyze live cells, DAPI- cells were 

selected, determining the population of interest. A range of 10,000-20,000 live cells per 

tube were acquired in most experiments. Data analysis was later performed using FlowJo 

Vx 10.0.7 software (BD Biosciences).  

Unstained controls were needed to set the laser voltages and single-stained controls were 

used to compensate the spectral overlap among fluorochromes. Fluorescence minus one 

(FMO) controls were needed to identify the additive effect of spreading from multiple 

fluorochromes into the detector of interest. Relative fluorescence intensity (RFI) was then 

calculated as the mean fluorescence intensity of Ab staining divided by the FMO, isotype 

or unstained. 

 

3.3.1.1. Immunophenotype 

Characterization of CAR effector cell and MM cell line phenotype was mainly performed 

by flow cytometry. CAR-target ligands (NKG2D-L and BCMA) on MM cells as well as 

CAR expression on NKAE cells (NKG2D-CAR and BCMA-CAR) were evaluated by 

this technique. NKG2D-CAR expression on NKAE cells was detected with a human α-

NKG2D Ab, while BCMA-CAR expression was analyzed on NKAE cells by the 

incubation with a rhBCMA protein conjugated to biotin and detected by streptavidin-PE. 

QuantibriteTM beads (Biolegend) were used to calculate the number of NKG2D molecules 

in NKG2D-CAR populations. 

In addition, the following expression markers were studied on the cell surface of CAR 

NKAE cells. 

• Activating or activation receptors: CD69, NKp30, NKp44, NKp46 and NKG2C 

• Inhibitory or exhaustion receptors: KIR2DL1/2DS5, KIR2L2/L3, KIR3DL1, 

KIR3DL2, NKG2A, PD-1, TIGIT and KLRG1 

• Adhesion receptors: DNAM-1 and CD62L 

• Costimulatory receptors: CD2 

• Maturation: CD57 



MATERIALS AND METHODS 

91 
 

• Cytokine receptors: CD25 (IL-2Rα), IL-15Rα and TGFBR-II  

• Apoptosis induction: TRAIL and FasL  

• Metabolic receptors: GLUT1 and CD71 

• ADCC and degranulation: CD16 and CD107a 

• Other receptors: Fas (FasL receptor) and CCR7 

 

HLA-E expression on primary MM cells as well as NKG2C, NKG2A and PD-1 

expression on cytotoxic T and NK cells from BM samples were analyzed using the panels 

below (table 10). We differentiated PC population into pathologic PC (pPC; CD138+ 

CD38+ CD19- CD56+) and normal PC (nPC; CD138+ CD38+ CD19- CD56+), following 

flow cytometry phenotype described for each patient by the Hematology Service from 

H12O. Then, RFI was calculated for each population. In order to obtain HLA-E 

expression ratio between pPC and healthy populations, pPC RFI was divided by nPC or 

CD34+ RFI. 

 

HLA-E expression on PC 

 450 FITC PE PerCP/Cy5.5 PE/Cy7 APC APC/Cy7 

HLA-E CD138 CD38 CD56 CD45 CD34 HLA-E CD19 

IgG1κ CD138 CD38 CD56 CD45 CD34 IgG1κ CD19 

 

NKG2C, NKG2A and PD-1 expression on cytotoxic T and NK cells 

 450 FITC PE PerCP/Cy5.5 PE/Cy7 APC APC/Cy7 

T/NK cells 1 DAPI NKG2C NKG2A CD8 CD3 CD56 CD16 

T/NK cells 2 DAPI - PD-1 CD8 CD3 CD56 CD16 

FMO DAPI - - CD8 CD3 CD56 CD16 

 

Table 10. Description of Abs used for HLA-E expression on PC and NKG2C, NKG2A and PD-1 

expression on CD8+ T cells and CD56+CD16+ NK cells. 

 

3.3.2. U-266 and RPMI-8226 ffLucGFP cell line sorting 

U-266 and RPMI-8226 cells were transduced with the ffLucGFP vector, as described in 

3.2.5. After one week, GFP signal was detected by flow cytometry to confirm 

transduction and GFP+ cells were sorted using BD FACSAria™ Fusion Cell Sorter (BD 

Biosciences), which contains 3 laser lines: 405 nm (violet), 488 nm (blue) and 640 nm 

(red). 



MATERIALS AND METHODS 

92 
 

3.4. Cytotoxicity assays 

3.4.1. Calcein release cytotoxicity assay against MM cell lines 

CAR CIML-NKAE and NKAE cell cytotoxicity against MM cell lines was evaluated in 

vitro through calcein release assay. Calcein-acetoxymethylester (calcein-AM) is a 

lipophilic ester that penetrates the plasma membrane. It is hydrolyzed in the cytosol by 

esterase enzymes to a polar green-fluorescent product (calcein) that is retained into the 

cells with intact membrane. When a cell is destroyed, calcein is released to the supernatant 

and the cytotoxicity can be evaluated due to the fluorescence detected in the supernatants 

of co-cultures.  

Target MM cells (50,000 per well) were stained with 3 μM Calcein-AM during 30 min at 

37ºC in RPMI-1640 medium with 100 IU/ml P/S. Target cells were then washed with 

PBS and co-cultured with CAR NK cells at different target:effector (T:E) ratios. 

Moreover, spontaneous lysis (calcein-stained target cells alone, in the absence of effector 

cells) and maximum lysis (calcein-stained target cells with 1% Triton X 100; Sigma 

Aldrich) controls were included. Every condition was carried out per triplicate in a non-

treated U-bottom 96-well plate (Corning) using NK cell corresponding medium without 

IL-2. Plates were centrifuged for 5 minutes at 1,200 rpm and incubated for 3 hours at 

37ºC. After the incubation, plates were again centrifuged, and supernatant was transferred 

into a 96-well black microplate (Corning). Fluorescence (excitation 488 nm; emission 

520 nm) was detected by the spectrophotometer VICTOR Nivo™ (PerkinElmer) (Fig. 

20) and specific lysis was calculated using the following formula: 

Specific lysis (%) = 
Co−culture lysis fluorescence −Spontaneous lysis fluorescence

Maximun lysis fluorescence−Spontaneous lysis fluorescence
x 100 

 

In NKG2A blocking assays, CAR NKAE cells were pre-treated with 10 μg/ml mouse α-

NKG2A clone Z199 (Beckman Coulter) or its corresponding isotype IgG2b (R&D 

systems) or human α-NKG2A BMS-986315 (Bristol-Myers-Squibb) or its corresponding 

isotype IgG1κ (Bristol-Myers-Squibb) Abs for 30 minutes at 37ºC. Later, blocked CAR 

NKAE cells were incubated with target calcein-stained cells, as described above. 



MATERIALS AND METHODS 

93 
 

 

Figure 20. Schematic representation of calcein release cytotoxicity assay. Created with BioRender.com. 

 

3.4.2. Clonogenic cytotoxicity assay against MM cell lines 

Cytotoxicity of CAR CIML-NKAE and NKAE cells was also performed against the MM 

cell line L-363. Around 40-50% of these tumor cells are clonogenic, which can grow 

forming colonies in semi-solid culture medium [466].  

CAR CIML-NKAE or NKAE cells were co-cultured at different T:E ratios with L-363 

cells, in non-treated U-bottom 96-well plates, in NK cell medium without IL-2. To 

establish the maximum number of colonies, L-363 cells were cultured in the absence of 

CAR NKAE cells and a control with NK cells alone was also included. Plates were 

centrifuged and incubated at 37ºC for 3 hours. Then, each co-coculture was seeded over 

methylcellulose-based medium without cytokines (MethoCult™ H4230; Stem cell) per 

triplicate in p35mm plates (Corning). Each plate contained 250 L-363 cells. After 14 days 

of incubation at 37ºC, colonies were counted using an inverted bright-field microscope 

(Nikon Diaphot, objective 4X).  

 

3.4.3. Cytotoxicity assay against primary MM cells through flow cytometry 

CAR NKAE cell cytotoxicity over primary MM cells and toxicity over CD34+ cells were 

performed by flow cytometry at 24 hours in a native context. 

BM samples from MM patients at diagnosis or progression were centrifuged at 350 g for 

10 minutes and plasma was collected and stored at 4 ºC. Then, BM samples were diluted 



MATERIALS AND METHODS 

94 
 

1:10 with PBS and added over 2:1 Ficoll-Paque. BM mononuclear cells (BMMCs) layer 

was collected, washed with PBS, and centrifuged at 1,600 rpm for 10 minutes. 1-2 x 105 

BMMCs were stained to analyze the percentage of PC in each sample (CD138+CD38+) 

by flow cytometry. 30,000 PC were plated per well within BMMCs and co-cultured with 

NKG2D-CAR NKAE cells at 1:10 T:E ratio or BCMA-CAR NKAE cells at 1:5 T:E ratio 

(pretreated with the α-NKG2A Ab or the corresponding isotype, as described in 3.4.1). A 

control with BMMCs alone was also used to determine the maximum PC survival within 

the experiment. Each condition was plated per triplicate in a non-treated U-bottom 96-

well plate using the NK cell medium, without IL-2, supplemented with 6% of the BM 

plasma. Plasma is added to maintain the TME present in the MM BM, which sustain PC 

survival, and could negatively impact in the adoptive cell therapy efficacy. Plates were 

centrifuged for 5 minutes at 1,200 rpm and incubated for 24 hours at 37ºC. After this time, 

plates were centrifuged again, supernatants were removed, and cells were washed with 

PBS. Then, cells were stained with CD38 FITC, CD138 PE/Cy7, CD45 PerCP/Cy5.5, 

CD34 APC and CD56 PE, as described in 3.3.1. After washing the staining, cells were 

resuspended in 150 µl of PBS-DAPI and 120 µl from each well were acquired by the 

cytometer at 500 µl/min. This experiment was performed using the Attune CytPix Flow 

Cytometer (Thermofisher), located in Vivia Biotech company. This cytometer allows the 

analysis of a concrete volume of sample, assuring more accurate results. The number of 

CD38+ CD138+ cells as well as CD34+ cells that remained in each condition was then 

analyzed. PC cytotoxicity and CD34+ toxicity were calculated using the following 

formula:  

Survival (%) =  
Number of PC or CD34+ cells per condition

Number of PC or CD34+ in the absence of effector cells
x 100 

 

Cytotoxicity or toxicity (%) = 100 - survival (%) 

 

3.4.4. Degranulation assay 

CAR CIML-NKAE and NKAE cell antitumor activity was also studied by measuring 

their degranulation levels. U-266, XG-1, ARP-1, RPMI-8226 and K562-mb21-41BBL 

target cells, 1 x 105 cells, were co-cultured with effector cells at 1:1 ratio per duplicate in 

NK cell medium without IL-2. CAR NK cells without target cells were also seeded as 

basal degranulation control. In α-NKG2A blocking assays, CAR NK cells were pre-



MATERIALS AND METHODS 

95 
 

treated as described in 3.4.1. CD107a PE or its isotype, IgG1κ PE, were added to each 

condition and incubated for 1 hour at 37ºC. Then, GolgiStop (BD) was added to all wells, 

following the manufacturer’s indications, and plates were centrifuged and incubated for 

3 more hours. Later, cells were centrifuged, washed with PBS and stained with CD56 

PerCP/Cy5.5 (XG-1 and K562-mb21-41BBL co-cultures and CAR NK cell controls), 

CD138 PE/Cy7 (U-266 and ARP-1 co-cultures) or CD45 PerCP/Cy5.5 (RPMI-8226 

co-cultures) Abs, as described in 3.3.1, to discriminate NK cells from MM cell lines in 

each case. Cells were resuspended in PBS-DAPI solution and acquired by the cytometer. 

CD107a+ cell percentage within live NK cells was analyzed. 

 

3.5. Cytokine detection assays 

3.5.1. Intracellular IFN-γ detection assay 

To detect intracellular IFN-γ production in CAR CIML-NKAE and NKAE cells, 1 x 105 

target cells (U-266, XG-1, ARP-1, RPMI-8226 and K562-mb21-41BBL) were co-

cultured with effector cells at 1:1 ratio per duplicate using NK cell medium without IL-2 

in non-treated U-bottom 96-well plates. GolgiStop was added and plates were centrifuged 

and incubated at 37ºC for 4 hours. Later, cells were centrifuged again, washed with PBS 

and stained with CD56 PerCP/Cy5.5 (XG-1 and K562-mb21-41BBL co-cultures and 

CAR NK cell controls), CD138 PE/Cy7 (U-266 and ARP-1 co-cultures) or CD45 

PerCP/Cy5.5 (RPMI-8226 co-cultures) Abs, to discriminate CAR NKAE cells from MM 

cell lines. After 30 minutes, cells were centrifuged, washed with PBS with 0.5% Bovine 

Serum Albumin (BSA) and 2 mM Ethylenediaminetetraacetic acid (EDTA) (PBE), 

centrifuged at 1,400 rpm for 5 minutes and resuspended in fix Solution A (Fix&Perm, 

Cytognos). After 10 minutes at RT, cells were washed with PBE buffer and resuspended 

in permeabilization Solution B (Fix&Perm, Cytognos) with the IFN-γ PE Ab or its isotype 

IgG1κ PE for 30 minutes at RT. Cells were washed again with PBE and resuspended in 

PBS to be acquired by the cytometer. IFN-γ+ cell percentage within live NK cells was 

analyzed. 

 

 

 



MATERIALS AND METHODS 

96 
 

3.5.2. Soluble cytokines quantification assay 

CAR NKAE effectors, 50,000 cells, were cultured with target cells (U-266, XG-1, RPMI-

8226 and K562-mb21-41BBL) at 1:2 T:E ratio in NK cell medium, without IL-2, in non-

treated U-bottom 96-well plates. For α-NKG2A studies, CAR NKAE cells were blocked 

with α-NKG2A Ab or its corresponding isotype for 30 minutes before the co-culture, as 

described in 3.4.1. After 24 hours, supernatants were collected and stored at -80ºC. 

Quantification of soluble cytokines from NK and target cell co-culture supernatants was 

performed by flow cytometry using LEGENDplex™ Human CD8/NK Panel (13-plex) 

(Biolegend), following the manufacturer´s instructions (Fig. 21). This panel allows 

simultaneous quantification of 13 human proteins, including IL-2, IL-4, IL-10, IL-6, IL-

17A, TNF-α, sFas, sFasL, IFN-γ, granzyme A, granzyme B, perforin, and granulysin. 

Besides, a customized LEGENDplex™ panel (IL-1β, IFN-α2, IFN-γ, TNF-α, MCP-1, IL-

6, IL-8, IL-10, IL-18, IL-15, IL-2, IL-3, TGF-β1) was designed for cytokines detection in 

MM patients and HD BM plasmas.  

LEGENDplex™ Data Analysis Software Suite was then used to analyze the flow 

cytometry data files. 

 

 

Figure 21. Schematic representation of LEGENDplex™ assay. Obtained from LEGENDplexTM protocol 

guide (Biolegend).  

 

 

 



MATERIALS AND METHODS 

97 
 

3.6. Toxicity assays 

3.6.1. Toxicity against PBMCs 

Evaluation of NKG2D- and BCMA-CAR NKAE cell toxicity was carried out against 

PBMCs from HD. PBMCs were isolated by density gradient centrifugation, as described 

in 3.2.1. Then, PBMCs were stained with calcein-AM and co-culture with the NKAE 

populations in NK cell medium without IL-2, as described in 3.3.1. 

 

3.6.2. Toxicity against CD34+ hematopoietic cells from CB 

The hematotoxicity of CAR CIML-NKAE and NKAE cells was evaluated over CD34+ 

cells isolated from CB samples. First, mononuclear cells were obtained by Ficoll density 

gradient centrifugation. Cells were then lysed with ACK lysis buffer (Gibco) and CD34+ 

cells were purified by immunomagnetic positive selection using CD34 MicroBead Kit 

(Miltenyi), following the manufacturer´s instructions. Purification of CD34+ cells was 

confirmed by flow cytometry (see 3.3.1) and these cells were frozen in 80% FBS with 

20% dimethyl sulfoxide (DMSO). 

On the day of the experiment, CD34+ cells were thawed in NK cell medium without 

cytokines. Thereafter, CAR CIML-NKAE or NKAE cells were co-cultured with CD34+ 

cells at different T:E ratios, using NK cell medium without IL-2, in non-treated U-bottom 

96-well plates. To establish the maximum number of colonies, CD34+ cells were cultured 

in the absence of effector cells and a negative control with NK cells alone was included. 

Plates were centrifuged and incubated at 37ºC, in normoxic conditions, for 4 hours. Then, 

each condition was seeded over a methylcellulose-based medium containing recombinant 

human stem cell factor (SCF), IL-3, EPO and granulocyte-macrophage colony-

stimulating factor (GM-CSF) (MethoCult™ H4434; Stem cell), per triplicate in p35mm 

plates. After 14 days incubation at 37ºC, Colony Forming Units-Granulocyte-

Macrophage (CFU-GM) and Burst Forming Units-Erythroid (BFU-E) were counted 

using an inverted bright-field microscope (Nikon Diaphot, objective 4X).  

 

 

 



MATERIALS AND METHODS 

98 
 

3.7. Proliferation assays 

3.7.1. Cell cycle analysis 

NKG2D-CAR CIML-NKAE or NKAE cells, 5 x 105 cells, were collected and washed 

with PBS in 5 ml round bottom test tubes. Cells were fixed by adding 70% cold ethanol 

drop by drop while vortexing for 1 minute. Samples were stored at -20ºC for at least 12 

hours. 

Before cell cycle analysis by flow cytometry, samples were washed twice with PBS and 

cell pellets were resuspended in PBS and incubated with 0.2 mg/ml RNAse A for 30 

minutes in darkness at RT. Propidium iodide (PI) at 0.04%, was added to each tube and 

PI staining was then analyzed by flow cytometry in linear scale (see 3.3.1). 

 

3.7.2. Ki67 cell proliferation analysis 

Ki67 protein expression indicates cell proliferation because this marker can be detected 

when cells are in G1, S, G2 and mitosis phases, but not in G0 [467]. 

As described for cell cycle assay, 5 x 105 NKG2D-CAR CIML-NKAE cells or NKAE 

cells were collected, washed, fixed with ethanol and stored at -20ºC for further analysis. 

Cells were then washed three times with Cell Staining Buffer (Biolegend) and stained 

with the Ki67 BV421 Ab. After 30 minutes in darkness at RT, cells were washed twice 

with this buffer and resuspended in 0.5ml of buffer for flow cytometry analysis (see 

3.3.1). 

 

3.8. Metabolism assays 

3.8.1. MitoTracker staining 

NKG2D-CAR CIML-NKAE or NKAE effectors, 1 x 105 cells, were collected, washed 

with PBS in 5 ml round bottom test tubes, and incubated for 10 min at 37ºC with 100 nM 

Mitotracker Green (Life Technologies). After incubation, cells were rinsed with PBS, 

centrifuged, and resuspended with PBS-DAPI solution prior to flow cytometry 

acquisition (see 3.3.1). 



MATERIALS AND METHODS 

99 
 

3.8.2. Mitochondrial Oxidative Phosphorylation (OXPHOS) system function 

Mitochondrial and glycolysis stress experiments with NKG2D-CAR CIML-NKAE or 

NKAE cell populations were performed by the Rare, Mitochondrial and Neuromuscular 

Diseases group from H12O, Madrid. 

To assess the OXPHOS system function, Oxygen Consumption Rate (OCR) was 

determined through high-resolution respirometry in a Seahorse XFp Extracellular Flux 

Analyzer (Agilent, Santa Clara, California; Fig. 22A). Mitochondrial stress test was 

performed through sequential injections of 2.5 µM oligomycin, 2 injections of fluoro-

carbonyl cyanide phenylhydrazone (FCCP), first one of 1.1 µM and the second one of 2.5 

µM, and then a mix of 1.3 µM rotenone and 1.3 µM antimycin A, prepared from XFp 

Cell Mito Stress Test (Agilent) reagents. Oligomycin, a complex V inhibitor, inhibits the 

ATP synthase and diminishes OCR; the mitochondrial uncoupler FCCP disrupts ATP 

synthesis, obtaining the maximal OCR; and rotenone and antimycin A inhibit the electron 

transport chain and decrease the OCR to a minimal value (Fig. 22B). 

NKG2D-CAR CIML-NKAE or NKAE cells, 30,000 per well, were resuspended in XF 

Seahorse Base pH 7.4 medium supplemented with 4.44 mM glucose, 0.08 mM sodium 

pyruvate and 0.66 mM glutamine and seeded on a XFp Cell Culture Miniplate. This plate 

was incubated at 37ºC without CO2 for 20 minutes. Then, the plate was introduced into 

the analyzer and sequential injections of the indicated compounds were carried out. The 

test determines the following parameters: 

- Basal OCR: OCR obtained before any compounds were added minus non-

mitochondrial OCR (nmOCR). 

- Oligomycin-sensitive OCR or ATP-linked respiration: it calculates the amount of 

OCR needed to pump protons that complex V would use to produce ATP. It is the 

basal OCR minus OCR after oligomycin injection. 

- Proton leak: it determines the proton flux that is not directed to ATP production. 

It is calculated as the oligomycin-sensitive OCR minus nmOCR. 

- Maximal OCR (maxOCR): FCCP forces a high energy demand to calculate the 

highest OCR the cell can achieve. It is the OCR value obtained after FCCP 

injection minus nmOCR. 

- Spare capacity: it is the reaction capacity of a cell to an energy demand. It is the 

difference between the maxOCR and the basal OCR. 



MATERIALS AND METHODS 

100 
 

- NmOCR: it is measured after rotenone and antimycin A injection, when all the 

electron transport chain is inhibited. 

 

 

Figure 22. Schematic representation of Seahorse OCR test. A) Seahorse XF Cell Mito Stress test profile. 

B) Target of action of Mito Stress test modulators on the electron transport chain. Obtained from Agilent. 

 

Moreover, the extracellular acidification rate (ECAR), which measures the concentration 

of free protons, was determined using the XFp Glycolysis Stress Test in the Seahorse XFp 

Extracellular Flux analyzer. For this assay, cells were resuspended in the same XF 

Seahorse Base medium without glucose and supplemented with 100 mM sodium pyruvate 

and 200 mM glutamine. Three sequential injections of 25 mM glucose, 2.6 µM 

oligomycin and 100 mM 2-desoxy-D-glucose (2-DG) were carried out to determine the 

following parameters (Fig. 23): 

- Non-glycolytic acidification: acidification that is not attributed to glycolysis. It is 

obtained after 2-DG injection. 

- Basal glycolysis: basal glycolytic activity of the cell, obtained from ECAR value 

after glucose injection minus non-glycolytic acidification or initial ECAR. 

- Maximal glycolytic capacity: obtained after mitochondrial ATP production 

disruption, due to oligomycin injection. It is determined by ECAR value after 

oligomycin injection minus initial ECAR. 

- Glycolytic reserve: capacity to increase glycolytic activity in response to an 

increase in energy demand. It is determined as ECAR after oligomycin addition 

minus basal glycolysis. 



MATERIALS AND METHODS 

101 
 

 

 

Figure 23. Schematic representation of Seahorse ECAR test. A) Seahorse XF Glycolysis Stress test 

profile. B) Target of action of Glycolysis Stress test modulators. Obtained from Agilent 

 

All measurements were normalized to the amount of protein in each well. Cells seeded in 

the microplates were lysed and proteins were quantified by Bradford assay (Bio-Rad). 

 

3.9. Western Blot (WB) 

To extract proteins from CAR NK cells, cell pellets were lysed in RIPA lysis buffer with 

protease (Roche) and phosphatase inhibitors (Merck). Samples were incubated for 1 hour 

on ice, vortexing occasionally. Then, they were centrifuged at 15,000 rpm for 30 minutes 

at 4ºC and supernatants, containing the extracted proteins, were collected. Protein 

samples were diluted in Laemmli loading buffer (Bio-Rad) containing sodium dodecyl 

sulfate (SDS) and 5% β-mercaptoethanol and incubated at 99ºC for 10 minutes in a block 

heater. These proteins were separated following a SDS-polyacrylamide gel 

electrophoresis (SDS-PAGE) under denaturing conditions. Electrophoresis was 

performed at constant voltage (120V) in electrophoresis buffer (25 mM Tris-HCl, 200 

mM glycine, 0.1% SDS, pH 8.3). 

After electrophoresis, proteins were transferred from the gels to nitrocellulose membranes 

(Amersham) at a constant current (350 mA) for 90 minutes using wet transfer. Then 

membranes were blocked with PBS containing 3% of BSA for 30 minutes at RT in 



MATERIALS AND METHODS 

102 
 

agitation. After this step, membranes were washed with TBS with 0.1% Tween20 (TBS-

T) and incubated with primary Abs overnight. The following day, membranes were 

washed three times with TBS-T and incubated with the secondary anti-mouse or anti-

rabbit Abs conjugated to Horseradish Peroxidase (HRP; Cell Signaling) for 30 minutes. 

Then, membranes were again washed three times with TBS-T and Abs bound to the target 

proteins were detected using commercial enhanced chemiluminescence (ECL) kits 

(Amersham) in the Gel DocTM EZ Documentation System (Bio-Rad). 

 

3.9.1. Detection of CAR expression on CAR NKAE cells 

Five million CAR NKAE cells were pelleted and processed, as previously described, to 

analyze the presence of CAR molecules in transduced NKAE cells. The CD3ζ signaling 

domain is present in the NKG2D- and BCMA-CAR sequences. Therefore, an α-CD3ζ 

primary Ab (#551033, BD Pharmigen) was used to identify the expression of CAR 

molecules within NKAE cell populations. P150 (#610473, BD Biosciences) detection 

was utilized as loading control. 

 

3.9.2. NKG2A signaling pathway 

To evaluate the phosphorylation state of ZAP-70/Syk, involved in NKG2A signaling 

pathway (Fig. 16), 5 x 106 isotype‑ or NKG2A‑pretreated CAR NKAE cells were 

stimulated with 5 x 105 RPMI‑8226 cells for 1h in NK medium with 2% AB serum and 

then pelleted. CAR NKAE cells as well as RPMI-8226 cells alone were also collected. 

Pellet samples were processed, as previously described, and a specific Ab against 

phospho-Zap-70 (Tyr319)/Syk (Tyr352) (#2701S; Cell Signaling) was used. ZAP‑70 

(#2705S, Cell Signaling) detection was utilized as loading control. 

 

3.10. Mass cytometry 

Surface and intracellular protein expression was studied by mass cytometry. For this 

purpose, 5 x 106 NKG2D-CAR NKAE cells were pretreated with either the human α-

NKG2A or its corresponding isotype IgG1κ, co-cultured at 1:10 T:E ratio with the MM 

cell line RPMI-8226 for 24 hours and frozen in 90% FBS + 10% DMSO. This experiment 



MATERIALS AND METHODS 

103 
 

was carried out and analyzed by the Department of Medical Oncology from Dana Farber 

Cancer Institute, Boston, USA. 

On the day of the experiment, cells were quickly thawed in a water bath, resuspended in 

PBS and pelleted. Cells were then stained with a viability stain. After washing with 

Maxpar cell staining buffer (Fluidigm), samples were stained with a cocktail of surface 

Abs (Table 11) (Standard Bio) for 15 minutes at RT. Next, cells were washed and fixed. 

After washing, cells were stained with Cell ID Intercalator Ir to label nucleated cells. 

Then, cells were rinsed with PBS and stored until acquisition on the same day. Samples 

were acquired on the Fluidigm XT (Standard Bio) mass cytometer. Data were analyzed 

with FlowJo and R programming. 

 

Antigen Metal 

CD16 145Nd 

CD27 167Er 

CD56 176Yb 

CD94 170Er 

CD117 (c-kit) 143Nd 

CD122 (IL-2/15Rβ) 156Gd 

CD127 (IL-7Rα) 165Ho 

CD158b 

(KIR2DL2/DL3) 
173Yb 

CD159a (NKG2A) 169Tm 

CD161 164Dy 

CD294 (CRTH2) 163Dy 

CD314 (NKG2D) 166Er 

CD328 (SIGLEC7) 152Sm 

CD335 (NKp46) 162Dy 

CD336 (NKp44) 155Gd 

CD337 (NKp30) 159Tb 

KIR3DL3 147Sm 
 

Table 11. Abs used for CyTOF experiment. 

 

3.11. Transcriptomic analysis by RNA sequencing (RNA-

seq) 

NKG2D-CAR CIML-NKAE and NKAE cells RNA expression differences from 4 

experiments were studied by RNA-seq. This technique was performed by the Department 



MATERIALS AND METHODS 

104 
 

of Hematology from H12O, Madrid, and analyzed by the Department of Biochemistry 

and Molecular Biology from Universidad Complutense de Madrid. 

RNA from these populations was extracted, as described in 3.2.6, and total RNA quality 

was determined in Bioanalyzer 2100 (Agilent Technologies). For this purpose, Agilent 

RNA 6000 Pico kit (Agilent Technologies) was used following the manufacturer’s 

instructions. 

RNA library generation was performed from 10 ng to 1 µg of total RNA following the 

KAPA RNA HyperPrep Kit with RiboErase (HMR) (KR1351 – v1.16) protocol for 

Illumina (San Diego, USA). KAPA RNA HyperPrep Kit and KAPA HyperPlus Kit 

(Roche) were utilized. 

First, oligonucleotides hybridization to rRNA was performed, rRNA was depleted 

through RNAse H and RNA was purified using 2.2X KAPA Pure Beads (Roche). 

Subsequently, hybridized oligonucleotides were digested by DNAse and RNA was again 

purified with 2.2X KAPA Pure Beads. rRNA-free RNA was then eluted, heat-fragmented 

and the first cDNA chain was generated. Afterwards, a second cDNA was synthesized 

and SeqCap Adapter Kit B (Roche) was used for adapter ligation. 

Later, samples were again purified using 0.63X and 0.7X KAPA Pure Beads and libraries 

were amplified utilizing HiFi polymerase. Libraries were 1X KAPA Pure Beads purified 

and quantified using KAPA Library Quant Kit (Illumina) ROX Low qPCR Mix (Roche), 

according to Ct comparison regarding 6 standards of Ct known concentrations (20 pM, 2 

pM, 0.2 pM, 0.02 pM, 0.002 pM and 0.0002 pM). 

Then, libraries were diluted until 4 nM, denaturalized, and again diluted following the 

NextSeq System: Denature and Dilute Libraries Guide (15048776 v09), until 1.8 pM. The 

8 samples from 4 paired experiments were loaded in a chip and sequenced in NextSeq 

500/550 System (Illumina) (Fig. 24). 



MATERIALS AND METHODS 

105 
 

 

Figure 24. Schematic representation of RNA-seq protocol. Obtained from BioRender gallery. 

 

Time and temperature specifications of each process developed in ProFlexTM Base 

(Applied Biosystems) or QuantStudioTM 5 Real-Time PCR Instrument (96-Well 0.1 ml 

Block; Applied Biosystems) are detailed in table 12. 

 

 

 

 

 

 

 

 

 

 

 

 



MATERIALS AND METHODS 

106 
 

Procedure Cycles Conditions 

Oligonucleotides 

hybridization and 

rRNA depletion 

1 95ºC / 2’ 

1 45ºC: - 0.1ºC / second 

1 45ºC / ∞ 

1 45ºC / 30’ 

1 4ºC / ∞ 

DNA digestion 
1 37ºC / 30’ 

1 4ºC / ∞ 

Fragmentation 1 85ºC / 6’ 

First cDNA chain 

synthesis 

1 25ºC / 10’ 

1 42ºC / 15’ 

1 70ºC / 15’ 

1 4ºC / ∞ 

Second cDNA chain 

synthesis 

1 16ºC / 30’ 

1 62ºC / 10’ 

1 4ºC / ∞ 

Libraries 

amplification 

1 98ºC / 45’’ 

11-15 

98ºC / 15’’ 

60ºC / 30’’ 

72ºC / 30’’ 

1 72ºC / 1’ 

1 4ºC / ∞ 

Quantitative PCR 

1 95ºC / 5’ 

35 
95ºC / 30’’ 

60ºC / 45’’ 

1 95ºC – 65ºC – 95ºC 
 

Table 12. Detailed description of all procedures carried out along RNA sequencing. 

 

RNA-seq data processing and analysis workflow were performed following these steps. 

First of all, RNA-seq data quality control was carried out using FastQC, to test sequencing 

quality and adapter presence. Then, reads were mapped to the homo sapiens GRCh38 

genome version (available in GENCODE v29), not only to align the reads to the reference 

genome but also for gene annotation. Fastq RNA-seq data were mapped using STAR-

2.7.0e and gene abundances were internally estimated using the HT-seq algorithm. Before 

the differential expression analysis, expression genes were filtered, keeping the genes 

whose read counts are above 1, and normalized, considering the estimates of the library 

size. Differential expression analysis was then performed using DESeq2 [468]. In order 

to detect differentially expressed genes between NKG2D-CAR NKAE and NKG2D-CAR 

CIML-NKAE cells, analysis was performed comparing differences between these 

populations, taking into account HD variability. Finally, principal components (PC) 

analysis was carried out, selecting the 3,000 genes with the higher variability among NK 

cell populations. PC1 reveals the most variation and PC2, the second most variation. 



MATERIALS AND METHODS 

107 
 

3.12. Bulk DNA methylation analysis  

Epigenomic analysis of NKG2D-CAR CIML-NKAE and NKAE cells was performed by 

Cancer epigenetic group, at Josep Carreras Leukaemia Research Institute, Barcelona. 

Following manufacturer recommendations, NK cells Genomic DNA was bisulfite-

converted using an EZ DNA Methylation Gold kit (Zymo Research, Orange, CA, USA). 

DNA methylation was then interrogated using the Infinium MethylationEPIC Kit 

(Illumina). Raw signal intensity data were initially preprocessed from resulting idat files 

using minfi Bioconductor package (v1.36.0). Several quality control steps were applied 

to minimize errors and remove erratic probe signals, such as failed probes (probes with 

detection P value > 0.01 in at least 10% of samples were removed), cross-reacting probes 

and probes that overlapped SNPs within ±1 bp of CpG sites, followed by background 

correction and dye-based normalization using ssNoob algorithm (single-sample normal-

exponential out-of-band). Normalized intensities were used to calculate DNA 

methylation levels (Beta values) for each CpG evaluated. All analyses were performed 

under R statistical environment (v4.0.3). 

 

3.13. Mouse models 

Non-obese diabetic (NOD) Cg-Prkdcscid Il2rgtm1Wjl Tg(IL15)1Sz/SzJ (NSG-Tg(Hu-

IL15)) mice were procured from Jackson Laboratory (#030890) and used to study CAR 

effector in vivo efficacy. NSG-Tg(Hu-IL15) strain has been modified with a rhIL-15 

knock-in to constitutively express physiological concentrations of human IL-15 (7.1 ± 0.3 

pg/ml) [469]. All the procedures were approved by the Comunidad de Madrid (under the 

PROEX 191.2/20) and according to the European Union Directive 2010/63/EU and the 

Spanish Royal Decree-Law RD53/2013. Mice were maintained in pathogen-free 

conditions and housed in ventilated cages at 20-25ºC, 55 ± 10% relative humidity, ad 

libitum fed and exposed to 12 hours light:dark (LD) cycles. At the time of experiments, 

6-8 week-old female mice were randomly assigned to the different experiment groups 

(n=4 mice per group). 

NSG-Tg(Hu-IL15) mice were sublethally irradiated with a whole-body 1.5 Gy gamma 

ray dose, 24 hours before tumor infusion. At day 0, 5 x 105 U-266 or RPMI-8226 



MATERIALS AND METHODS 

108 
 

ffLucGFP cells were intravenously administered via tail vein. After 72 hours, CAR NK 

effectors were intravenously injected. 

Mice body weight was measured weekly and end point criteria (20% weight loss, asthenia 

or paraplegia) were evaluated daily. At this point, mice were sacrificed by anaesthesia 

(75 µg/gr ketamine and 500 µg/gr dexmedetomidine intraperitoneal (i.p)) and tissues 

were collected for further analysis. 

 

3.13.1. In vivo tumor biodistribution analysis by bioluminescence imaging (BLI) 

Tumor burden monitoring was weekly or fortnightly performed by BLI using In-Vivo 

Xtreme Preclinical Optical/X-ray Imaging System (Bruker Sciences) or IVIS Lumina 

XRMS Series III (Perkin Elmer) until the end of the experiment. Mice were anaesthetized 

with 2% isoflurane and 200 mg/kg of an aqueous solution of D-luciferin potassium salt 

(Thermosfisher) were intraperitoneally administered 10 minutes prior to image 

acquisition. Binning, f-stop and time exposure were automatically set for each picture and 

luminescence was expressed in photons per second per mm2 and overlaid on the white 

light image. 

 

3.13.2. CAR NK cell infiltration analysis in PB samples 

CAR effector in vivo follow-up was carried out at 7-, 14- and 29-days post-administration. 

PB from tail vein was extracted in EDTA tubes and CAR NK cells were detected by flow 

cytometry. CD45 PerCP/Cy5.5 and CD56 APC Abs were used for NK cell gating and 

rhBCMA/Streptavidin PE or NKG2D PE for CAR+ cells detection, as described in 3.3.1. 

 

3.13.3. Human population engraftment analysis in PB and BM at necropsy 

At end point, mouse necropsy was conducted and PB, femurs and tibias were extracted 

for further analysis. BM was flushed out into PBE with a syringe. Samples were filtered 

(0.3 µm) and erythrocytes were lysed with the ACK lysis buffer. MM cells were detected 

by GFP expression and CD138 PE/Cy7 staining, while CAR NK cells were analyzed by 

CD45 PerCP/Cy5.5, CD56 APC and rhBCMA/Streptavidin PE or NKG2D PE staining, 

after flow cytometer acquisition. 



MATERIALS AND METHODS 

109 
 

3.14. Statistical analysis 

GraphPad Prism 8 was used for all statistical analysis (GraphPad Software, CA, USA). 

Data are shown as mean ± Standard Error of Mean (SEM). First, Shapiro-Wilk and 

Levene tests were conducted to evaluate normality and homoscedasticity, respectively in 

the studied variables.  

Paired or unpaired two-tailed Student t-tests were applied between two parametric groups, 

one-way analysis of variance (ANOVA) following Tukey’s post-hoc test was carried out 

for two groups with only one variable and two-way ANOVA analysis was performed 

following Tukey´s multiple comparison post-hoc test for two groups that had two 

different variables. Mann-Whitney-Wilcoxon U test was used to compare two 

independent non-parametric variables, while Kruskal-Wallis test was performed for three 

or more groups comparison. In vivo mouse model survival curves were analyzed by 

applying the Kaplan-Meier method and Mantel-Cox test. Additionally, Kaplan-Meier 

analysis of PFS and OS between MM patients at diagnosis was carried out. Hazard ratio 

(HR) and corresponding 95% CIs are from a Cox proportional hazard model. One-sided 

p value was derived from a log-rank test. Statistical significance was shown as follows: p 

< 0.05 (*), p < 0.01 (**), p < 0.001 (***), p < 0.0001 (****). 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

110 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

111 
 

 

 

 

 

 

 

4. RESULTS 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

112 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



RESULTS 

113 
 

4.1. HLA-E/NKG2A immune checkpoint is upregulated in 

MM patients but α-NKG2A blocking antibodies are not 

efficient as monotherapy treatment 

4.1.1. HLA-E is overexpressed across the progression of MM 

To evaluate whether HLA-E is overexpressed in MM cells, we first studied HLA-E 

expression on malignant PC from MM patients in comparison to PC from HD. A 

compilation of 20 HD BM samples as well as 54 MM patient samples showed that HLA-

E is overexpressed on PC from MM patients (17.31 ± 1.99 RFI) with respect to PC from 

HD (10.31 ± 1.29 RFI) (Fig. 25A). 

We further assessed HLA-E expression on pPCs compared to healthy populations, such 

as nPCs and CD34+ progenitor cells, from those MM patients at different disease stages. 

We stratified MM patient samples into diagnosis (n=31) and progressions (n=25), and 

incorporated MGUS patients (n=12) into the study. The analysis of HLA-E expression 

levels revealed that this biomarker is overexpressed in pPC with respect to nPC in 59% 

of MM symptomatic patients and 78% of patients at progression. The expression ratio 

achieves significantly higher values in patients at progression (1.72 ± 0.2-fold), compared 

to patients at diagnosis (1.08 ± 0.12-fold) or MGUS (0.96 ± 0.19-fold) (Fig. 25B). By the 

same token, we also detected superior levels of HLA-E in pPC over CD34+ cells in 70% 

of MM symptomatic patients and 84% of patients at progression, with the highest 

overexpression values in progression cases (2.2 ± 0.24-fold) (Fig. 25B). 



RESULTS 

114 
 

 

Figure 25. HLA-E is overexpressed on PC from MM patients, especially at progression, but it does 

not significantly correlate with a higher OS or PFS. A) Flow cytometry analysis of HLA-E expression 

on PC from HD (n=20) or symptomatic MM patient BM samples (n=54). The mean ± SEM is shown. B) 

At the left, flow cytometry analysis of HLA-E expression ratio between pPC and nPC from MGUS (n=11) 

or symptomatic MM patient samples at diagnosis (n=23) or progression (n=18). At the right, HLA-E 

expression ratios between pPC and CD34+ cells from MGUS (n=12) or symptomatic MM patient samples 

at diagnosis (n=31) or progression (n=25). The mean ± SEM is shown. Below, percentages of patients with 

pPC/nPC or pPC/CD34+ HLA-E expression ratios >, < or equal to 1 are represented as circle chart. C) 

Kaplan-Meier analysis of progression-free survival (PFS) and overall survival (OS) between MM patients 

at diagnosis (n=32). Hazard ratio (HR) and corresponding 95% CIs are from a Cox proportional hazard 

model stratified by HLA-E ratio pPC/CD34+ ≤ 1.1 or > 1.1. *p<0.05; **p<0.01. 

 

High HLA-E expression has been associated with poor patient survival in solid tumors 

[379, 384-386, 388, 389], but its relevance as biomarker in MM progression has not been 



RESULTS 

115 
 

properly studied. Our latter analysis, which shows HLA-E overexpression on MM cells 

with regard to healthy populations across MM progression, raised the evaluation of HLA-

E as a prognostic factor in MM. For this purpose, we stratified MM patients at diagnosis 

in two groups: high HLA-E expression (patients with a pPC/CD34+ HLA-E expression 

ratio >1.1; n=14) and low HLA-E expression (patients with a pPC/CD34+ HLA-E 

expression ratio ≤1.1; n=18). No significant differences were found neither in OS nor in 

PFS. However, high HLA-E expression patients had a median PFS of 25 months, while 

the median PFS within the low HLA-E expression group was not reached (HR 0.64; 95CI 

0.22-1.86) (Fig. 25C). 

Altogether, these results suggest that HLA-E expression increases with the progression 

of MM but a further study, in a larger number of patients, would be required to better 

evaluate the use of this biomarker to predict MM patient prognosis. 

 

4.1.2. IFN-γ is highly expressed in BM MM patients and increases HLA-E levels, 

while BTZ decreases its expression on MM cells 

To understand the possible effect of TME on HLA-E expression on MM cells, we studied 

the presence of different immunosuppressive or inflammatory cytokines (TGF-β1, IL-1β, 

IL-6, IL-3, TNF-α, MCP-1, IL-2, IL-8, IFN-α, IL-15, IL-18, IL-10 and IFN-γ) in MM 

patient BM plasmas, including patients at diagnosis and progression, compared to HD. 

Of note, neither HD nor patients had any sign of infection at the time BM samples were 

extracted. Results show that the concentration of TGF-β1 (66.76 ± 13.81 pg/ml vs 16.97 

± 2.95 pg/ml), MCP-1 (2,807 ± 398.6 pg/ml vs 1,316 ± 120.9 pg/ml), IL-6 (23.78 ± 3.49 

pg/ml vs 6.91 ± 1.69 pg/ml) and IFN-γ (14.35 ± 2.9 pg/ml vs 6.85 ± 0.27 pg/ml) were 

significantly higher in the plasma of BM from MM patient compared to HD (Fig. 26A).  



RESULTS 

116 
 

 

Figure 26. BM plasma of MM patients has an increased concentration of IFN-γ, cytokine that, along 

with BTZ, modulates HLA-E expression. A) Cytokine quantification in BM plasma from HD (n=14) and 

MM patients (n=32) by LEGENDplex bead-based immunoassay. The mean ± SEM is shown. B) Flow 

cytometry analysis of HLA-E expression modulation on MM cell lines RPMI-8226 and NCI-H929 and the 

20 nM BTZ-resistant RPMI-8226 R20 and NCI-H929 R20 cells (n=3) or C) on MM patient PC at diagnosis 

(n=3) or progression (n=2), without treatment or after 24-hour treatment with 20 nM BTZ, 500 ng/ml IFN-

γ or both. The mean ± SEM is shown. ns: not significant; *p<0.05; **p<0.01; ***p<0.001. 

 

Given that IFN-γ has been described to increase HLA-E expression on MM cells [393], 

we studied this association in depth. 

With this aim, NCI-H929 and RPMI-8226 MM cells were treated with IFN-γ for 24 hours. 

Flow cytometry analysis demonstrated that IFN-γ is able to significantly increase HLA-



RESULTS 

117 
 

E expression, 2.87 ± 0.42-fold on RPMI-8226 and 2.48 ± 0.24-fold NCI-H929 MM cells, 

over untreated MM cells (Fig. 26B). In addition, we tested the induction of HLA-E in 

MM primary cells after 24-hour IFN-γ exposure over BMMC cultures. Similarly, IFN-γ 

treatment is capable of enhancing HLA-E expression on PC of MM patients at diagnosis 

(from 6.89 ± 4.39 RFI without treatment to 14.59 ± 5.76 RFI) and progression (from 

10.13 ± 1.84 RFI without treatment to 17.28 ± 3.6 RFI) (Fig. 26C).  

Since previous studies demonstrated that BTZ treatment reduces HLA-E expression on 

MM cells [85], we asked whether the activity of this proteasome inhibitor might be 

compromised in the presence of IFN-γ. BTZ treatment slightly decreases HLA-E 

expression on the studied MM cell lines, but it is able to significantly counteract the 

induction of HLA-E expression caused by IFN-γ on RPMI-8226 cells (2.87 ± 0.42-fold 

after IFN-γ vs 0.66 ± 0.07-fold after IFN-γ and BTZ) and NCI-H929 cells (2.48 ± 0.24-

fold after IFN-γ vs 0.98 ± 0.07-fold after IFN-γ and BTZ) (Fig. 26B). HLA-E reduction 

due to BTZ treatment is also observed in PC, even after IFN-γ exposure from patients at 

diagnosis (14.59 ± 5.76 RFI vs 7.77 ± 3.22 RFI) and progressions (17.28 ± 3.6 RFI vs 

9.18 ± 1.54 RFI) (Fig. 26C). 

MM patients can acquire BTZ resistance after first-line standard treatments [98]. 

Therefore, we studied the combination effect of IFN-γ and BTZ over MM cells resistant 

to 20 nM of BTZ: RPMI-8226 R20 and NCI-H929 R20. HLA-E expression induction on 

RPMI-8226 R20 treated with IFN-γ is 2.67 ± 0.42-fold, but BTZ treatment cannot 

significantly decrease this value (2.49 ± 0.62-fold). Similarly, the increment of HLA-E 

expression caused by IFN-γ on NCI-H929 R20 (2.38 ± 0.2-fold), is maintained after BTZ 

exposure (1.87 ± 0.2-fold) (Fig. 26B). 

These results indicate that IFN-γ treatment augments HLA-E presence on the surface of 

both cell lines and primary MM cells, while BTZ decreases HLA-E expression, even the 

higher levels achieved after IFN-γ exposure. However, BTZ treatment cannot reduce IFN-

γ-induced HLA-E expression on BTZ-resistant MM cell lines, which could mimic the 

BTZ-resistant MM patient scenario. 

 

 



RESULTS 

118 
 

4.1.3. NKG2A is overexpressed on MM cytotoxic NK cells at progression but α-

NKG2A treatment does not reactivate their cytotoxicity against tumor cells 

in native cultures 

In addition to the HLA-E study on PC from MM patients, we analyzed the expression of 

their activating and inhibiting receptors, NKG2C and NKG2A, respectively, on cytotoxic 

NK (CD56dimCD16+) and T (CD3+CD8+) cells from these patients.   

Flow cytometry results revealed that there is a higher percentage of cytotoxic NK cells 

expressing NKG2A than NKG2C in all stages (Fig. 27A). No differences were found 

regarding NKG2C expression on NK cells (17.17 ± 4.1% on MGUS, 12.2 ± 2.52% on 

MM at diagnosis and 28.71 ± 6.11% at progression). Remarkably, the percentage of 

NKG2A+ cells is significantly higher in patients at progression (49.24 ± 7.61%), 

compared to patients at diagnosis (29.56 ± 4.82%), which suggests that the NK population 

from these patients may be inhibited by the HLA-E/NKG2A immune checkpoint. 

Regarding T cells, NKG2C and NKG2A expression is minimal in all the studied patients, 

less than 20% of positive cells.  



RESULTS 

119 
 

 

Figure 27. Cytotoxic NK cells from MM patients at progression have increased NKG2A expression, 

but α-NKG2A Abs in monotherapy cannot restore NK cell activity against primary PC. Flow 

cytometry analysis A-B) of NKG2C, NKG2A (A) and PD-1 (B) expression on CD8+ T and CD56dim CD16+ 

NK cells from BM MGUS (n=13) and MM patient samples at diagnosis (n=19) or at progression (n=16). 

C) Flow cytometry analysis of PC survival in cultures of BMMCs from MM patients after treatment with 

the mouse (Z199) or human (BMS-986315) α-NKG2A Abs or their corresponding isotypes (mouse IgG2b 

and human IgG1κ, respectively) (n=5). Means ± SEM are shown. ns: not significant; *p<0.05; **p<0.01. 

 

In addition, we have studied the expression of another relevant immune checkpoint in 

MM, PD-1. The percentage of PD-1+ T cells is 43.78 ± 3.91% at MGUS, 47.61 ± 3.26% 

at diagnosis, increasing significantly in patients at progression 60.3 ± 3.36% (Fig. 27B). 

By contrast, low percentages of MM cytotoxic NK cells express this inhibitory receptor 



RESULTS 

120 
 

(4.29 ± 2.35% on MGUS, 4.06 ± 1.17% on MM at diagnosis and 11.26 ± 2.64% at 

progression). Altogether, these results suggest that NKG2A is not the predominantly 

inhibitory receptor expressed in MM CD8+ T cells, where PD-1 might be a relevant 

inhibitory signal. 

Since both HLA-E on pPC and NKG2A on cytotoxic NK cells are highly expressed on 

MM patients at progression, we assessed the disruption of the inhibitory signal in order 

to reactivate NK cell cytotoxicity against the tumor cells. BMMCs from MM patients 

were cultured for 24 hours with a mouse (Z199) or human (BMS-986315) α-human 

NKG2A Ab or their corresponding isotypes, mouse IgG2b (mIgG2b) or human IgG1κ 

(hIgG1κ), respectively. Neither the Z199 nor the BMS-986315 Abs are able to restore 

patient NK cell activity against primary MM cells, not exceeding 8% of lysis reduction 

(Fig. 27C). 

Taken together, these results demonstrate that, despite the HLA-E/NKG2A immune 

checkpoint axis is upregulated in MM patients at progression, the treatment with α-

NKG2A Abs, as single agents, is not effective reducing pPC number in vitro. 

 

4.2. NKG2A blockade improves NKG2D-CAR and BCMA-

CAR NKAE cell anti-MM cytotoxic activity 

4.2.1 NKG2D- and BCMA-CAR NKAE cells are efficiently generated 

Based on HLA-E overexpression on MM cells and considering that α-NKG2A 

monotherapy has not restored NK cell activity, correlating with clinical trial results in 

solid malignancies [407, 408], we studied the combination of α-NKG2A Abs with a novel 

immunotherapy approach. Regarding that clinical responses obtained with CAR NK cells 

might be improved, we determined to evaluate the combination of α-NKG2A blocking 

Abs with two CAR NK cell therapies for MM treatment. 

To generate CAR NKAE cells from HD, PBMCs were obtained from 18 ml of PB and 

co-cultured with the irradiated feeder cell line, K562-mb21-41BBL. Once NKAE cells 

were purified at day 6 of culture, these cells were transduced with NKG2D- or BCMA-

CAR lentivectors at MOI 5 and 16, respectively. 



RESULTS 

121 
 

Phenotype analysis of these populations, NKG2D-CAR NKAE and BCMA-CAR NKAE 

cells, was performed by flow cytometry (Fig. 28A). Although NK cell percentage is lower 

than 10% in PBMCs at day 0 of culture (7.98 ± 2.96% among cultures for NKG2D-CAR 

NKAE cells and 9.91 ± 2.4% for BCMA-CAR NKAE cells), after activation, expansion, 

purification, and transduction, an average purity of 91.96 ± 2.05% CD56+CD3- cells in 

NKG2D-CAR and 94.1 ± 3.11% CD56+CD3- cells in BCMA-CAR NKAE cell cultures 

was obtained at day 14. Of note, NK cells at both time points are mainly cytotoxic (around 

70-90% CD56+CD16+). Moreover, despite T cells are a predominant subset at the 

beginning of NKAE cell expansion, with a mean percentage above 40% (38.64 ± 10.22% 

among cultures for NKG2D-CAR NKAE cells and 50.14 ± 9.14% for BCMA-CAR 

NKAE cells), NK cell expansion and subsequent purification reduce CD3+ cells in both 

cultures to values lower than 0.5% (0.19 ± 0.1% in NKG2D-CAR NKAE cell cultures 

and 0.43 ± 0.38 in BCMA-CAR NKAE cell cultures at day 14) (Fig. 28B). 



RESULTS 

122 
 

Figure 28. High-purity and cytotoxic NKG2D- and BCMA-CAR NKAE cell generation from HD PB. 

A) Representative flow cytometry dot plot of NK cells (CD56+CD3-), cytotoxic NK cells (CD56+CD3-

CD16+), and T cells (CD3+) from NKG2D- and BCMA-CAR NKAE cell cultures at day 0 and day 14. B) 

Flow cytometry analysis of the percentage of NK cells (CD56+CD3-), cytotoxic NK cells (CD56+CD3-

CD16+), and T cells (CD3+) cells in NKG2D- or BCMA-CAR NKAE cell cultures at day 0 and day 14 

(n=6). Means ± SEM are shown. *p<0.05; **p<0.01; ****p<0.0001. 

 

At day 6 after transduction, flow cytometry and q-PCR analysis were performed to 

evaluate CAR transduction efficiency in NKAE cells. The percentage of transduction, 

analyzed by flow cytometry, is 34.05 ± 6.32% for NKG2D-CAR and 22.21 ± 4.83% for 

BCMA-CAR (Fig. 29A, B). q-PCR data show there to be 0.82 ± 0.22 and 0.55 ± 0.17 

vector copies per cell on average of NKG2D- and BCMA-CAR, respectively (Fig. 29B). 



RESULTS 

123 
 

The protein expression of both CAR in NKAE cells was also corroborated by western 

blotting using an α-CD3ζ Ab. To obtain the theoretical molecular weight of CAR 

molecules, nucleotide sequences were translated to protein sequence with ExPASy 

translate tool and these amino acid sequences were applied to calculate the protein 

molecular weight with ExPASy compute pI/Mw. Western blot analysis shows proteins at 

the expected molecular mass of NKG2D-CAR (43.5 kDa) and BCMA-CAR (53.9 kDa) 

(Fig. 29C). 

These results demonstrate the feasibility of generating NKG2D- and BCMA-CAR NKAE 

cells, achieving a high percentage of pure and cytotoxic NK cells, with a low percentage 

of T cell contamination, and being potently transduced with CAR molecules. 



RESULTS 

124 
 

 

Figure 29. NKAE cells are stable and efficiently transduced with lentivectors containing NKG2D- or 

BCMA-CAR. A-B) Representative flow cytometry dot plot (A) and analysis (B, left) of NKG2D- and 

BCMA-CAR expression on the surface of NKAE cells at 6 days post-transduction (n=7). Vector copy 

number (VCN) quantification (B, right) of NKG2D- and BCMA-CAR in NKAE cells by q-PCR (n=7). The 

mean ± SEM is shown. C) Western blot analysis of NKG2D- and BCMA-CAR expression in NKAE cells 

after 6 days post-transduction. Lysates of untransduced, NKG2D-CAR and BCMA-CAR NKAE cells were 

separated by SDS-PAGE. A mouse anti-human CD3ζ Ab was used to detect the expression of endogenous 

and chimeric CD3ζ proteins. p150 was used as loading control. 

 

4.2.2. NKG2A is highly expressed on CAR NKAE cells 

To study the modulation of NKG2 family receptors along NK cell expansion, we analyzed 

NKG2C and NKG2A expression on NKG2D- and BCMA-CAR NKAE cells. Regarding 



RESULTS 

125 
 

NKG2D, we only studied the differential expression of this receptor between day 0 and 

day 14 on BCMA-CAR NKAE cells, to avoid NKG2D-CAR expression bias. 

Flow cytometry results show that 12.91 ± 5.2% of NKG2D-CAR NKAE cells were 

NKG2C+ at day 0 vs 17.11 ± 4.23% at day 14, whereas 31.3 ± 13.37% of BCMA-CAR 

NKAE cells were NKG2C+ at day 0 vs 13.35 ± 3.69% at day 14 (Fig. 30A). While 

NKG2C levels remained stable after NK cell expansion, NKG2A expression significantly 

enhanced on NKG2D-CAR NKAE cells (59.27 ± 4.24% at day 0 vs 90.49 ± 2.76% at day 

14) and BCMA-CAR NKAE cells (48.46 ± 6.05% at day 0 vs 85.23 ± 3.66% at day 14). 

The expression of NKG2D also increased from 11.1 ± 1.74% at day 0 to 64.79 ± 11.52% 

at day 14 in BCMA-CAR NKAE cells (Fig. 30A). 

Since the majority of CAR NKAE cells were NKG2A+, we studied the effect of several 

cytokines, present in MM TME (IL-6, IL-10, TGF-β, and IFN-γ) or normally used to 

stimulate NK cells (IL-2, IL-15, IL-21, and 4-1BBL), to determine which of them can 

induce NKG2A expression on NK cells. NKG2A expression on regulatory 

(CD56brightCD16-) and cytotoxic (CD56dimCD16+) NK cells, purified from HD PBMCs, 

was measured by flow cytometry. IL-15 significantly increases NKG2A expression on 

both regulatory (162.21 ± 25.37 RFI with IL-15 vs 102.56 ± 7.14 RFI without stimulation) 

and cytotoxic (101.7 ± 11.05 RFI with IL-15 vs 54.94 ± 4.59 RFI without stimulation) 

NK cell subpopulations (Fig. 30B), as well as its combination with cytokines utilized in 

ex vivo culture and expansion of NK cells. Moreover, IL-2 in combination with IL-21, 

with or without 4-1BBL, also augments NKG2A expression. Notably, this last 

combination, IL-2, IL-21, and 4-1BBL, the one employed by our group to generate NKAE 

cells (IL-21 and 4-1BBL are expressed on the membrane of the feeder cell), significantly 

increases NKG2A expression on regulatory (149.55 ± 10.75 RFI with IL-2, IL-21, and 4-

1BBL vs 102.56 ± 7.14 RFI without stimulation) and cytotoxic (101.9 ± 12 RFI with IL-

2, IL-21, and 4-1BBL vs 54.94 ± 4.59 RFI without stimulation) NK cells. 

These results suggest that all the cytokines commonly used in NK cell ex vivo expansion 

protocols increase the expression of the inhibitory receptor NKG2A, probably to 

counteract the high activation and expansion of these cells, defined by the increase of 

NKG2D expression. 



RESULTS 

126 
 

 

Figure 30. Cytokines used to expand NK or CAR NK cells ex vivo promote NKG2A expression. A) 

Flow cytometry analysis of the percentage of NKG2C+, NKG2A+, and NKG2D+ cells within CD56+CD3- 

populations on NKG2D- or BCMA-CAR NKAE cell cultures at day 0 and day 14 (n=6). Means ± SEM are 

shown. B) Flow cytometry analysis of NKG2A expression on regulatory (CD56brightCD16-) and cytotoxic 

(CD56dimCD16+) NK cells after 48-hour NK cell culture with different cytokines: IL-2 (50 IU/ml), IL-6 (20 

ng/ml), IL-10 (50 ng/ml), IL-15 (50 ng/ml), IL-21 (50 ng/ml), TGF-β (10 ng/ml), 4-1BBL (200 ng/ml), and 

IFN-γ (500 ng/ml) or combinations of these cytokines (n=9). Means ± SEM are shown. ns: not significant; 

*p<0.05; **p<0.01; ***p<0.001; ****p<0.0001 

 

 



RESULTS 

127 
 

4.2.3. CAR NKAE cells produce a high concentration of IFN-γ and increase   

HLA-E tumor expression after co-culture with MM cells 

Once we have studied the NKG2A modulation on NK cells by cytokines present in the 

TME or used in NK cell activation protocols, we investigated whether CAR NKAE cells 

can modulate HLA-E expression on MM cell lines. 

We first studied HLA-E expression on the MM cell lines used in the subsequent 

experiments: 2.64 ± 0.23 RFI on U-266, 3.64 ± 0.54 RFI on RPMI-8226, 4.48 ± 0.28 RFI 

on ARP-1 and 4.48 ± 0.18 RFI on XG-1. Moreover, we included the feeder cell line, 

K562-mb21-41BBL, as a HLA-E- cell line (RFI of 0.23 ± 0.09) (Fig. 31A). 

To study the impact of CAR NKAE treatments on HLA-E expression on MM cell lines, 

we co-cultured both CAR therapies with the XG-1 MM cell line and analyzed HLA-E 

presence on its surface at 0, 3, and 24 hours. At 3 hours, HLA-E expression is similar to 

basal levels but, at 24 hours, HLA-E RFI significantly increases up to 10.17 ± 2.44 on 

XG-1 cells after NKG2D-CAR NKAE cell co-culture (vs 2.33 ± 0.1 RFI on XG-1 without 

NK treatment) and 11.61 ± 5.81 after BCMA-CAR NKAE cell co-culture (vs 2.29 ± 0.25 

RFI on XG-1 without NK treatment) (Fig. 31B). 

Bearing in mind that CAR NK cells secrete IFN-γ [284], and this cytokine augments 

HLA-E expression on MM cells [378], we quantified the amount of IFN-γ produced by 

CAR NKAE cells for 24 hours. Flow cytometry analysis revealed that CAR NKAE cells, 

in the absence of tumor cells (590.87 ± 136.75 pg/ml in NKG2D-CAR NKAE cells and 

97.41 ± 9.31 pg/ml in BCMA-CAR NKAE cells) or in co-culture with XG-1 cells (693.14 

± 200.69 pg/ml in NKG2D-CAR NKAE cells and 236.05 ± 53.24 pg/ml in BCMA-CAR 

NKAE cells), release a high concentration of IFN-γ, compared to XG-1 cultures (4.24 ± 

0.14 pg/ml) (Fig. 31C). 

Taken together, these findings suggest that HLA-E expression may further enhance on 

MM cells after CAR NK cell treatment, increasing the resistance to CAR NK cell therapy 

mediated by HLA-E/NKG2A immune checkpoint.  

 



RESULTS 

128 
 

 

Figure 31. HLA-E expression increases on MM cells after co-culture with CAR NKAE cells that 

release high concentrations of IFN-γ. A) Flow cytometry analysis of HLA-E expression on U-266, RPMI-

8226, ARP-1 and XG-1 MM cells and the feeder cell K562-mb21-41BBL (n=5). Means ± SEM are shown. 

B) Flow cytometry analysis of HLA-E expression on XG-1 cells alone or in co-culture with NKG2D- (n=4) 

or BCMA-CAR NKAE cells (n=3) at 0, 3 and 24 hours. The mean ± SEM is shown. C) IFN-γ quantification 

in NKG2D- or BCMA-CAR NKAE cell supernatants alone or after 24 hours co-culture with XG-1 MM 

cells, at 1:2 T:E ratio, by LEGENDplex bead-based immunoassay (n=3). *p<0.05; **p<0.01; ***p<0.001. 

 

4.2.4. NKG2A blockade enhances CAR NKAE cell cytotoxicity against MM cell 

lines 

Given that a high percentage of NKG2D- and BCMA-CAR NKAE cells are NKG2A+, 

and that HLA-E is overexpressed on MM cell lines and primary MM cells, whose 

expression is even higher after CAR therapies co-cultures, we used the two α-NKG2A 

Abs (mouse Z199 and human BMS-986315) to disrupt this inhibitory immune 

checkpoint. 

First, we evaluated the expression of CAR targets, NKG2D-L and BCMA, on different 

MM cell lines. U-266 (17.77 ± 0.61 RFI) and ARP-1 (6.29 ± 0.48 RFI) cells have a high 

NKG2D-L expression, whereas XG-1 (2.52 ± 0.15 RFI) and RPMI-8226 (4.09 ± 0.53 

RFI) cell NKG2D-L expression is low. U-266 (15.14 ± 1.1 RFI) and RPMI-8226 (14.49 

± 3.93 RFI) cells have high expression of BCMA, while XG-1 (7.17 ± 0.84 RFI) and 

ARP-1 (7.17 ± 0.84 RFI) cells present lower levels (Fig. 32A).  



RESULTS 

129 
 

To test the efficacy of NKG2A blockade to improve CAR NKAE cell killing capacity 

against MM cell lines, XG-1, ARP-1, U-266, and RPMI-8226 cells were co-cultured with 

CAR effectors pretreated with the mouse (Z199) or human (BMS-986315) α-NKG2A 

Abs, or their corresponding isotypes. Calcein-release assays demonstrated that blocking 

with either the mouse or the human α-NKG2A Abs significantly increases NKG2D- (Fig. 

32B) and BCMA-CAR NKAE cell cytotoxicity (Fig. 32C) against all MM cell lines at 

different T:E ratios. 

In subsequent studies, we tested the efficacy of both α-NKG2A Abs after increasing HLA-

E expression on RPMI-8226 cells with a 24-hour IFN-γ treatment. Calcein-release assay 

results indicate a lower CAR NKAE cytotoxicity against RPMI-8226 cells when 

pretreated with IFN-γ. However, even in this increased resistance setting, both α-NKG2A 

blocking Abs were also able to potentiate NKG2D- and BCMA-CAR NKAE anti-MM 

activity (Fig. 32D). 

Altogether, these in vitro experiments demonstrate that the blockade of the inhibitory 

receptor NKG2A fosters CAR NKAE cytotoxic activity against HLA-E+ MM cell lines. 



RESULTS 

130 
 

 

Figure 32. NKG2D- and BCMA-CAR NKAE cells in combination with α-NKG2A blocking Abs 

exhibit higher cytotoxicity against MM cell lines, even after 24h-treatment of tumor cells with IFN-

γ. A) Flow cytometry analysis of NKG2D-L (ULBP1, ULBP 2/5/6, ULBP 3 and MICA/B) and BCMA 

expression on U-266, RPMI-8226, XG-1, and ARP-1 MM cells (n=6). Means ± SEM are shown. B-C) 

Specific lysis of NKG2D- (B) and BCMA-CAR NKAE cells (C) treated with either the mouse (Z199) or 

the human (BMS-986315) α-NKG2A Abs or their corresponding isotypes (mouse IgG2b and human IgG1κ, 

respectively) against the MM cells U-266, XG-1, ARP-1, and RPMI-8226, at different T:E ratios, analyzed 

by 3h-Calcein release assay (n=3-6). The mean ± SEM is shown. D) Specific lysis of NKG2D- and BCMA-

CAR NKAE cells treated with either the mouse (Z199) or the human (BMS-986315) α-NKG2A Abs or 

their corresponding isotypes (mouse IgG2b and human IgG1κ, respectively) against the MM cells RPMI-

8226, at 1:1 ratio, analyzed by 3h-Calcein release assay (n=3). Where indicated, RPMI-8226 cells are 

pretreated with IFN-γ for 24 hours. The mean ± SEM is shown. *p<0.05; **p<0.01; ***p<0.001; 

****p<0.0001. 

 



RESULTS 

131 
 

4.2.5. NKG2A blockade increases CAR NKAE degranulation levels and IFN-γ 

release during co-culture with MM cell lines 

Another hallmark of NK cell activation by tumor cells is the degranulation capacity. NK 

cells hold cytotoxic granules containing perforins and granzymes, among other cytolytic 

proteins, which are directly released when these cells interact with target cells. Since the 

protein CD107a, also known as LAMP-1, is expressed on the membrane of these 

cytotoxic granules, the detection of this molecule has been described as a marker of NK 

and CD8+ T cells degranulation, an indirect method to measure effector cytotoxicity [470] 

(Fig. 33A). 

Both NKG2D- and BCMA-CAR NKAE cells pretreated with the α-NKG2A Abs heighten 

their degranulation levels against all the MM cell lines tested, but not against the HLA-

E- feeder cell K562-mb21-41BBL. Against ARP-1 cell line, 20.4 ± 4.47% or 27.04 

± 4.38% of NKG2D-CAR NKAE cells pretreated with mIgG2b or hIgG1κ, 

respectively, are CD107a+, but increase to 35.98 ± 3.98% or 44.27 ± 5.21% in cells 

pretreated with Z199 or BMS-986315 α-NKG2A Abs, respectively. The pretreatment of 

BCMA-CAR NKAE cells with α-NKG2A Abs also potentiates degranulation levels 

(22.91 ± 1.17 % or 24.73 ± 4.03 % NKAE cells pretreated with mIgG2b or hIgG1κ 

vs 39.26 ± 2.95 % or 42.5 ± 3.77 % NKAE cells pretreated with Z199 or BMS-986315 

α-NKG2A Abs) (Fig. 33B). 

 



RESULTS 

132 
 

 



RESULTS 

133 
 

Figure 33. Higher degranulation capacity and IFN-γ release are exhibited by α-NKG2A-pretreated 

NKG2D- and BCMA-CAR NKAE cells. A) Representative flow cytometry dot plot of CD107a staining 

in NKG2D- and BCMA-CAR NKAE cells pretreated with α-NKG2A blocking Abs (Z199 or BMS-986315) 

or their corresponding isotypes (mIgG2b or hIgG1κ, respectively) and co-cultured with the ARP-1 MM 

cells for 4h. B) Flow cytometry analysis of CD107a expression on NKG2D- (n=5) and BCMA-CAR NKAE 

cells (n=3) pretreated with α-NKG2A blocking Abs (Z199 or BMS-986315) or their corresponding isotypes 

(mIgG2b or hIgG1κ, respectively), alone or after 4-hour co-culture with the XG-1, U-266, ARP-1, and 

RPMI-8226 MM cell lines, and the feeder cell HLA-E-, K562-mb21-41BBL, at 1:1 ratio. The mean ± SEM 

is shown. C) Quantification of IFN-γ release after NKG2D- or BCMA-CAR NKAE cells pretreated with 

either the mouse (Z199) or human (BMS-986315) α-NKG2A Abs or their corresponding isotypes (mouse 

IgG2b or human IgG1κ, respectively) upon 24-hour co-culture with the MM cells U-266, XG-1, and RPMI-

8226, at 1:2 T:E ratio, by LEGENDplex bead-based immunoassay (n=8). The mean ± SEM is shown. ns: 

not significant; *p<0.05; **p<0.01. 

 

As the NKG2A blockade enhances CAR NKAE cell degranulation, we studied which 

cytokines or cytolytic proteins are increased. NKG2A-blocked NKG2D- and BCMA-

CAR NKAE cells were cultured alone or with the MM cells U-266, XG-1 or RPMI-8226, 

or the feeder cell line K562-mb21-41BBL for 24 hours. Then, supernatants were collected 

and IL-2, IL-4, IL-6, IL-10, IL-17A, IFN-γ, TNF-α, soluble Fas, soluble FasL, granzyme 

A, granzyme B, perforin and granulysin release were analyzed by flow cytometry. Of 

note, the concentration of IFN-γ measured into NKG2D-CAR NKAE cell co-cultures 

against RPMI-8226 increases up to 236.23 ± 58.89 pg/ml and 282.8 ± 67.22 pg/ml in 

cultures of CAR NK cells pretreated with mouse or human α-NKG2A Abs, compared to 

88.52 ± 30.57 pg/ml and 95.78 ± 32.78 pg/ml in cells pretreated with the mIgG2b or 

hIgG1κ, respectively. In addition, the pretreatment of BCMA-CAR NKAE cells with 

mouse or human α-NKG2A Abs raises IFN-γ concentration to 682.01 ± 103.99 pg/ml and 

760.87 ± 154.15 pg/ml, respectively, whereas the concentration with their corresponding 

isotypes is 359.43 ± 70.85 pg/ml and 493 ± 112.39 pg/ml (Fig. 33C). 

Collectively, these results demonstrate that breaking off the HLA-E/NKG2A interaction 

by means of NKG2A blocking Abs potentiates NKG2D-CAR and BCMA-CAR NKAE 

cell cytolytic capacity against MM cell lines. 

 



RESULTS 

134 
 

4.2.6. α-NKG2A treatment does not significantly affect CAR NKAE cell 

activation or inhibition marker expression, but increases ZAP70/Syk 

phosphorylation 

To investigate whether NKG2A blockade has an impact on NK cell phenotype, the 

expression of different activation and inhibition surface markers was analyzed by mass 

(Fig. 34A) and flow cytometry (Fig. 34B) on NKG2D- and BCMA-CAR NKAE cells 

pretreated with the human α-NKG2A Ab or its corresponding isotype after co-culture with 

RPMI-8226 cells for 24 hours. 

 

 

Figure 34. NKG2D-CAR NKAE cell immunophenotype is not markedly altered by NKG2A blockade 

but ZAP70/Syk phosphorylation is potentiated after tumor stimulation. A) Mass cytometry analysis of 

several markers in NKG2D-CAR NKAE cells (CD56+) pretreated with the BMS-986315 α-NKG2A Ab or 

the human IgG1κ isotype, after 24-hour co-culture with the MM cells RPMI-8226 (n=3). B) Flow cytometry 



RESULTS 

135 
 

analysis of different markers expressed on the surface of NKG2D-CAR NKAE cells (CD56+CD3-) after 

pretreatment with the BMS-986315 α-NKG2A Ab or the human IgG1κ isotype, and 24-hour co-culture 

with the MM cells RPMI-8226 (n=3). C) Western blot analysis of ZAP-70 (Tyr319) / Syk (Tyr352) 

phosphorylation in NKG2D- and BCMA-CAR NKAE cells, pretreated with BMS-986315 α-NKG2A Ab 

or the human IgG1κ isotype after 1-hour co-culture with the RPMI-8226 MM cells. Where indicated, 

RPMI-8226 cells, pretreated with BMS-986315 Ab or IgG1κ isotype, as well as NKG2D- and BCMA-CAR 

NKAE cells, are collected at time 0. Cell lysates were separated by SDS-PAGE. A rabbit anti-human 

phospho-ZAP-70 (Tyr319) / Syk (Tyr352) Ab was used to detect the phosphorylation of ZAP-70/Syk. ZAP70 

was used as loading control. *p<0.05. 

 

The detection of NKG2A by both cytometry techniques is diminished, indicating that α-

NKG2A blocking Ab binds to NK cells and compete with α-NKG2A staining and/or 

foster the reduction of the receptor in the cell surface by internalization. Apart from that, 

NKG2A blockade did not significantly affect the expression of other NK cell surface 

receptors studied, including CD69, CD25, NKp30, NKp44, NKp46, NKG2C, and PD-1. 

Consequently, intracellular signaling was examined by assessing the phosphorylation of 

Syk and zeta-chain-associated protein kinase 70 (ZAP-70) by western blot. The co-culture 

with RPMI-8226 cells increments the phosphorylation of Syk (Tyr352) and ZAP-70 

(Tyr319) in human isotype-pretreated NKG2D- and BCMA-CAR NKAE cell co-cultures 

but increases even more with the blockade of NKG2A (Fig. 34C). 

Overall, these results suggest that blocking NKG2A does not modify the expression of 

the main activating and inhibiting receptors of CAR NK cells, but intracellular signaling 

pathways seem to be differentially activated. 

 

4.2.7. Pretreatment with α-NKG2A Abs enhances the cytotoxicity of NKG2D- 

and BCMA-CAR NKAE therapies over MM primary cells without 

compromising CD34+ population 

After studying the ability of the blockade of NKG2A to strengthen the antitumor potency 

of both CAR NKAE treatments against MM cell lines, we sought to investigate the killing 

activity over primary PC and CD34+ progenitors from MM patients. BM samples of 7 

newly diagnosed patients and 2 differently treated patients at progression were used for 

these assays (for detailed patient characteristics see Table 13). 



RESULTS 

136 
 

 Age Status ISS Treatment Isotype PC (%) 
NKG2D-L 

(RFI) 

BCMA 

(RFI) 

HLA-E 

(RFI) 

CD34+ 

(%) 

#1 73 ND - - IgG kappa 6.11 21.45 32.12 4.72 3.41 

#2 62 ND R-ISS II - IgG kappa 36.7 21.68 41.83 4.63 0.25 

#3 89 ND ISS-III - IgG kappa 36.03 5.65 27.76 6.39 1.9 

#4 79 ND ISS-III - IgG kappa 13.55 4.03 13.58 11.97 1.46 

#5 53 ND R-ISS II - IgG kappa 8.38 3.32 13.62 11.37 1.95 

#6 71 ND ISS-II - Light chain kappa 8.55 16.54 20.1 4.21 0.69 

#7 58 ND R-ISS III - IgG lambda 46.56 7.04 16.15 5.67 2.62 

#8 59 RR ISS-II DRd (1st line) IgG lambda 18.6 15.94 27.6 8.99 0.67 

#9 57 RR R-ISS II DKd (3rd line) IgG kappa 39.3 12.12 30.33 5.25 0.13 

 

Table 13. Characteristics of MM patients used in ex vivo experiments. ISS: International Staging 

System; PC: plasma cell; RFI: relative fluorescence intensity; NKG2D-L: NKG2D ligands; ND: newly 

diagnosed; RR: relapsed or refractory; DRd: daratumumab, lenalidomide and dexamethasone; DKd: 

daratumumab, carfilzomib and dexamethasone; Ig: immunoglobulin. 

 

To mimic the TME surrounding PC in MM, BMMCs from patients were obtained through 

density gradient and cultures were performed in the presence of the own patient BM 

plasma. The percentage of CD34+ cells and PC, as well as the expression of NKG2D-L, 

BCMA, and HLA-E within this population, were studied by flow cytometry (Table 13).  



RESULTS 

137 
 

 

Figure 35. NKG2D- and BCMA-CAR NKAE cells pretreated with α-NKG2A Abs exhibit a higher 

antitumor activity against PC from MM patients, with low toxicity over CD34+ cells from the same 

patient and PBMCs from HD. A) Representative flow cytometry dot plots of PC (CD38+CD138+) and 

CD34+ cells gating in BCMA-CAR NKAE cell and BMMC 24h-co-cultures. B) Specific lysis of NKG2D- 

and BCMA-CAR NKAE cells treated with either the mouse (Z199) or the human (BMS-986315) α-NKG2A 

Ab, or their corresponding isotypes (mIgG2b or hIgG1κ), against PC (CD38+CD138+) or CD34+ cells from 

MM patient BMMCs for 24 hours, analyzed by flow cytometry (Attune CytPix Flow Cytometer). PC : 

CAR NKAE cell ratio (1:10 for NKG2D-CAR NKAE cell experiments and 1:5 for BCMA-CAR NKAE 

cell experiments) was determined taking into account the percentage of PC within BMMCs. The mean ± 

SEM is shown (n=5). C) Specific lysis of NKG2D- and BCMA-CAR NKAE cells treated with either the 

mouse (Z199) or the human (BMS-986315) α-NKG2A Ab, or their corresponding isotypes (mIgG2b and 

hIgG1κ), against HD PBMCs analyzed by 3h-Calcein release assay. The mean ± SEM is shown (n=6). ns: 

not significant; *p<0.05; **p<0.01. 



RESULTS 

138 
 

An increased cytotoxic activity against PC was achieved at 1:10 T:E ratio with NKG2D-

CAR NKAE cells pretreated with α-NKG2A Abs (77.15 ± 4.07% with Z199 and 83.37 ± 

4.74% with BMS-986315) compared to NKG2D-CAR NKAE cells pretreated with their 

corresponding isotypes (58.79 ± 4.49% with mIgG2b and 62.27 ± 2.78% with hIgG1κ) 

(Fig. 35A, B). Similarly, α-NKG2A Ab pretreated BCMA-CAR NKAE cells increment 

the tumor cell lysis at 1:5 T:E ratio (79.55 ± 4.64% with Z199 and 85.03 ± 4.04% with 

BMS-986315) with respect to isotype pretreated cells (57.87 ± 3.79% with mIgG2b and 

57.11 ± 6.28% with hIgG1κ). No significant killing activity was observed over CD34+ 

progenitors with NKG2A blocked NKG2D- (33.26 ± 8.2% with Z199 and 31.92 ± 8.1% 

with BMS-986315) and BCMA-CAR NKAE therapies (31.1 ± 9.58% with Z199 and 

19.76 ± 7.59% with BMS-986315) compared to isotype pretreated NKG2D- (18.16 ± 

6.31% with mIgG2b and 17.19 ± 4.38% with hIgG1κ) and BCMA-CAR NKAE cells 

(14.94 ± 7.26% with mIgG2b and 12.85 ± 6.25% with hIgG1κ). 

Considering the improved efficacy of CAR populations pretreated with the α-NKG2A 

Abs, we performed additional toxicity assays against mature blood cells from HD. None 

of the combinations proposed cause toxicity against these PBMCs from HD. The 

maximum percentage of cell lysis was 17.8 ± 4.32% obtained with α-NKG2A blocked 

NKG2D-CAR NKAE cells at a ratio of 1:4 (T:E) (Fig. 35C). 

Taken together, these data demonstrate that NKG2A blockade enhances the cytolytic 

capacity of NKG2D- and BCMA-CAR NKAE cells against primary MM cells whereas 

exhibits low hematotoxicity. 

 

4.2.8. Pretreatment of NKG2D-CAR NKAE cells with the BMS-986315 α-

NKG2A blocking Ab increases survival in disseminated MM xenograft 

mouse model 

Once we had demonstrated the in vitro antitumor efficacy and reduced hematotoxicity of 

the combination of α-NKG2A blocking Abs with NKG2D- or BCMA-CAR NKAE cells, 

we tested its anti-MM activity in vivo. As only the human α-NKG2A Ab provided by 

BMS, BMS-986315, could be used in clinical practice, mouse experiments were carried 

out comparing this human Ab and its corresponding IgG1κ isotype treatment. 



RESULTS 

139 
 

For this purpose, we selected the mouse strain NOD.Cg-Prkdcscid Il2rgtm1Wjl 

Tg(IL15)1Sz/SzJ, NSG-Tg(Hu-IL15) with physiological serum levels of human IL-15 

that facilitates the maintenance of functional human NK cells. Additionally, as we have 

demonstrated that IL-15 promotes NKG2A expression, the use of NSG-Tg(Hu-IL15) 

strain enables the examination of these therapies efficacy in a more challenging/resistant 

scenario. 

In a first experiment, 5 x 105 RPMI-8226 ffLucGFP cells (Fig. 36A) were intravenously 

injected in sublethally irradiated mice. After 72 hours, 8 x 106 NKG2D- or BCMA-CAR 

NKAE cells pretreated with either the human α-NKG2A Ab or the corresponding hIgG1κ 

isotype were intravenously infused. To guarantee the blockade of NKG2A on CAR 

NKAE cells for a longer time, we intraperitoneally administered 250 µg of the hIgG1κ 

isotype or the α-NKG2A Ab at day 0 and day 11. Scheme treatments are shown in figure 

36B. The CAR-NK transduction efficiency on the day of infusion was 41.6% for 

NKG2D-CAR and 49.5% for BCMA-CAR (Fig. 36C). In addition, NKG2A blockade on 

CAR NKAE cells was corroborated by flow cytometry. The expression of NKG2A cells 

in NKG2D‑CAR NKAE population diminished from 83.61 RFI in cells pretreated with 

hIgG1κ to 11 RFI in cells incubated with α-NKG2A. Similarly, BCMA-CAR NKAE cells 

reduced the expression of NKG2A cells from 133.6 RFI in cells pretreated with hIgG1κ 

to 13.46 RFI in α-NKG2A‑treated cells. 



RESULTS 

140 
 

 



RESULTS 

141 
 

Figure 36. Intraperitoneally infusion of BMS-986315 α-NKG2A Ab in mice decreases NKG2D- and 

BCMA-CAR NKAE cell percentages in vivo. A) Representative flow cytometry dot plot of GFP 

expression on RPMI-8226 and RPMI-8266 ffLucGFP cells purified by cell sorting. B) Experimental design 

of the MM disseminated orthotopic NSG-Tg(Hu-IL15) mouse model generated through the intravenous 

(i.v.) infusion of 5 x 105 RPMI-8226 ffLucGFP cells followed by an i.v. infusion of 8 x 106 NKG2D- or 

BCMA-CAR NKAE cells pretreated with either the human α-NKG2A (BMS-986315) or its corresponding 

isotype (hIgG1κ) at day 3. Additionally, 250 µg of the α-NKG2A or the hIgG1κ isotype Abs were 

intraperitoneally (i.p.) infused at day 0 and day 11. PB was extracted for NK cell detection analysis by flow 

cytometry at day 7 and 14 after CAR NKAE cell infusion. Tumor burden was monitored weekly by BLI. 

C) Flow cytometry dot plot of NKG2D-CAR+ and BCMA-CAR+ cell percentages at infusion. D) Flow 

cytometry analysis of the percentage of NK cells and NKG2A+ and CAR+ within NK cells from mouse PB 

samples at day 7 and 14 after NK cell infusion (n=4). The mean ± SEM is shown. E) Imaging of tumor 

burden, monitored by bioluminescence, at the indicated time points from NSG-Tg(Hu-IL15) mice bearing 

RPMI-8226 ffLucGFP without treatment or treated with intraperitoneal Ab infusion and NKG2D- or 

BCMA-CAR NKAE cells, preincubated with either BMS-986315 or the hIgG1κ isotype. *p<0.05; 

**p<0.01; ***p<0.001; ****p<0.0001. 

 

Strikingly, few percentages of NK cells were detected in mice treated with the α-NKG2A 

Ab at day 7 (4.55 ± 0.87% in α-NKG2A NKG2D-CAR NKAE cell vs 8.43 ± 1.79% IgG1κ 

NKG2D-CAR NKAE cell treated mice and 3.41 ± 0.89% in α-NKG2A BCMA-CAR 

NKAE cell vs 11.58 ± 1.09% IgG1κ BCMA-CAR NKAE cell treated mice) and day 14 

(1.06 ± 0.13% in α-NKG2A NKG2D-CAR NKAE cell vs 1.41 ± 0.43% IgG1κ NKG2D-

CAR NKAE cell treated mice and 0.73 ± 0.24% in α-NKG2A BCMA-CAR NKAE cell 

vs 2.08 ± 0.41% IgG1κ BCMA-CAR NKAE cell treated mice) (Fig. 36D). All NK cells 

from α-NKG2A treated mice were NKG2A-, while high NKG2A expression was 

maintained in NK cells from isotype-treated mouse groups. Remarkably, CAR expression 

was superior in α-NKG2A pretreated populations compared to the isotype at day 7: 85.48 

± 0.87% in α-NKG2A NKG2D-CAR NKAE cells vs 72.43 ± 1.33% IgG1κ NKG2D-CAR 

NKAE cells and 67.58 ± 2.01% in α-NKG2A BCMA-CAR NKAE cells vs 60.73 ± 0.97% 

IgG1κ BCMA-CAR NKAE cells. 

First results from BLI studies shown that α-NKG2A treatment in combination with both 

CAR therapies delayed disease progression but, after 44 days, tumor burden was quite 

similar among all treated groups. Moreover, most mice did not have tumor cells in BM at 

necropsy, but they developed extramedullary tumors (Fig. 36E). 

Due to the decreased NK cell percentage in the α-NKG2A-treated groups, probably 

caused by a fratricide effect among NKG2A- cells, and the inability to properly measure 

MM cells within their niche when using the RPMI-8226 ffLucGFP model, we tested the 

efficacy of these combination therapies against another MM cell line, the U-266 



RESULTS 

142 
 

ffLucGFP (Fig. 37A), maintaining the pretreatment CAR NK cell approach, but 

eliminating the intraperitoneal Ab administrations. 

A pilot experiment was undertaken to find the optimal CAR NK cell dose range. 5 x 105 

U-266 ffLucGFP cells were intravenously injected in sublethally irradiated NSG-Tg(Hu-

IL15) mice. After 3 days, 8 x 106 NKG2D- or BCMA-CAR NKAE cells pretreated with 

either the human α-NKG2A Ab or the hIgG1κ isotype were intravenously administered 

(Fig. 37B). As in the previous experiment, CAR expression (Fig. 37C) (6.9% NKG2D-

CAR+ and 17.8% BCMA-CAR+) and the blockade of NKG2A on NK cells in α-NKG2A 

pretreated mice (5 NKG2A RFI in α-NKG2A NKG2D-CAR NKAE cells vs 112.78 RFI 

in hIgG1κ NKG2D-CAR NKAE cells and 6.29 NKG2A RFI in α-NKG2A BCMA-CAR 

NKAE cells vs 106.42 RFI in IgG1κ BCMA-CAR NKAE cells) were corroborated on the 

day of NK cell infusion. 



RESULTS 

143 
 

 

Figure 37. NKG2D- and BCMA-CAR NKAE cells persist for at least 29 days in NSG-Tg(Hu-IL15) 

mice bearing U-266 ffLucGFP MM cells. A) Representative flow cytometry dot plot of GFP expression 

on U-266 ffLucGFP cells purified by cell sorting. B) Experimental design of the disseminated MM NSG-

Tg(Hu-IL15) mouse model generated through the intravenous (i.v.) infusion of 5 x 105 U-266 ffLucGFP 

cells followed by an i.v. infusion of 8 x 106 NKG2D- or BCMA-CAR NKAE cells pretreated with either 

the human α-NKG2A (BMS-986315) or its corresponding isotype (hIgG1κ) at day 3. PB was extracted for 

NK cell detection analysis by flow cytometry at day 7, 14, and 29 after CAR NKAE cell infusion. Tumor 



RESULTS 

144 
 

burden was monitored fortnightly by BLI. C) Flow cytometry dot plot of the NKG2D-CAR+ and BCMA-

CAR+ cell percentages at infusion. D) Representative flow cytometry dot plot of NK cells gating in 

NKG2D- or BCMA-CAR NKAE cells pretreated with BMS-986315 α-NKG2A Ab or its isotype (hIgG1κ) 

at day 7 post-NK cell infusion. E) Flow cytometry analysis of the percentage of NK cells and NKG2A+ and 

CAR+ within NK cells from PB mouse samples at day 7, 14, or 29 after NK cell infusion (n=4). The mean 

± SEM is shown. 

 

PB samples collected at day 7, 14, and 29 post-NK administration revealed the CAR 

expression stability and NK cell persistence for at least 29 days post-therapy infusion. At 

day 14, the percentage of human NK cells was 18.38 ± 3.73% in hIgG1κ and 17.63 ± 

1.32% in BMS-986315 pretreated NKG2D-CAR NKAE cell mouse group and 17.33 ± 

1.84% in hIgG1κ and 16.95 ± 4.98% in BMS-986315 pretreated BCMA-CAR NKAE cell 

mouse group. No significant differences in NK cell frequencies or CAR expression were 

observed between mice infused with hIgG1κ or α-NKG2A-pretreated CAR NK cells. 

Moreover, since intraperitoneally infusions of BMS-986315 were replaced by 

pretreatment, at day 7, all CAR NKAE expressed a high percentage of NKG2A (Fig. 37 

D, E). 



RESULTS 

145 
 

 

Figure 38. NKG2D-CAR NKAE cells pretreated with either the BMS-986315 Ab or its isotype 

significantly reduce tumor burden in NSG-Tg(Hu-IL15) mice bearing U-266 ffLucGFP MM cells. A) 

Imaging of tumor burden monitored by bioluminescence at the indicated time points from NSG-Tg(Hu-

IL15) mice bearing U-266 ffLucGFP cells without treatment or treated with NKG2D-CAR NKAE cells, 

preincubated with either BMS-986315 or the hIgG1κ isotype. B) Kaplan-Meier survival curve of mice 

described in (A) (n=4). C) Flow cytometry analysis of HLA-E expression on U-266 ffLucGFP cells in 

culture or collected from BM mice at necropsy (n=4). The mean ± SEM is shown. D) Flow cytometry 

analysis of GFP+ tumor cells in BM mice at necropsy from the indicated experimental groups (n=4). The 

mean ± SEM is shown. E) Flow cytometry dot plot of residual tumor cells in the BM of indicated mice in 

(A). Flow cytometry analysis of the number of U-266 ffLucGFP cells detected in the BM of mice #1, #2, 

and #3 from α-NKG2A-pretreated NKG2D-CAR NKAE cell group, sacrificed at day 150. One U-266 

ffLucGFP untreated mouse analysis is shown. F) Imaging of tumor burden monitored by bioluminescence, 

at the indicated time points from NSG-Tg(Hu-IL15) mice bearing U-266 ffLucGFP cells without treatment 



RESULTS 

146 
 

or treated with BCMA-CAR NKAE cells, preincubated with either BMS-986315 or the hIgG1κ isotype. 

*p<0.05; **p<0.01. 

 

Regarding NKG2D-CAR NKAE cells, both pretreated, α-NKG2A (undefined) and 

isotype (˃135.5 days), CAR NKAE cells significantly prolong mice survival compared 

to the vehicle group (67.5 days) (Fig. 38A, B). Of note, HLA-E expression was 

significantly superior on U-266 ffLucGFP from mice analyzed at necropsy (14.24 ± 3.34 

RFI) compared to U-266 ffLucGFP in cell culture (4.64 ± 0.74 RFI) (Fig. 38C). 

Tumor cells were detected at necropsy in BM mice by flow cytometry (Fig. 38D). 

Remarkably, after 150 days, 2 mice from the isotype and 3 from the NKG2A blockade 

group did not have clinical signs of the disease or BLI signal and, consequently, the 

experiment was discontinued. Flow cytometry analysis was conducted to assess residual 

MM cells in the BM of these mice: #1 (16 PC on 4.64 x 106 viable events), #2 (2 PC on 

4.01 x 106 viable events), and #3 (2 PC on 1.28 x 106 viable events) (Fig. 38E). 

Concerning BCMA-CAR NKAE cells, all mice were rescued from death (Fig. 38F). 

These results indicated that both NKG2D- and BCMA-CAR NKAE therapies were able 

to eliminate tumor burden in 62.5% and 100% of mice, respectively. For that reason, we 

reduced the number of infused CAR NK cells to test the efficacy of NKG2A-blocked 

CAR NK cells under suboptimal conditions, mimicking low therapy:tumor cell 

proportions in RRMM patients. 

For this purpose, 4 x 106 NKG2D- or BCMA-CAR NKAE cells pretreated with either the 

human α-NKG2A Ab or the hIgG1κ isotype, were infused in mice sublethally irradiated 

72h after U-266 ffLucGFP infusion (5 x 105 cells), as indicated in Figure 39A. Once 

again, we confirmed CAR expression (37.2% NKG2D-CAR+ and 30.1% BCMA-CAR+) 

(Fig. 39B) and NKG2A blockade in all infused populations (18.2 NKG2A RFI in α-

NKG2A NKG2D-CAR NKAE cells vs 126.39 RFI in hIgG1κ NKG2D-CAR NKAE 

cells, and 12.78 NKG2A RFI in α-NKG2A BCMA-CAR NKAE cells vs 71.51 RFI in 

hIgG1κ BCMA-CAR NKAE cells). 



RESULTS 

147 
 

 

Figure 39. Lower doses of NKG2D- and BCMA-CAR NKAE cells can be also detected at least 29 

days from infusion in NSG-Tg(Hu-IL15) mice bearing U-266 ffLucGFP MM. A) Experimental design 

of the disseminated MM NSG-Tg(Hu-IL15) mouse model generated through the intravenous (i.v.) infusion 

of 5 x 105 U-266 ffLucGFP cells followed by an i.v. infusion of 4 x 106 NKG2D- or BCMA-CAR NKAE 

cells pretreated with either the human α-NKG2A (BMS-986315) or its corresponding isotype (IgG1κ) at 

day 3. PB was extracted for NK cell detection analysis by flow cytometry at day 7, 14, and 29 after CAR 

NKAE cell infusion. Tumor burden was monitored fortnightly by BLI. B) Flow cytometry dot plot of 

NKG2D-CAR+ and BCMA-CAR+ cell percentages at infusion. C) Flow cytometry analysis of the 

percentage of NK cells and NKG2A+ and CAR+ within NK cells from mouse PB samples at day 7, day 14, 

or day 29 after NK cell infusion (n=4). The mean ± SEM is shown. 

 

Despite having halved the cell therapy dose compared to the previous in vivo study, NK 

cells remain detected at least 29 days post infusion in all experimental mice, showing also 

a stable CAR expression (Fig. 39C). Regarding NKG2D-CAR therapy, both CAR NK 



RESULTS 

148 
 

cell conditions increased mice survival over untreated (71 days), but the blockade of the 

NKG2A significantly delayed tumor development (˃143 days) compared to the isotype 

group (93.5 days) (Fig. 40A, B). Quantification of the luminescence photon flux signal 

was measured along the experiment (Fig. 40C) and tumor cells in the BM of all mice were 

analyzed at necropsy by flow cytometry (Fig. 40D). At day 154, the experiment was 

discontinued and tumor cell infiltration in the BM of the remaining live mice (#1 and #2) 

was studied (Fig. 40E).   

Surprisingly, hIgG1κ and α-NKG2A-pretreated BCMA-CAR NKAE cells rescued all 

mice from death (Fig. 40F). Therefore, we could not evaluate the differences in antitumor 

activity between the BCMA-CAR NKAE cell treatments. 

 



RESULTS 

149 
 

Figure 40. BMS-986315 pretreatment on NKG2D-CAR NKAE cells significantly prolongs mouse 

survival in NSG-Tg(Hu-IL15) mice bearing U-266 ffLucGFP MM cells.  A) Imaging of tumor burden 

monitored by bioluminescence, at the indicated time points from NSG-Tg(Hu-IL15) mice bearing U-266 

ffLucGFP cells without treatment or treated with NKG2D-CAR NKAE cells, preincubated with either 

BMS-986315 or the hIgG1κ isotype. B) Kaplan-Meier survival curve of mice described in (A) (n=4). C) 

Quantification of luminescence per mouse at different time points. D) Flow cytometry analysis of 

CD138+GFP+ tumor cells in BM mice at necropsy from the indicated experimental groups (n=4). The mean 

± SEM is shown. E) Flow cytometry dot plot of tumor cells in the BM of mice indicated in (A). Flow 

cytometry analysis of the number of U-266 ffLucGFP cells detected in the BM of mice #1 and #2 from α-

NKG2A-pretreated NKG2D-CAR NKAE cell groups, sacrificed at day 154. One U-266 ffLucGFP 

untreated mouse is shown. F) Imaging of tumor burden monitored by bioluminescence at the indicated time 

points from NSG-Tg(Hu-IL15) mice bearing U-266 ffLucGFP cells without treatment or treated with 

BCMA-CAR NKAE cells, preincubated with either BMS-986315 or the hIgG1κ isotype. *p<0.05; 

**p<0.01. 

 

Taken together, these results demonstrate that NKG2A blockade improves the in vivo 

killing efficacy of NKG2D-CAR NK cells, but further experiments with lower CAR NK 

cell doses would be required to evaluate the effect of blocking NKG2A in BCMA-CAR 

NKAE cells. 

 

4.3. NKG2D-CAR CIML-NKAE cells have superior efficacy 

against MM cells in vitro and in vivo 

4.3.1. Generation of NKG2D-CAR NKAE and CIML-NKAE cells from HD PB 

samples is feasible 

CIML-NK cells are a unique approach to heighten in vivo expansion, persistence, and 

antitumor responses of NK cells [437, 457]. As the memory phenotype is induced through 

a short exposure to IL-12, IL-15, and IL-18 on fresh purified NK cells [432], we asked 

whether we could potentiate our NKG2D-CAR NKAE therapy with additional 

characteristics described for CIML cells by the incubation with this cytokine cocktail. 

NKG2D-CAR NKAE cells were generated as previously detailed and, right after 

lentivector removal, were exposed to IL-12, IL-15, and IL-18 for 16 hours. After 5-7 days, 

experiments to characterize this population were carried out. 

Characterization of NKG2D-CAR CIML-NKAE and NKAE cells was performed by flow 

cytometry (Fig. 41A). These CAR NKAE cells were generated from PB samples with a 

low percentage of NK cells (5.78 ± 0.74% of CD56+ cells), mostly CD16+ (75.33 ± 



RESULTS 

150 
 

5.93%), and a high percentage of T cells (55.48 ± 5.97% of CD3+ cells). At day 14, both 

NKG2D-CAR NKAE cell cultures achieved a purity of more than 96% CD56+ cells 

(97.85 ± 0.39% in the CIML-NKAE and 96.24 ± 0.79% in NKAE cultures), most of them 

CD16+ (83.97 ± 2.68% in the CIML population and 88.08 ± 2.72% in the NKAE subset), 

with extremely low contamination of CD3+ cells (0.02 ± 0.01% in CIML-NKAE cells 

and 0.04 ± 0.02% in NKAE cells) (Fig. 41B). 

Flow cytometry data showed no significant differences in NKG2D-CAR transduction 

levels between both populations, with a mean value of 15.65 ± 2.85% in NKAE cells and 

19.89 ± 3.49% in the CIML-NKAE cells (Fig. 41C, D). VCN results were similar between 

therapies, 0.25 ± 0.08 copies per cell in NKG2D-CAR NKAE cells and 0.31 ± 0.07 copies 

per cell in NKG2D-CAR CIML-NKAE cells (Fig. 41D). Remarkably, transduction 

percentages, analyzed by flow cytometry, and VCN values, obtained by q-PCR, have a 

high correlation (R2=0.971 in NKG2D-CAR NKAE cells and R2=0.9372 in NKG2D-

CAR CIML-NKAE cells) (Fig. 41E). Additionally, we used QuantibriteTM beads to 

specifically compare the amount of NKG2D molecules on the surface of these 

populations. It is important to take into account that the number of molecules includes not 

only the CAR constructs but also the endogenous NKG2D receptor. Results show that 

transduction with the NKG2D-CAR enhances the presence of this receptor on both 

NKG2D-CAR CIML-NKAE cells (2,087.13 ± 278.46 molecules on the CAR subset vs 

832.58 ± 170.57 on untransduced CIML-NKAE cells) and NKG2D-CAR NKAE cells 

(1,667.35 ± 366.92 molecules on the CAR population vs 631.47 ± 234.63 on 

untransduced NKAE cells) (Fig. 41F). 



RESULTS 

151 
 

 

Figure 41. CIML-NKAE and NKAE cells are efficiently generated and stably transduced with 

lentivectors containing NKG2D-CAR. A) Representative flow cytometry dot plot of NK cells 

(CD56+CD3-), cytotoxic NK cells (CD56+CD16+), and T cells (CD3+) from NKG2D-CAR CIML-NKAE 

and NKAE cell cultures at day 0 and day 14. B) Flow cytometry analysis of the percentage of NK cells 

(CD56+CD3-), cytotoxic NK cells (CD56+CD16+), and T cells (CD3+) cells in NKG2D-CAR CIML-NKAE 

and NKAE cell cultures at day 0 and day 14 (n=6). Means ± SEM are shown. C) Representative flow 

cytometry dot plot of NKG2D-CAR expression on the surface of CIML-NKAE and NKAE cells at 6 days 

post-transduction. D) Flow cytometry analysis and vector copy number (VCN) quantification by q-PCR of 

NKG2D-CAR in CIML-NKAE and NKAE cells at day 6 post-transduction (n=8). The mean ± SEM is 

shown. E) Correlation between the percentage of NKG2D-CAR+ cells and the VCN per cell on CIML-

NKAE and NKAE cells after 6 days from transduction (n=8). F) Quantification of the NKG2D molecules 



RESULTS 

152 
 

on the cell surface of the CIML-NKAE and NKAE cells, untransduced or transduced with the NKG2D-

CAR lentivector (n=5) by QuantibriteTM beads for flow cytometry analysis at 6 days post-transduction. 

Means ± SEM are shown. G) Flow cytometry analysis of the percentage of NKG2C+, NKG2A+ and PD-1+ 

cells within CD56+CD3- populations on NKG2D-CAR CIML-NKAE and NKAE cell cultures at day 0 and 

day 14 (n=6). Means ± SEM are shown. ns: not significant; ***p<0.001; ****p<0.0001. 

 

Moreover, we studied the expression changes regarding other NKG2 receptors, NKG2C 

and NKG2A, and PD-1. NKG2C (29.59 ±  4.05% NKG2C+ cells within NKG2D-CAR 

CIML-NKAE cells and 31.02 ± 5.93% in NKG2D-CAR NKAE cells at day 14 vs 24.39 

± 4.93 at day 0) and PD-1 expression (23.29 ± 8.94% PD-1+ cells within  NKG2D-CAR 

CIML-NKAE cells and 23.97 ± 10.32% in NKG2D-CAR NKAE cells at day 14 vs 12.59 

± 2.04 at day 0) did not experiment any change during CAR NKAE cell generation, while 

NKG2A expression (89.46 ± 3.42% NKG2A+ cells within NKG2D-CAR CIML-NKAE 

cells and 83.88 ± 6.08% in NKG2D-CAR NKAE cells at day 14 vs 52.49 ± 7.44 at day 

0) augmented after activation and expansion of NK cells (Fig. 41G). 

These results indicate the feasibility of generating NKG2D-CAR CIML-NKAE cells 

from NKG2D-CAR NKAE cells, with similar CAR transduction efficiency and 

expression of NKG2 and PD-1 receptors.    

 

4.3.2. NKG2D-CAR CIML-NKAE cells exhibit higher cytotoxicity and 

degranulation capacity against MM cell lines 

To test their anti-MM activity, both NKG2D-CAR populations were co-cultured with the 

MM cell lines XG-1 and RPMI-8226 at different T:E ratios. NKG2D-CAR CIML-NKAE 

cells are significantly more efficient in killing these two MM cell lines, XG-1 (58.03 ± 

2.88% specific lysis vs 41.23 ± 5.18%, at 1:8 T:E ratio) and RPMI-8226 (80.02 ± 4.45% 

specific lysis vs 62.81 ± 7.11%, at 1:8 T:E ratio), than NKG2D-CAR NKAE cells (Fig. 

42A). 

Furthermore, we studied the antitumor ability of NKG2D-CAR CIML-NKAE or NKAE 

cells against the MM cell line, L-363, which has the capacity to grow forming colonies 

in methylcellulose semi-solid medium. At 4:1 T:E co-culture, NKG2D-CAR CIML-

NKAE cells significantly reduce L-363 colony percentage (53.93 ± 11.29%), compared 

to NKG2D-CAR NKAE cells (85.59 ± 4.34%) (Fig. 42B). 



RESULTS 

153 
 

 

Figure 42. Anti-MM and degranulation activity is augmented in NKG2D-CAR CIML-NKAE cells. 

A) Specific lysis of NKG2D-CAR CIML-NKAE and NKAE cells against the MM cells RPMI-8226 (n=6) 

and XG-1 (n=7) analyzed by 3h-Calcein release assay. The mean ± SEM is shown. B) At the left, 

quantification of the percentage of L-363 colonies after 14 days of co-culture with NKG2D-CAR CIML-

NKAE or NKAE cells in methylcellulose semi-solid medium (n=3). At the right, representative images of 

plates containing L-363 cells alone or after co-culture with CAR NKAE cells at different T:E ratios. CAR 

NKAE cell control plates are also shown. C) At the left, representative flow cytometry dot plot of CD107a+ 

staining in NKG2D-CAR CIML-NKAE and NKAE cells after 4h co-culture with RPMI-8226 cells. At the 

right, flow cytometry analysis of CD107a expression on NKG2D-CAR CIML-NKAE and NKAE cells 

(n=4) alone or after 4-hour co-culture with the MM cells U-266 and RPMI-8226, or the feeder cell K562-

mb21-41BBL at 1:1 ratio. The mean ± SEM is shown. *p<0.05; **p<0.01; ***p<0.001; ****p<0.0001. 

 

To corroborate the higher cytolytic potential of NKG2D-CAR CIML-NKAE cells, 

degranulation assays were performed. A compilation of 4 experiments shows that 



RESULTS 

154 
 

NKG2D-CAR CIML-NKAE cells have increased degranulation levels, compared to 

NKG2D-CAR NKAE cells, against the MM cell lines U-266 (71.39 ± 4.21% vs 64.48 ± 

3.41%) and RPMI-8226 cells (37.34 ± 9.97% vs 27.12 ± 7.57%) (Fig. 42C). 

Altogether, although NKG2D-CAR NKAE cells display high cytotoxicity and 

degranulation levels against MM target cells, the induction of the memory phenotype in 

these cells significantly improves their killing activity. 

 

4.3.3. Cytotoxic cytokine production is increased in NKG2D-CAR CIML-NKAE 

cells  

Because of cytokine release profile is key for NK cell killing activity, we studied whether 

NKG2D-CAR CIML-NKAE cells secrete more cytotoxic cytokines or cytolytic proteins 

after interaction with tumor cells. First, intracellular IFN-γ was measured by flow 

cytometry in both therapies after their co-culture with two MM cell lines, U-266 and 

RPMI-8226. In addition, we included in the assay the modified feeder cell K562-mb21-

41BBL, used to generate these populations, as CIML-NK cells have been described to 

quickly react to previous stimuli [431]. 

Results show that NKG2D-CAR CIML-NKAE cells present high intracellular IFN-γ 

levels against U-266 (44.87 ± 7.2% vs 28.21 ± 5.84%), RPMI-8226 (41.68 ± 11.01% vs 

25.99 ± 7.06%) and K562-mb21-41BBL (65.32 ± 8.21 vs 47.64 ± 8.75%), compared to 

NKG2D-CAR NKAE cells (Fig. 43A). 



RESULTS 

155 
 

 

Figure 43. NKG2D-CAR CIML-NKAE cells have higher IFN-γ intracellular production and IFN-γ, 

perforin, sFasL and granzyme A release, compared to NKG2D-CAR NKAE cells. A) At the left, 

representative flow cytometry dot plot of IFN-γ staining in NKG2D-CAR CIML-NKAE and NKAE cells 

after 4h co-culture with RPMI-8226 cells. At the left, flow cytometry analysis of IFN-γ intracellular 

expression on NKG2D-CAR CIML-NKAE and NKAE cells alone or after 4-hour co-culture with the MM 

cells U-266 and RPMI-8226, and the feeder cell K562-mb21-41BBL, at ratio 1:1 (n=4). The mean ± SEM 

is shown. B) Quantification of cytokine/cytolytic proteins (IFN-γ, soluble FasL, granzyme A, granzyme B, 

perforin and granulysin) in supernatants of NKG2D-CAR CIML-NKAE or NKAE cells 24-hour alone or 

co-cultured with the MM cells U-266, XG-1, and RPMI-8226, or the feeder cell line K562-mb21-41BBL, 

at 1:2 T:E ratio, by LEGENDplex bead-based immunoassay (n=9). The mean ± SEM is shown. ns: not 

significant; *p<0.05; **p<0.01; ****p<0.0001. 

 

To characterize the secretion profile of NK cells, a panel of 13 cytotoxic and cytolytic 

proteins, including IL-2, IL-4, IL-6, IL-10, IL-17A, IFN-γ, TNF-α, soluble Fas, soluble 

FasL, granzyme A, granzyme B, perforin, and granulysin, was analyzed by flow 

cytometry. NKG2D-CAR NKAE and CIML-NKAE cells were cultured alone or with the 

U-266, XG-1, RPMI-8226 and K562-mb21-41BBL cells for 24h and the collected 

supernatant was used for cytokine/cytolytic protein release determination. As shown in 



RESULTS 

156 
 

Figure 43B, NKG2D-CAR CIML-NKAE cells secrete higher concentrations of IFN-γ, 

perforin, granzyme A and soluble FasL, alone or after co-culture with tumor cell lines, 

compared to NKG2D-CAR NKAE cells. After 24h co-culture of NKG2D-CAR CIML-

NKAE cells with U-266 cells, the concentration of IFN-γ was 2,939.18 ± 734.2 pg/ml, of 

perforin 1,294.6 ± 232.39 pg/ml, of granzyme A 4764.72 ± 772.3 pg/ml, and of soluble 

FasL 179.73 ± 26.98 pg/ml compared to 922.48 ± 411.23 pg/ml IFN-γ, 810.73 ± 86.68 

pg/ml perforin, 2,608.64 ± 690.49 pg/ml granzyme A, and 109.98 ± 15.85 pg/ml soluble 

FasL in NKG2D-CAR NKAE cell co-cultures. 

Overall, these results show that memory induction on NKG2D-CAR NKAE cells is able 

to modify the cytokine and cytolytic protein production, enhancing NK cell antitumor 

efficacy. 

 

4.3.4. NKG2D-CAR CIML-NKAE cells have enhanced proliferation activity in 

vitro 

One of the main characteristics of the CIML population is its high proliferation and 

persistence potential [436, 439]. For this reason, we studied NKG2D-CAR CIML-NKAE 

proliferative capacity through two assays: by measuring the percentage of Ki67+ cells 

within the total population and the percentage of cells that are duplicating their DNA and 

entering mitosis (phases S, G2, and M). An average of 77.94 ± 2.62% of NKG2D-CAR 

CIML-NKAE cells are Ki67+, compared to 71.73 ± 2.66% of NKG2D-CAR NKAE cells 

(Fig. 44A). Consistently, the percentage of NKG2D-CAR CIML-NKAE cells entering 

mitosis is higher in NKG2D-CAR CIML-NKAE cells (32.33 ± 2.2%) than in NKG2D-

CAR NKAE population (24.19 ± 2.47%) (Fig. 44B). 

Both assays demonstrate that the exposure of NKG2D-CAR NKAE cells to IL-12, IL-15, 

and IL-18 enhances cell proliferation capacity. 



RESULTS 

157 
 

 

Figure 44. Cytokine-induction of memory promotes proliferation activity and glycolytic state in CAR 

NKAE cells. A) Representative flow cytometry histogram of Ki67 staining of NKG2D-CAR CIML-NKAE 

and NKAE cells (left). Flow cytometry analysis of the percentage of Ki67+ cells on NKG2D-CAR CIML-

NKAE and NKAE cell populations (right). B) Representative flow cytometry histogram of cell cycle 

propidium iodide-stained NKG2D-CAR CIML-NKAE and NKAE cells (left). Flow cytometry cell cycle 

analysis of the percentage of S-G2-M cells on NKG2D-CAR CIML-NKAE and NKAE cells (right). The 

mean ± SEM is shown (n=14). C) At the left, ECAR measured by Seahorse assay comparing NKG2D-CAR 

CIML-NKAE and NKAE cells. At the right, bar graphs summarize the glycolytic activity (n=5). The mean 

± SEM is shown. D) OCR measured by Seahorse assay comparing NKG2D-CAR CIML-NKAE and NKAE 

cells (n=5). E) MitoTracker Green flow cytometry staining in NKG2D-CAR CIML-NKAE and NKAE cells 

(n=4). The mean ± SEM is shown. ns: not significant; **p<0.01; ***p<0.001 

 

4.3.5. NKG2D-CAR CIML-NKAE cells have increased glycolysis 

As the short exposure to IL-12/IL-15/IL-18 prompts a sustained metabolic switch of NK 

cells towards glycolysis [453], we compared the ECAR and OCR between both 

populations. Our results suggest that NKG2D-CAR CIML-NKAE cells have increased 

glycolysis (32.94 ± 2.63 mpH/min) compared to NKG2D-CAR NKAE cells (20.72 ± 1.77 

mpH/min) (Fig. 44C). However, NKG2D-CAR CIML-NKAE cells do not have superior 

mitochondrial oxidative phosphorylation compared to NKG2D-CAR NKAE cells (Fig. 

44D). 

To assess whether the differences in glycolysis were caused by cellular mitochondrial 

content, we evaluated mitochondrial mass using MitoTracker Green staining. Results 

show that there is no differential mitochondrial mass between NKG2D-CAR CIML-

NKAE and NKAE cells. 



RESULTS 

158 
 

These results suggest that NKG2D-CAR CIML-NKAE has an increase glucose 

metabolism. 

 

4.3.6. NKG2D-CAR NKAE and CIML-NKAE cells have a differential 

transcriptome and epigenome profile  

To better characterize the NKG2D-CAR CIML-NKAE cells, RNA-seq transcriptome 

analysis was performed in both populations. First, samples were analyzed based on the 

PC variations. Considering the 3,000 genes with the higher variability between NKG2D-

CAR NKAE and CIML-NKAE cells, the most variation (PC1) and the second most 

variation (PC2) were determined. PC1 had a 52% of variance while PC2 had 26%. As 

expected, most differences are between individuals and not by cell type (Fig. 45A).  

However, 162 differentially expressed genes (DEG) were found between CAR NKAE 

populations. It is noteworthy the upregulated expression of genes involved in cell 

proliferation, including CCND2 (cyclin D2) [471], TNRF11A (RANK) [472], MANF 

[473], and CD71 [453] in NKG2D‑CAR CIML-NKAE cells, as well as the chemokine 

receptors CCR4 and CCR7. Additionally, CABLES1 [474] and KLRG1 [475] expression 

is downregulated in this population, proteins that negatively regulate NK cell growth and 

function. By contrast, higher expression levels of the immune checkpoint TIGIT were 

showed in NKG2D-CAR NKAE cells. 

To understand the biological significance of these 162 DEG in NK cells, gene ontology 

(GO) analysis was conducted. NKG2D-CAR CIML-NKAE cells are enriched for 

pathways related to migration, regulation of IL-12 production and chemotaxis. The ratio 

(quantity of genes that belong to a specific group) and the odd ratio (probability of a DEG 

to belong to a specific group), including its −log10(APVAL), are represented in figure 

45A. 



RESULTS 

159 
 

 

Figure 45. NKG2D-CAR CIML-NKAE cells have a differential epigenetic and transcriptomic profile, 

compared to NKG2D-CAR NKAE cells. A) At the left, principal component analysis among 4 paired 

NKG2D-CAR CIML-NKAE and NKAE cells. In the middle, volcano plot of RNA-seq transcriptomic 

analysis. At the right, gene ontology biological process enrichments from 162 DEG. B) At the left, 

supervised hierarchical clustering of differentially methylated CpG sites between NKG2D-CAR CIML-

NKAE and NKAE cells (n=4). At the right, gene ontology analysis of genes containing CpG sites 

differentially expressed in the NKG2D-CAR CIML-NKAE cells, obtained from Reactome pathway. 

 

To further investigate the differences associated with cytokine memory induction, the 

epigenetic profile of NKG2D-CAR CIML-NKAE and paired NKG2D-CAR NKAE cells 

from 4 HD was analyzed. DNA methylation levels differed between both populations at 

115 CpG sites, located in regions that potentially impact on the expression of 86 genes. 

GO analysis revealed that the most overrepresented pathways are involved in cytokine 

signaling, cytoskeletal remodeling, and cell‑extracellular matrix interactions (Fig. 45B). 

Taken together, the transcriptomic and epigenomic analyses reveal different expression 

profiles between CAR effectors due to the exposure to IL-12, IL-15, and IL-18. 

 



RESULTS 

160 
 

4.3.7. NKG2D-CAR NKAE and CIML-NKAE cells exhibit differential 

expression of extracellular markers 

To validate the results obtained by RNA-seq and to identify additional NK cell 

extracellular markers with different expression levels between both CAR populations, a 

detailed flow cytometry phenotyping was conducted.  

Results reveal that NKG2D-CAR CIML-NKAE cells have higher expression of the 

costimulatory protein CD2 (1,747 ± 222.6 vs 1,409 ± 192.1 RFI), the activation receptor 

CD25 (164.6 ± 31.59 vs 80.12 ± 21.92 RFI) and the transferrin receptor CD71 (495.2 ± 

134.3 vs 325.4 ± 84.1 RFI), compared to NKG2D-CAR NKAE cells. By contrast, the 

expression of the activating receptors NKp30 (113.1 ± 14.04 vs 131.4 ± 17.36 RFI), 

NKp44 (28.18 ± 4.681 vs 53.26 ± 8.505 RFI), NKp46 (79.45 ± 11.77 vs 91.93 ± 12.81 

RFI) is downregulated on NKG2D-CAR CIML-NKAE cells. Similarly, NKG2D-CAR 

CIML-NKAE cells have lower levels of CD57 (6.78 ± 2.12 vs 11.71 ± 3.38 RFI) and the 

glucose transporter GLUT1 (37.64 ± 13.39 vs 57.99 ± 19.65 RFI). Moreover, CD62L 

expression (49.26 ± 16.39 vs 15.11 ± 2.2 RFI) is slightly higher in NKG2D-CAR CIML-

NKAE cells, while NKG2A expression diminishes (415.6 ± 100.5 vs 807.4 ± 218.5 RFI), 

although these last results are not statistically significant (Fig. 46). 

These results suggest that NKG2D-CAR CIML-NKAE cells have increased activation 

(CD25) as well as a high costimulation, through CD2. Besides, CD71 is highly expressed, 

corroborating RNA-seq results. However, NCRs expression is lower compared to 

NKG2D-CAR NKAE cells, although functional assays have demonstrated a higher 

cytotoxicity. Remarkably, decreased CD57 expression means NKG2D-CAR CIML-

NKAE population is less mature. 



RESULTS 

161 
 

 

Figure 46. NKG2D-CAR CIML-NKAE cells have a higher expression of CD25, CD2 and CD71, while 

display a reduced expression of NCRs, GLUT1, and CD57. Flow cytometry analysis of different markers 

expressed on the surface of NKG2D-CAR CIML-NKAE or NKAE cells (n=7). *p<0.05; **p<0.01. 

 

4.3.8. NKG2D-CAR CIML-NKAE cells display higher cytotoxicity against 

primary MM cells with low toxicity over CD34+ hematopoietic stem cells 

After demonstrating NKG2D-CAR CIML-NKAE cell superior antitumor activity against 

MM cell lines, we tested their efficacy against primary MM cells. For this purpose, 

NKG2D-CAR NKAE and CIML-NKAE cells were co-cultured with BMMCs from 3 

MM patients (Table 13) for 24 hours, as previously described. 

The cytotoxic capacity of NKG2D-CAR NKAE cells was 29.72 ± 2.43% at ratio 1:10 

(T:E), whereas the NKG2D-CAR CIML-NKAE population achieved a mean value of  

53.65 ± 7.23% . Remarkably, the killing activity against CD34+ progenitors did not show 

significant differences between both populations (8 ± 4.35% in NKG2D-CAR NKAE 

cells and 10.61 ± 5.55% in NKG2D-CAR CIML-NKAE cells) (Fig. 47A). 

To corroborate the absence of toxicity of NKG2D-CAR CIML-NKAE cells against 

healthy tissues, clonogenic assays were performed in methylcellulose semi-solid medium 

by co-culturing both CAR NK therapies with CD34+ cells, obtained from CB. At the 

highest T:E ratio 1:16, CFU-GM colony percentage is similar to the condition without 

effector cells and BFU-E colony percentage only decreases by a mean of 23% (Fig. 47B). 

In both cases, colony average percentages did not significantly differ with respect to the 

culture with CD34+ cells alone.  



RESULTS 

162 
 

Taken together, these results reveal that NKG2D-CAR CIML-NKAE cells exert a higher 

antitumor activity against primary MM cells without provoking hematotoxicity. 

 

 

Figure 47. Cytokine-induction of memory in NKG2D-CAR NKAE cells significantly improves the 

lysis of PC from MM patients, without compromising CD34+ population from the same patients or 

isolated from CB samples. A) Specific lysis of NKG2D-CAR CIML-NKAE or NKAE cells against PC 

(CD38+CD138+) or CD34+ progenitors from MM patient BMMCs after 24h-co-culture analyzed by flow 

cytometry (Attune CytPix Flow Cytometer). PC : CAR NKAE cell ratio (1:10) was determined taking into 

account the percentage of PC within BMMCs. The mean ± SEM is shown (n=3). B) Quantification of the 

percentage of CFU-GM (at the top) and BFU-E colonies (at the bottom) after 14 days co-cultured of CD34+ 

cells with NKG2D-CAR CIML-NKAE or NKAE cells at different T:E ratios in methylcellulose semi-solid 

medium (n=4). Means ± SEM are shown. ns: not significant; *p<0.05. 

 

4.3.9. NKG2D-CAR CIML-NKAE cells display heightened antitumor efficacy in 

a disseminated MM xenograft mouse model 

NKG2D-CAR CIML-NKAE cell in vivo efficacy and persistence were evaluated in a 

disseminated MM xenograft NSG-Tg(Hu-IL15) mouse model. At day 0, sublethally 

irradiated mice were intravenously injected with 5 x 105 U-266 ffLucGFP cells. After 3 

days, 4 x 106 NKG2D-CAR NKAE or CIML-NKAE cells were intravenously 

administered (Fig. 48A). The percentage of NKG2D-CAR+ cells was 37.2% in NKG2D-

CAR NKAE cells and 48.1% in NKG2D-CAR CIML-NKAE cells (Fig. 48B). 



RESULTS 

163 
 

 

Figure 48. The percentage of NKG2D-CAR CIML-NKAE cells in mouse PB is superior, compared to 

NKG2D-CAR NKAE cells at day 7 post-NK cell infusion. A) Experimental design of the MM 

disseminated orthotopic NSG-Tg(Hu-IL15) mouse model generated through the intravenous  (i.v.) infusion 

of 5 x 105 U-266 ffLucGFP cells followed by an i.v. infusion of 4 x 106 NKG2D-CAR CIML-NKAE and 

NKAE cells at day 3. PB was extracted for NK cell detection analysis by flow cytometry at day 7, 14, and 

29 after CAR NKAE cell infusion. Tumor burden was monitored fortnightly by BLI. B) Flow cytometry 

dot plot of the percentage of NKG2D-CAR+ cells at infusion. C) Flow cytometry analysis of the percentage 

of NK cells and CAR+ NK cells from mouse PB samples at day 7, day 14, or day 29 after NK cell infusion 

(n=4). The mean ± SEM is shown. *p<0.05; ****p<0.0001. 

 

PB analysis was carried out at 7-, 14-, and 29-days post-therapy infusion to evaluate NK 

cell persistence in mice. At day 7, the average infiltration of human NK cells was 

significantly higher in the NKG2D-CAR CIML-NKAE cells (17.73 ± 2.67%), compared 

to the NKG2D-CAR NKAE cells (9.64 ± 1.87%) (Fig. 48C). In addition, at day 14, NK 



RESULTS 

164 
 

cell percentage was still superior in the memory group and, remarkably, at this point, the 

percentage of NKG2D-CAR+ CIML-NKAE cells was significantly higher (49.75 ± 

0.81%), which could signify that CAR+ cells were proliferating more in the memory 

group, compared to NKG2D-CAR NKAE cells (40.25 ± 0.41%). However, at day 29, the 

percentage of NK cells was similar in both groups, suggesting that NKG2D-CAR CIML-

NKAE cells do not exhibit a higher persistence, compared to NKG2D-CAR NKAE cells. 

 

 

Figure 49. NKG2D-CAR CIML-NKAE cells significantly prolong mice survival in NSG-Tg(Hu-IL15) 

mice bearing U-266 ffLucGFP MM cells.  A) Imaging of tumor burden monitored by bioluminescence, 

at the indicated time points from NSG-Tg(Hu-IL15) mice bearing U-266 ffLucGFP cells without treatment 

or treated with NKG2D-CAR CIML-NKAE or NKAE cells. B) Kaplan-Meier survival curve of mice 

described in (A) (n=4). C) Quantification of luminescence per mouse at different time points. D) Flow 

cytometry of CD138+GFP+ tumor cells in BM mice at necropsy from the indicated experimental groups 

(n=4). The mean ± SEM is shown. E) Flow cytometry dot plots of residual tumor cells in the BM of the 

indicated mice in (A). Flow cytometry analysis of the number of U-266 ffLucGFP detected in the BM of 



RESULTS 

165 
 

mice #1 and #2 from NKG2D-CAR CIML-NKAE cell groups, sacrificed at day 154. One U-266 ffLucGFP 

untreated mouse is shown. *p<0.05; **p<0.01. 

 

In this experiment, both CAR NK cell therapy delayed tumor progression, but NKG2D-

CAR CIML-NKAE cells significantly increased mice survival (˃140 days) compared to 

the NKG2D-CAR NKAE cell treatment (93.5 days), and the vehicle group (71 days) (Fig. 

49A, B). Quantification of the luminescence photon flux signal was measured along the 

experiment (Fig. 49C) and tumor cells in the BM of all mice were analyzed by flow 

cytometry (Fig. 49D). Notably, 2 mice did not have clinical signs of disease neither BLI 

signal in NKG2D-CAR CIML-NKAE cells at day 150 and the experiment was 

discontinued. Residual MM cells in the BM of these mice by flow cytometry were 

detected: 60 PC on 1.07 x 106 viable events in mouse #1 and 88 PC on 1.15 x 106 viable 

events in mouse #2 (Fig. 49E). 

Altogether, these results support that cytokine memory induction of NKG2D-CAR 

NKAE cells potentiates anti-MM activity in vivo. 

 

4.3.10. NKG2A blockade enhances NKG2D-CAR CIML-NKAE cells 

cytotoxicity against MM cells in vitro 

Due to all the results above-described and, taking into account that NKG2D-CAR CIML-

NKAE cells also express high levels of NKG2A, we combined both approaches: memory 

induction through IL-12, IL-15, and IL-18 exposure with Z199 NKG2A blockade. 

In vitro cytotoxicity assays performed against XG-1, ARP-1, and RPMI-8226 MM cell 

lines demonstrated that the blockade of NKG2A is able to increase even more NKG2D-

CAR CIML-NKAE cell antitumor efficacy. At a ratio 4:1 (T:E) against RPMI-8226 cells, 

the specific lysis of NKG2D-CAR NKAE cells pretreated with isotype is 23.55 ± 4.16% 

and with NKG2A Ab rises to 49.91 ± 4.52%. Interestingly, isotype pretreated NKG2D-

CAR CIML-NKAE cells achieve 37.66 ± 4.28% of specific lysis, which increases up to 

62.43 ± 3.23% in when pretreated with α-NKG2A Ab (Fig. 50). 

Overall, although NKG2D-CAR CIML-NKAE cells produce a high antitumor response 

with respect to CAR NKAE cells, NKG2A blockade can potentiate their cytolytic activity.   



RESULTS 

166 
 

 

Figure 50. The blockade of NKG2A significantly augments NKG2D-CAR CIML-NKAE efficacy 

against MM cell lines. Specific lysis of NKG2D-CAR CIML-NKAE and NKAE cells treated with either 

the mouse α-NKG2A Ab (Z199) or its corresponding isotype (mIgG2b) against the MM cells ARP-1 (n=3), 

RPMI-8226 (n=4), and XG-1 (n=5) analyzed by 3h-Calcein release assay. The mean ± SEM is shown. 

Statistical analysis is included in the table. ns: not significant; *p<0.05; **p<0.01; ***p<0.001; 

****p<0.0001. 

 

 

 

 

 

 

 

 

 

 



 

167 
 

 

 

 

 

 

 

 

 

 

 

5. DISCUSSION 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

168 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



DISCUSSION 

169 
 

Although MM overall survival has increased over the last decades due to the therapeutic 

armamentarium available, MM is still an incurable disease. Within the TME, MM cells 

are sustained by soluble factors and direct cell-cell interactions that promote their 

proliferation, while exhausting NK and T cells, especially in the RRMM scenario. This 

immunosuppressive context arises the need to find new strategies to eradicate this tumor. 

In this setting, the main objective of this thesis is the generation of enhanced allogeneic 

CAR NK cell therapies for RRMM patients, in combination with immune checkpoint 

inhibitors to improve CAR NK cell efficacy, as well as with the induction of memory 

properties to increase CAR NK cell persistence. 

On the one hand, based on previous preclinical and clinical studies not only in other 

hematological malignancies, but also in solid tumors, immune checkpoints seem to play 

a key role in NK and T cell dysfunction, promoting tumor escape. In concrete, HLA-E, a 

non-classical class I molecule, has been found overexpressed in many tumors [348, 377, 

379-383, 399], correlating with lower overall survival [379, 384-386, 388, 389]. As the 

treatment with α-KIR and α-PD-1 or α-PD-L1 Abs to disrupt immune checkpoint 

inhibitory signals has reported severe toxicities in MM patients [356, 476], the interaction 

of HLA-E and NKG2A has become an interesting alternative to potentiate NK cell 

activity in this tumor. In this thesis, we have tested the combination of CAR NK cells 

redirected to two main targets in MM, NKG2D ligands and BCMA, with a novel human 

clinical-candidate α-NKG2A Ab, to overcome this shortcoming. 

On the other hand, in vivo persistence after adoptive transfer has been previously reported 

to correlate with clinical responses. NK cell short lifespan and limited proliferative 

capacity caused by the host immunosuppressive TME may be responsible for the low 

sustenance of CAR NK therapies. In this study, we have analyzed the in vivo efficacy and 

persistence of a new cytokine-induced memory-like NKG2D-CAR NKAE cell therapy 

for RRMM patients. 

 

 

 

 



DISCUSSION 

170 
 

5.1. HLA-E is overexpressed in pathologic plasma cells from 

RRMM patients compared to healthy populations 

In this first set of experiments, our results show that HLA-E expression is increased on 

PC from a large cohort of MM patients (n=54) compared to HD (Fig. 25A). Similar results 

were described by other groups that also studied HLA-E expression by flow cytometry 

on PC from 30 NDMM patients and 7 HD [125] and 8 MM patients and 1 PCL patient 

[378].  

To our knowledge, this is the first time that HLA-E expression on pPC has been compared 

to normal populations (nPC and CD34+ cells) from the same patients on MGUS, MM at 

diagnosis and MM at progression BM samples, resulting in an increased HLA-E 

expression in patients at progression, compared to the other stages. This fact suggests that 

HLA-E expression augments along MM progression and that its presence on the cell 

membrane is significantly high compared to healthy cells (Fig. 25B). Comparison with 

CD34+ cells is intuitive because a high HLA-E expression would predict a possible multi-

lineage cytopenia after HLA-E/NKG2A targeted therapies. Due to the improved clinical 

management of CRS, pancytopenia is currently the main severe adverse after CAR T cell 

treatments [477]. 

Since HLA-E expression is increased in patients at progression, we then compared PFS 

between MM patients at diagnosis with high (pPC/CD34+ ratio > 1.1) and low 

(pPC/CD34+ ratio ≤ 1.1) HLA-E expression. Our results show a shorter PFS tendency in 

the high HLA-E expression ratio, but a larger cohort of patients would be needed to 

propose HLA-E expression ratio as a prognostic factor. Unfortunately, a higher HLA-E 

expression ratio between pPC and CD34+ cells does not impact on patient overall survival 

(Fig. 25C). Since mediated responses by immune effectors in the BM may be fine-tuned 

via different signals beyond HLA-E/NKG2A axis, these results may suggest that other 

factors could be dominantly affecting MM survival. Although HLA-E may not result in a 

biomarker of the progression of MM, it could be conditioning adoptive therapy responses. 

Nevertheless, Lagana’s study reported a shorter PFS within the high HLA-E expression 

group after studying 436 NDMM patients, but this measurement was inferred from HLA-

A data from CoMMpass study, as HLA-E expression is not available in this dataset [387]. 

In line with the impact of HLA-E over prognostic markers, Yang’s group stratified the 



DISCUSSION 

171 
 

previously-mentioned 30 NDMM patients and obtained a significant correlation between 

HLA-E expression and an advanced ISS stage and a high cytogenetic risk [125]. 

Additionally, we have studied the relevance of several cytokines present in TME, not only 

to unveil possible cytokines involved in HLA-E modulation, but also their possible 

impact on CAR adoptive therapy efficacy. Previously, some groups have already reported 

different cytokine expression levels in this tumor compared to healthy samples. IL-10, 

MCP-1, IL-6, CXCL2, IL-2 and TNF-α, among others, have been found increased in 

samples from MM patients, while lower levels of IL-3, IL-13 and GM-CSF have been 

observed [478, 479]. However, IFN-γ levels in MM patients are controversial. Pérez-

Andrés et al. described a higher IFN-γ concentration (around 20 pg/ml) in BM samples 

from MM patients compared to HD [130], while Sharma’s and Zheng’s studies found 

lower levels of this cytokine in PB samples [478, 480]. In our project, we have analyzed 

the concentration of diverse cytokines in BM samples from 32 patients at different stages, 

obtaining higher levels of MCP-1, IL-6, in line with previous published results [478, 479], 

TGF-β, and IFN-γ (around 15 pg/ml), compared to HD (Fig. 26A). This result suggests, 

according to Pérez-Andrés et al., that IFN-γ levels are increased in the MM 

microenvironment, in the BM but not in the PB of these patients, where it may play a key 

role in the immune escape of this tumor. In this sense, although IFN-γ is essential in NK 

and T cell immunomodulation, this cytokine can also upregulate molecules, which 

mediates immune effector exhaustion (eg. PD-L1), in tumor cells [137]. 

In this context, we tested the influence of this proinflammatory cytokine on HLA-E 

expression. Previous studies had already reported that IFN-γ increased HLA-E expression 

on tumor cells [348, 377, 394, 395], including MM cells [349, 393]. IFN-γ 

transcriptionally augments HLA-E mRNA [481]. Moreover, HLA-E is stabilized on the 

cell surface when loading signal peptides derived from HLA molecules [88]. In the 

presence of IFN-γ or TNF-α, a dynamic equilibrium between the formation of the 

proteasome and the immunoproteasome occurs. Furthermore, BTZ is able to inhibit the 

β5 subunit of the proteasome along with the β5 induced subunit in the 

immunoproteasome, causing ER stress and a lack of loading peptides on HLA-E, 

reducing its presence on the cell membrane [482]. 

Our results also corroborated this effect over primary MM cells, from diagnosis and 

progression samples (Fig. 26B), and the MM cell lines RPMI-8226 and NCI-H929 (Fig. 

26C). Moreover, we included, in this set of experiments, two MM cell lines resistant to 



DISCUSSION 

172 
 

20 nM BTZ, to study the effect of this cytokine over BTZ-resistant MM cells. As we 

expected, IFN-γ also augments HLA-E presence on these cells, suggesting that BTZ-

resistant patients would also have increased HLA-E expression on PC after IFN-γ 

exposure. Additionally, BTZ treatment diminishes HLA-E expression on sensitive MM 

cells, even if its expression had been previously induced by IFN-γ. Remarkably, since 

HLA-E expression was not reduced after IFN-γ exposure on BTZ-resistant cells, HLA-E 

levels may remain high in MM BTZ-resistant patients with increased IFN-γ levels within 

the BM.  

 

5.1.1. α-NKG2A Ab monotherapy does not restore immune effector activity in 

BM MM patient samples 

Given the high expression of HLA-E detected in MM cells, we studied the expression of 

its activating receptor NKG2C and its inhibiting receptor NKG2A on cytotoxic NK and 

T cells. In our results, there is a higher percentage of NK cells expressing the inhibiting 

receptor NKG2A than the activating NKG2C in BM samples from MM patients, reaching 

50% of NKG2A+ NK cells in samples from patients at progression (Fig. 27A). On the 

contrary, cytotoxic tumor-infiltrating T cells expressed a very low percentage of NKG2A+ 

or NKG2C+ cells, while around 60% of these T cells were PD-1+(Fig. 27B).  These studies 

suggest that NK cell activity against tumors, but not T cells, may be impaired by the 

inhibitory signaling fostered by the binding of NKG2A to HLA-E in the tumor cells. Thus, 

the disruption of this axis could reactivate NK cell antitumor efficacy. 

Related to these results, around 60% of tumor-infiltrating NK cells have been found 

NKG2A+ in B-cell lymphoma bearing mice [399], although no significant differences 

were detected regarding this receptor between the NK cells from CLL patients and HD 

[348]. Moreover, low expression of NKG2A has been reported on tumor-infiltrating T 

cells from mice bearing B-cell lymphoma [399] and gynecological cancer samples at 

diagnosis [379], but these studies did not analyze the expression of NKG2A on NK cells. 

The importance of this inhibitory receptor in patient survival is unclear. High expression 

of NKG2A has been found to correlate with a poor prognosis in liver cancer [483] and 

HNSCC, while it has been associated with a failure to achieve remission in AML patients 

[103]. By contrast, a higher expression of NKG2A has been reported to correlate with a 

better survival in breast cancer and no associations were found in other solid tumors [484].  



DISCUSSION 

173 
 

Regarding α-NKG2A treatments in monotherapy, Monalizumab, a humanized α-NKG2A 

blocking Ab, has reported activity restoring immune effector activity in CLL and EBV+ 

lymphomas in vitro [348], but it has not shown efficacy in vivo in immunocompetent mice 

against B-cell lymphomas [399] and PDAC [383]. Due to commercial market issues and 

availability, our study has been focused on the use of another α-NKG2A Ab, the BMS-

986315, which specifically neutralizes the NKG2A receptor, compared to a standard 

mouse α-NKG2A Ab, the Z199, widely used on diverse studies. Battin et al. reported a 

strong binding of Z199 to NKG2A, as well as a high efficacy targeting HLA-E+ tumor 

cells that were loading different leader peptides [402]. Moreover, Z199 activity was 

compared to the use of PEBL, designed with the same scFv, where both showed a 

significant increase in NK cell cytotoxicity [394]. In keeping with this, the efficacy of the 

human α-NKG2A included in this study, provided by BMS (BMS-986315), is being tested 

in two clinical trials against NSCLC (NCT06094296) and solid tumors (NCT04349267), 

in combination with other drugs, but neither preclinical nor clinical efficacy data have 

been published yet. 

In this thesis we have compared, for the first time, to our knowledge, the efficacy of two 

α-NKG2A Abs (Z199 and BMS-986315) over BMMCs from MM patients. These cultures 

imply a native context, including PC as well as NK, T, DCs, among others, supplemented 

with own sample plasma, to test the effect of these Abs in the most similar conditions to 

the TME. Neither the use of the human nor the mouse α-NKG2A Ab were able to 

reactivate NK cell activity against MM cells (Fig. 27C). Probably, the challenging culture 

context would explain our results in this experiment. However, our findings are consistent 

with the negative outcomes reported in clinical trials, where Monalizumab only achieved 

modest responses in HNSCC [408] and gynecologic malignancies [407]. For this reason, 

most clinical trials are now combining Monalizumab with other blocking Abs and have 

reported better results [410, 411]. Having demonstrated that α-NKG2A Abs in 

monotherapy have not shown efficacy and, considering the reported toxicities with 

common mAbs (α-PD-1 and α-KIR) in MM patients, we hypothesized that by combining 

CAR NK adoptive therapy and α-NKG2A Abs, it might be possible to foster NK 

antitumor responses still further. 

 



DISCUSSION 

174 
 

5.2. NKG2D- and BCMA-CAR NKAE cells are efficiently 

generated 

CAR NK cells have emerged as a safer but equally [293], or even more effective therapy 

[155, 485], compared to CAR T cells in vitro. In the clinical setting, unlike CD19 CAR 

NK cells showed equal outcomes with regard to CAR T cells, BCMA-CAR NK therapies 

have reported modest and low durable responses (Table 4). This fact has promoted the 

development of new strategies to enhance CAR NK cell antitumor efficacy.  

CAR NK cell preclinical studies have used different specificities to target MM, as CS1, 

CD138, BCMA and CD19 [206, 330-333]. Our lab has been focused on two targets, 

BCMA and NKG2D-L, utilizing a BCMA-CAR based on a scFv that specifically binds 

BCMA and a NKG2D-CAR designed from the natural ectodomain of NKG2D receptor. 

BCMA has been described as the first MM target with limited expression on healthy cells. 

In fact, extraordinary outcomes of the two FDA-approved BCMA-CAR T therapies (ide-

cel and cilta-cel) have been reported. This target is virtually expressed on all PC, as we 

have demonstrated in different MM cell lines (Fig. 32A) as well as in primary MM cells 

(Table 13). Moreover, NKG2D-L have been also found overexpressed on MM cells, 

especially on the side population [155], which suggests the importance of targeting these 

ligands in this tumor. However, few clinical trials using NKG2D-CAR T cells have not 

met the expected results in MM patients [160, 266]. Probably, the poor efficacy may be 

caused by the absence of lymphodepletion prior to CAR T infusion, the use of first-

generation NKG2D-CAR or the low levels of NKG2D-L in those patients. However, we 

have detected a high expression of the NKG2D-L, not only on MM cell lines (Fig. 32A), 

but also on all the primary MM cells we have analyzed (Table 13). 

One of the main benefits of CAR NK therapy, regarding target-cell interaction, is that 

CAR NK cells are not restricted to HLA-TCR recognition. CAR NK cells can be used in 

an allogeneic context, without the need of engineering the TCR, as allogeneic CAR T 

cells, being a universal and off-the-self treatment option for cancer patients. Allogeneic 

CAR T cells have inferior clinical outcomes, probably due to genotoxicity, caused by 

gene-editing, and immunogenicity [486]. 

Expansion and transduction of CAR NK cells are quite challenging. CAR NK cells can 

be generated from different sources, although most preclinical studies use PB 



DISCUSSION 

175 
 

leukapheresis [286], PB buffy coats [335, 487] or umbilical CB [419]. Our study is based 

on obtaining a great amount of CAR NK cells from 18 ml of PB from HD, a less invasive 

and expensive method. However, most sources, including ours, need feeder cells (e.g 

K562-mb21-41BBL) to expand and activate NK cells [155, 488], resulting in a diverse 

NK cell clonal repertoire [489]. The use of K562-mb21-41BBL in co-culture with 

PBMCs generated a large amount of NKAE cells (up to 4 x 108 cells, 100-fold, data not 

shown), almost 100% CD56+, around 70% of them were cytotoxic (CD56+CD16+) and a 

very low contamination of CD3+ cells was observed (lower than 0.5%) (Fig. 28B). 

Interestingly, this feeder cell or the K562-mb15-41BBL have been used in numerous 

clinical trials, demonstrating a large NK cell expansion ratio. Besides, no toxicity or 

oncogenic risk were reported due to their manufacturing use, as these cells have to be 

previously irradiated [255, 490, 491]. 

Another key point is NK cell transduction. Lentiviral vectors are the most selected method 

for CAR delivery. These particles can infect dividing and non-dividing cells, while γ-

retrovirus only infect dividing cells and can produce a higher insertional oncogenesis risk 

compared to lentivirus, as they are normally inserted near transcriptional start sites, 

reasons why they have been discarded from many studies. Moreover, lentiviral 

transduction implies CAR integration in the DNA, between exons and introns, 

preferentially far from transcriptional start sites, and a long-lasting CAR expression on 

the cell membrane, compared to the transient expression obtained by mRNA or DNA 

transfection methods (3 or 15 days, respectively). Commonly, lentiviral vectors are 

pseudotyped with the VSV-G envelope. However, its receptor, the LDL-R, is poorly 

expressed on activated NK cells. For this reason, to favor the interaction between 

lentivectors and NK cells, several strategies as the use of cationic polymers (polybrene or 

protamine sulfate) to avoid virion-cell repulsion, crosslinking agents as Retronectin or 

Vectofusin-1, spinoculation or multiple rounds of transduction have been utilized to 

enhance CAR integration efficiency [492]. We have optimized NKG2D- and BCMA-

CAR transduction on our NKAE cells using retronectin-coated plates and spinoculation 

to enhance CAR integration. Moreover, due to differences in CAR expression between 

NKG2D- and BCMA-CAR NKAE cells, we have used a higher MOI (MOI 8) and 2 

rounds of transduction for BCMA-CAR. Our results showed an average of 34% and 22% 

of CAR efficiency expression in NKG2D- and BCMA-CAR NKAE cells, respectively, 

quite similar between populations, as it was also demonstrated after VCN measurement 



DISCUSSION 

176 
 

(0.82 vs 0.55 copies/cell) (Fig. 29B). Other groups have reported similar data in NK cells 

derived from PB. Published results from our group described a median of 20% NKG2D-

CAR+ NKAE cells, derived from PB from MM patients [155], while Herrera et al. 

achieved around 45% of CAR+ NK cells, after 7 days from transduction [325]. However, 

Muller et al. only reached a median of 10% CAR+ cells with lentiviral vectors [487], 

according to several studies that have reported the superior refractoriness of PB NK cells 

to engineering. Moreover, a higher sensitivity to apoptosis after lentiviral infection has 

been reported [493]. Nevertheless, NK cells derived from CB are more permissive [492], 

as described in clinical trials, achieving a median of 50% CAR+ NK cells [326]. Besides, 

unpublished results from our lab have showed a 67% and 53% of NKG2D- and BCMA-

CAR+ NKAE cells when generating these therapies from CB. 

To improve the transduction of NK cells derived from PB, some groups have proposed 

the use of another envelope, the Baboon envelope (BaEV), which receptors, ASCT-1 and 

ASCT-2 are highly expressed on activated NK cells [492, 494]. Dong et al. reported a 

median of 65% transduction efficiency when using BaEv lentiviral vectors [458], 

although our experience with the production of lentivectors containing the BaEv envelope 

resulted in 100-fold low titers, compared to VSV-G lentivectors. Due to the high lability 

of BaEv and that our transduction efficiency in NK cells was similar to VSV-G (around 

30%), we have not used this envelope in this thesis. Moreover, other intrinsic antiviral 

mechanisms may be hampering NK cell transduction, which can be inhibited by other 

molecules (eg. BX-795) [324]. 

Several publications reported an increased NKG2A and NKG2D expression on NK cells 

after ex vivo activation [155, 405, 495]. Tognarelli et al. demonstrated that, compared to 

resting NK cells, cytokine stimulation increases the percentage of NKG2A+ and NKG2D+ 

NK cells up to 80 and 100%, respectively [393]. According to these reports, we have 

observed a higher percentage of NKG2A+ and NKG2D+ CAR NKAE cells, 90 and 60%, 

respectively (Fig. 30A), suggesting that the activation of NK cells not only augments 

activating receptors such as NKG2D but also inhibiting receptors, as NKG2A, to 

counterbalance this excessive activation, which could produce an autoimmune effect. 

Moreover, we have studied the impact of different cytokines, used to generate NKAE 

cells or that are present in the MM microenvironment, on NKG2A expression. In 48-hour 

NK cell cultures, we showed a mainly IL-15-dependent NKG2A induction on cytotoxic 

NK cells (RFI of 100 with IL-15 compared to RFI of 55 on unstimulated NK cells). The 



DISCUSSION 

177 
 

same fold was obtained after IL-2 and IL-21 addition, cytokines commonly used to 

expand NK cells. However, TGF-β did not increase NKG2A expression (Fig. 30B). IL-

21, IL-15, IL-12, IL-10 and TGF-β have been described to induce NKG2A expression on 

NK cells [396]. Although we cannot properly compare NKG2A induction measure to 

other studies, Hromadnikova et al. reported an increase in the percentage of NKG2A+ 

cells up to 55 and 52% after IL-2 or IL-15 addition, respectively, compared to 30 and 45% 

without cytokines, after 4 days of culture [496]. Moreover, Skak et al. also described a 

synergistic effect of IL-2 and IL-21, raising NKG2A percentage from 41% without 

stimulation, to 59% with IL-2 and IL-21. These results indicate that all distinct cytokines 

that could be used to activate and expand NK cells would increase NKG2A expression, 

being an important inhibitory signal for NKAE cells when binding HLA-E on tumor cells.  

According to what we observed in primary MM cell analysis, the MM cell lines used in 

this project (U-266, RPMI-8226, XG-1, and ARP-1) also express different levels of HLA-

E, results that were previously reported by Sarkar et al. [378] (Fig. 31A). Although HLA-

E expression may seem low (RFI of 2.5-4.5), MM cells co-culture with NKG2D- or 

BCMA-CAR NKAE cells significantly increases the expression of HLA-E on MM cell 

lines (RFI of 10 and 12, respectively) (Fig. 31B). The high IFN-γ release by NKG2D- 

and BCMA-CAR NKAE cells (around 600 pg/ml) suggests the potential role of this 

cytokine in MM resistance to CAR NK therapy (Fig. 31C), which implies a positive 

feedback, as CAR NKAE cells produce more IFN-γ, when they are co-cultured with cell 

lines, increasing even more HLA-E expression on target cells. A similar study was 

performed on AML cells, where they observed an increase in HLA-E expression, induced 

by the IFN-γ secreted by immature NK cells [497]. In this way, tumor cells are protected 

from NK cell lysis via HLA-E/NKG2A signaling. 

 

5.3. The blockade of NKG2A enhances NKG2D- and 

BCMA-CAR NKAE efficacy against MM cells 

NKG2A, as well as KIRs, are involved in NK cell licensing. Therefore, a total elimination 

of NKG2A could induce hyporesponsiveness in NK cells. Zhang et al. suggested that 

partial deletion of NKG2A is more favorable than the complete blockade of this axis 

[498]. In this setting, different approaches have been carried out to disrupt NKG2A/HLA-



DISCUSSION 

178 
 

E binding. Kamiya et al. described the use of PEBL, molecules that concomitantly bind 

NKG2A and the ER to retain NKG2A receptors, impeding its presence on the cell 

membrane and therefore its binding to HLA-E [394]. However, two recent publications 

have knocked out by CRISPR-Cas9 the gene that codifies NKG2A, KLRC1, on NK cells, 

showing enhanced cytotoxicity against MM (80 vs 60% at 1:1 ratio against MM1.S cells 

and 40 vs 18% against U-266 pretreated with IFN-γ) [349] and solid tumors [495]. 

Nevertheless, most studies have been carried out using an α-NKG2A blocking Ab. 

Combination of ex vivo NK cells with α-NKG2A Abs enhances their antitumor activity 

against leukemia [404], CLL [348], PDAC [383], HNSCC [395] and MM [393]. This last 

study observed a significant increase in activated NK cell cytotoxicity in combination 

with the Z199 against different MM cell lines (75 vs 65% at 1:2 T:E ratio against U-266), 

even if they were previously incubated with IFN-γ to augment HLA-E expression (75 vs 

50% at 1:2 T:E ratio against U-266 pretreated with IFN-γ). Since CAR NK cells have not 

been studied in combination with complete or partial elimination of NKG2A against 

tumors, we cannot properly compare our results. We found that both mouse (Z199) and 

human (BMS-986315) α-NKG2A Abs significantly enhance NKG2D- and BCMA-CAR 

NKAE cells cytolytic activity against the MM cell lines U-266 (70 vs 50% at 1:1 ratio 

with NKG2D-CAR NKAE cells), RPMI-8226, XG-1 and ARP-1 (Fig. 32B) and primary 

cells from MM patients (Fig. 35B). Besides, we also corroborated the efficacy of these 

combinations against the RPMI-8226 cell line, after 24-hour exposure to IFN-γ (57 vs 

27% at 1:1 ratio with Z199 pretreated NKG2D-CAR NKAE cells) (Fig. 32D). 

Remarkably, experiments against primary cells were performed co-culturing both 

pretreated CAR NKAE cells with BMMCs. These BMMCs contained not only PC, but 

also immune cells, including immunosuppressive cells and CD34+ cells. Moreover, 

soluble factors (ADO, IDO, TGF-β) and platelets contained in the plasma of BM samples 

were added to sustain PC and test the efficacy of CAR NKAE cells in a challenging 

scenario. In this context, we have tested the efficacy of the α-NKG2A pretreatment on 

CAR NKAE cells mimicking TME, where both therapies have demonstrated great 

antitumor activity. Of note, although we did not observe differences in terms of efficacy 

between both CARs against MM cell lines, we used a lower T:E ratio for BCMA-CAR 

comparisons suggesting that BCMA-CAR has a higher efficacy eliminating PC (80 vs 

50% 1:5 T:E),  than NKG2D-CAR (80 vs 60% at 1:10 T:E) (Fig. 35B). Blockade of 

NKG2A on activated NK cells also enhanced cytotoxicity against AML primary cells, 

previously treated with IFN-γ (around 50 vs 30% at 1:1 or 1:2 T:E) [394] and knock out 



DISCUSSION 

179 
 

of NKG2A on activated NK cells resulted in an enhanced lysis of IFN-γ against primary 

MM cells (40 vs 20% at 1:1 T:E) [349]. 

Furthermore, we have measured the degranulation capacity of both CAR therapies against 

the MM cell lines previously mentioned and against the feeder cell K562-mb21-41BBL 

that lacks HLA-I and HLA-E expression. Again, both combinations demonstrated an 

increased degranulation capacity compared to isotypes except against the feeder cell, 

where, as expected, no differences were found due to the absence of HLA-E on the target 

cells (Fig. 33B). By contrast, Mahaweni et al. did not observe differences on 

degranulation levels after NKG2A blockade on NK cells against primary MM cells, only 

some efficacy was described when KIR mismatched samples were selected [406]. In 

addition, Oei et al. suggested that the presence of CAR molecules on NK cells overcomes 

the inhibition caused by the interaction between NKG2A and HLA-E, although they do 

not combine CAR NK cells with NKG2A blockade [499]. Our results demonstrate that 

even CAR NKAE cells have enhanced antitumor efficacy, the blockade of the inhibitory 

receptor NKG2A is able to increase even more this effect. 

Additionally, NKG2D- and BCMA-CAR NKAE cells secreted high concentrations of 

cytolytic cytokines. However, only IFN-γ concentration was significantly higher after 

CAR NKAE cells blockade with the α-NKG2A Abs, when co-cocultured with MM cell 

lines (Fig. 33C). Published results from our group regarding NKG2D-CAR NKAE cells, 

generated from MM patient PB samples, reported low levels of these cytokines alone (a 

median of 50 pg/ml of IFN-γ) [155] compared to the average of 600 pg/ml that we 

obtained. Notably, since cytokine concentration is augmented when exposed to MM cells, 

we cannot compare the release of other soluble molecules with the ones obtained in the 

cited publication [155]. Besides, Bexte et al. did not find significant differences in IFN-γ 

production between KLRC1 KO NK cells and non-treated NK cells after co-culture with 

tumor cells, but a higher tendency in the KO group was noticed [349]. Remarkably, our 

blocked cells neither expressed IL-6 nor IL-10 (data not shown), cytokines directly related 

to CRS. Although there is no direct correlation between cytokines produced in vitro and 

in patients, the absence of IL-6 and IL-10 production in vitro may suggest the low risk of 

this side effect when using CAR NK therapies in vivo [326]. 

This improved anti-MM activity could be expected to produce toxicities. On the one hand, 

most preclinical studies in MM have not studied CAR NK cell toxicity against healthy 

cells. Leivas et al. demonstrated low toxicity of NKG2D-CAR NKAE cells against a lung 



DISCUSSION 

180 
 

cell line (25%) and HD PBMCs (20%), but a considered lysis against a colon cell line 

(40%), at 1:4 T:E ratio [155]. However, these healthy cell lines may not have the same 

marker profile as healthy tissues, where NKG2D-L expression might be lower. Moreover, 

CD19 CAR NK cells used in clinical trials have shown transitory grade 4 hematologic 

toxic events (neutropenia and lymphopenia), but both trials have associated them with the 

lymphodepleting chemotherapy prior to CAR NK cell infusion (NCT03056339 [326] and 

NCT05020678). NKG2D-CAR NK therapy (NKX101) caused grade 3 or higher anemia 

and neutropenia in 50% of AML and MDS patients (3/6; NCT04623944), probably 

related to the lymphodepletion treatment. On the other hand, monalizumab, as a single 

agent, has not shown severe hematologic side effects in clinical trials. A phase II study 

treating HNSCC tumors detected grade 1 or 2 anemia in 7% of patients [408], while a 

gynecologic tumor clinical trial reported grade 3 or higher anemia in 2 out of 40 patients 

[407]. In this thesis, we have tested NKG2D- and BCMA-CAR NKAE cells toxicity when 

pretreated with the α-NKG2A Abs against HD PBMCs (Fig. 35C) and CD34+ from MM 

patients in a native-context culture (Fig. 35B). Toxicity against PBMCs was quite low, 

less than 20%, using the α-NKG2A NKG2D-CAR treatment, as Leivas et al. reported 

[155], and less than 5% of killing with the BCMA-CAR combined therapy, at the highest 

ratio (1:4 T:E). Moreover, we chose to test toxicity against hematopoietic progenitors 

because one of the most severe CAR T cell side effects is prolonged cytopenias, 

associated with a higher risk of infections [500]. Notably, the combination of α-NKG2A 

Abs with NKG2D- or BCMA-CAR NKAE cells displayed low cytolytic activity against 

MM patient CD34+ cells, suggesting that they concomitantly lyse primary PC without 

affecting these hematopoietic stem cells, in line with most CAR NK cells and 

monalizumab clinical trial results, separately. Remarkably, the use of the BMS-986315 

Ab did not increase the toxicity caused by CAR NKAE cells incubated with the isotype 

(Fig. 35B). 

Unfortunately, we have not found differential expression on diverse expression markers 

studied by flow cytometry and CyTOF (Fig. 34A, B), as other groups have already 

reported when disrupting HLA-E/NKG2A axis [349, 394], except from a downregulation 

on NKG2A expression. By contrast, as a paradigmatic example, Mac Donald et al. 

reported a higher expression of NKG2C on KLRC1KO NK cells, which partially 

participates in the augmented cytotoxicity of the KO NK cells, and suggested that 

NKG2C increase is caused by the NK cell expansion using IL-21. However, we activated 



DISCUSSION 

181 
 

and expanded NK cells with the feeder cell K562-mb21-41BBL that contains IL-21 and 

did not observe an increased expression of NKG2C during NK cell expansion or when 

treated with the α-NKG2A Ab. 

Although differences in cytotoxicity are appreciated between CAR NKAE cells 

pretreated with the α-NKG2A Ab or the isotype and other markers may be affecting the 

activating/inhibiting NK balance, we have found an increased phosphorylation of 

downstream ZAP70/Syk kinases after blocking with BMS Ab (Fig. 34C), which could 

explain the higher efficacy of this combination per se. Further phosphoproteomic studies 

of other kinases are needed to unveil the complete signaling pattern. 

 

5.3.1. Pretreatment of NKG2D-CAR NKAE cells with BMS-986315 α-NKG2A 

Ab significantly reduces tumor burden in vivo 

First in vivo attempt in immunodeficient NSG-Tg(Hu-IL15) mice bearing RPMI-8226 

ffLucGFP cells comprised NKG2D- or BCMA-CAR NKAE cells pretreated with either 

the human BMS-986315 or the hIgG1κ and additional infusions of the corresponding Abs 

to maintain the blockade of NKG2A over time. PB analysis at day 7 and day 14 revealed 

a significant NK cell percentage decrease within α-NKG2A pretreated CAR NK cells, 

which may be caused by the prolonged NKG2A absence, fostering CAR NKAE cell 

recognition of their siblings and inducing fratricide (Fig. 36). Of note, this effect is not 

caused by an enhanced ADCC, as Fc region from BMS-986315 Ab is mutated [414]. 

Ruggeri et al. also described this effect after repeated doses of α-NKG2A Ab in mice, 

suggesting NK cell auto-killing [404]. However, this fratricide effect has not been 

reported in vitro. For example, Bexte et al. knocked-out the NKG2A encoded gene, 

KLRC1, and demonstrated that the lack of NKG2A did not impact on NK cell viability. 

These cells even had a 10-fold expansion after 18 days from CRIPSR-Cas9 edition [349]. 

Since NKG2A blocked CAR NKAE cell persistence was reduced, we tested the efficacy 

of the blockade of NKG2A on both therapies, NKG2D- and BCMA-CAR NKAE cells, 

only as a pretreatment, against the modified U-266 ffLucGFP cell line in the disseminated 

orthotopic NSG-Tg(Hu-IL15) mouse model. As we and others have reported [378], 

although HLA-E expression is quite low in this cell line, it increases in vivo (5 vs 15 RFI) 

(Fig. 38C), becoming an appropriate model to assess our treatments. 



DISCUSSION 

182 
 

CAR NK cells have shown efficacy in vivo against MM cells. Regarding NKG2D-CAR, 

our group has already published that NKG2D-CAR NKAE cells efficiently eliminate 

tumor burden in 25% of mice against the same cell line, U-266 ffLucGFP, although there 

are some differences between that study and ours. Leivas et al. generated NKG2D-CAR 

NKAE cells from MM patients PB, not HD samples, and for the in vivo experiment they 

infused this therapy at a 1:3 T:E ratio in NSG mice [155]. Our first in vivo experiment 

against U-266 ffLucGFP cells was designed at a 1:16 T:E ratio. Under these 

circumstances, even mice from the isotype control treatment were rescued from dead 

(50%), and in the α-NKG2A blocking therapy, 75% were MRD- (Fig. 38). Using a low 

T:E ratio, 1:8, we significantly delayed tumor development in the α-NKG2A NKG2D-

CAR NKAE cells treatment (>143 days) compared to the isotype (93.5 days) (Fig. 40), 

suggesting that the disruption of the HLA-E/NKG2A axis enhances the antitumor efficacy 

of CAR NK cells. 

Other studies have also reported the benefit of hampering the binding of NKG2A to HLA-

E in NK cells, out of CAR context. Ruggeri et al. rescued from death 100% of 

immunodeficient mice bearing AML, pretreating activated NK cells with monalizumab 

α-NKG2A Ab [404]. Similarly, α-NKG2A PEBL-transduced NK cells were able to 

eliminate Ewing sarcoma cells in 5 out of 7 immunodeficient mice, after 269 days of 

follow-up [394]. In addition, the CRISPR-Cas9 KO of the KLRC1 gene on NK cells 

delayed breast cancer proliferation (59 vs 48 days) in NSG-Tg(Hu-IL15) mice [495]. 

Unfortunately, in our two U-266 ffLucGFP in vivo experiments (1:16 and 1:8 T:E), we 

could not evaluate differences between α-NKG2A blocked BCMA-CAR NKAE cells and 

the isotype (Fig. 38 and 40). These results indicate that BCMA-CAR has more efficacy 

than NKG2D-CAR in eradicating MM cells, as we also detected against primary cells. 

New experiments with low BCMA-CAR NKAE cell doses are ongoing. Previous studies 

have also reported high efficacy of this BCMA-CAR on NK cells in MM mouse models. 

Ng et al. engineered PB NK cells to co-express BCMA-CAR and CXCR4 to improve 

CAR NK cells homing to the BM, obtaining a higher mice survival over the BCMA-CAR 

NK cells group (1:6 T:E ratio) [335]. In line with BCMA-CAR NK cell modifications, 

Cichocki et al. modified iPSC-derived NK cells to express a BCMA-CAR, mbIL-15 and 

IL-15R, CD38 KO as well as a non-cleavable CD16, to augment ADCC in combination 

with daratumumab, and showed their efficacy at an extremely high ratio (1:150 T:E ratio) 

against the MM cell MM1.S [340]. Our experience suggests that redirecting BCMA-CAR 



DISCUSSION 

183 
 

NKAE cells with chemokines receptors or daratumumab to increase their potential are 

not needed to fully eradicate tumor burden in our MM bearing mouse models. 

The enhanced NKG2D- and BCMA-CAR NKAE cell antitumor efficacy is probably 

caused by the long persistence of CAR NKAE cells in mice. We were able to detect NK 

cells in all mice treated with CAR NKAE therapies up to 29 days and, in both U-266 

ffLucGFP experiments, CAR expression remained stable (Fig. 37 and 39). Of note, no 

significant differences were obtained regarding NK cell percentage in mice between 

CAR-NKAE cells pretreated with the α-NKG2A Ab and the isotype. These results suggest 

that the release of human IL-15 in these mice plays a key role in NK cell persistence. 

Moreover, the expression levels of transgenic human IL-15 expressed by NSG-Tg(Hu-

IL15) mice are similar to the levels that can be found in humans (around 7.1 pg/ml) [469], 

suggesting that our model resembles the IL-15-milieu that these therapies would find in 

patients. Several studies have recommended the addition of interleukins, such as IL-2 or 

IL-15, to promote NK cell sustenance in mice [501]. Another strategy is the incorporation 

of the human IL-15 and the CAR sequence in a bicistronic construct, as a soluble factor 

or as a membrane ligand [336, 340, 419]. In particular, Liu et al. obtained a longer mice 

survival when CB CAR NK cells auto-produced IL-15, with a bicistronic construct, up to 

68 days, compared to CAR NK cells lacking IL-15, and demonstrated that exogenous IL-

15 results more toxic than the production of this cytokine by engineering CAR NK cells. 

By contrast, Thangaraj et al. detected NK cells at necropsy (63-91 days) in all NK 

treatments in the mouse BM from NSG-bearing MM, suggesting that no exogenous or 

IL-15 self-production on NK cells is needed to maintain these cells for extended periods 

[390]. Seemly in contrast, none of our mice had NK cells at necropsy (later than 90 days). 

Altogether, the combination of the human α-NKG2A with NKG2D-CAR NKAE cells 

displays high efficacy in vivo and could be a novel therapy for RRMM treatment. 

 

 

 



DISCUSSION 

184 
 

5.4. NKG2D-CAR CIML-NKAE cells exhibit a higher 

efficacy than NKG2D-CAR NKAE cells in vitro and in 

vivo 

Previous exposures to haptens, virus or even tumor cells have been reported to induce a 

higher proliferation in NK cells, especially after a second exposure to the stimulus [423, 

424]. Interestingly, Fehniger et al. tested the effect of different interleukins (IL-12, IL-15 

and IL-18) on NK cell functions [430] and, in 2009, Cooper et al. combined these three 

molecules and first described the induction of memory-like properties on NK cells 

through cytokine exposure, the CIML-NK cells [431]. Thus, the discovery of memory 

properties in NK cells has resulted in a multitude of studies trying to unveil the 

mechanisms underlying this phenomenon.  

Most studies obtained NK cells from PB leukapheresis or buffy coats and cultured them 

overnight with IL-12, IL-15, and IL-18. Conventional NK cells were maintained with IL-

15 only as a control [437, 442], although this population is obsolete, and is no longer use 

in modern clinical trials. Moreover, CAR transduction was normally performed the day 

after cytokine exposure [454, 456, 458]. Our study wanted to combine the benefits of 

generating CAR NKAE cells from a small amount of PB and the co-culture with the 

feeder cell K562-mb21-41BBL with the induction of memory properties. Moreover, we 

have compared the CAR CIML population with our best-developed therapy, the CAR 

NKAE cell, not the conventional IL-15-activated NK cells. Only a recent study has 

generated CAR CIML-NK cells using a similar strategy. He et al. first purified NK cells 

and exposed them to IL-12, IL-15, and IL-18 for 16 hours. Then, memory-like NK cells 

were co-cultured with the feeder cell to expand and activate these NK cells and after 4 

days, they were retrovirally transduced with a CD19-CAR [457]. In this thesis, CAR 

NKAE cells were generated as previously described. First, PBMCs were co-cultured with 

the feeder cell to increase NK cell number and activation and after NK cell purification, 

NKAE cells were lentivirally transduced with a NKG2D-CAR. The following day, CAR 

NKAE cells were exposed to IL-12, IL-15, and IL-18. This way, since memory is induced 

after CAR NKAE cells are generated, NK cell reprogramming takes place after. However, 

He et al. interfere CIML reprogramming with feeder cell expansion, which could modify 

memory-like proper characteristics.  



DISCUSSION 

185 
 

NKG2D-CAR CIML-NKAE or NKAE cells were efficiently generated from HD PB. At 

day 14, after activation, purification, and transduction, both populations were almost 

100% CD56+, most of them CD56+CD16+, with less than 1% of CD3+ cells (Fig. 41B). 

Although a lower expression of CD16+ has been reported after IL-12 and IL-18 addition 

[502], we did not observe significant differences between groups. Moreover, NKG2D-

CAR CIML-NKAE cells significantly augmented NKG2A expression compared to day 

0, while NKG2C and PD-1 expression remained unchanged (Fig. 41G), suggesting that 

NKG2A is still a critical checkpoint after memory induction, as Berrien-Elliott et al. also 

described in conventional CIML-NK cells  [445]. 

NKG2D-CAR transduction, using lentivectors with VSV-G envelope, was efficient and 

stable after 6 days, as we demonstrated by flow cytometry (around 20% of NKG2D-CAR 

CIML-NKAE cells were CAR+) and q-PCR (Fig. 41D). Remarkably, there was a high 

correlation (R2=0.93) between the percentage of NKG2D-CAR+ NK cells and the VCN, 

indicating that the integration of the CAR construct correlates with its expression on the 

cell membrane. Besides, we could quantify the number of NKG2D-CAR molecules per 

cell (around 1,000 molecules) (Fig. 41F). Although there were no significant differences 

in NK cell transduction between the two populations, a higher tendency was observed in 

the NKG2D-CAR CIML-NKAE cells. Our results are in line with other studies that used 

lentivectors and the VSV-G envelope. They detected around 15-20% of CAR+ NK cells 

and also suggested a higher transduction efficacy in the CIML population [454, 456]. 

However, it has been reported that the use of BaEv-lentiviral vectors [458] or RD114 

feline envelope-gammaretroviral vectors [457] could achieve a higher percentage of 

CAR+ CIML-NK cells, 65% and 70%, respectively. Although transduction efficacy using 

the VSV-G envelope is lower, our NKG2D-CAR CIML-NKAE cells have demonstrated 

a relevant anti-MM activity. 

NKG2D-CAR CIML-NKAE cells exhibited a significantly higher cytotoxicity against 

MM cell lines (Fig. 42A) and notably, against primary MM cells (Fig. 47A). This last 

experiment was performed maintaining TME conditions, where we also demonstrated the 

superiority of the CIML population. Besides, we detected an increased efficacy against 

the clonogenic MM cell line L-363 (40% survival at 2:1 T:E ratio) (Fig. 42B), indicating 

that NKG2D-CAR CIML-NKAE cells are able to efficiently kill clonogenic tumor cells, 

also described by Leivas et al., using NKG2D-CAR NKAE cells (70% of colony survival 

at 1:1 ratio) [155]. Moreover, we observed that NKG2D-CAR CIML-NKAE cells 



DISCUSSION 

186 
 

degranulate more (Fig. 42C) and have a higher percentage of IFN-γ+ NK cells compared 

to the control population (Fig. 43A), NKG2D-CAR NKAE cells, alone or during co-

culture with MM cell lines. These results are in line with several studies where CIML-

NK cells have demonstrated a higher cytolytic capacity against other hematological 

malignancies as ALL [437, 447] and AML [437, 442, 445], compared to conventional NK 

cells. Regarding CAR CIML-NK cells, these cells also displayed a higher cytotoxicity, 

degranulation and IFN-γ production against B-cell malignancies [457], lymphomas [456] 

and HNSCC [454], with regard to conventional CAR NK cells. These studies also 

compared CAR CIML-NK cell efficacy with untransduced CIML-NK cells, as their aim 

was demonstrating CAR superiority too. In this thesis, since we aimed to improve the 

functionality of our best effector, the NKG2D-CAR NKAE cells, the differences that we 

studied are due to the induction of the memory properties, not CAR transduction. 

Moreover, we have also studied the release of cytolytic proteins. NKG2D-CAR CIML-

NKAE cells secreted more soluble FasL, IFN-γ, granzyme A and perforin compared to 

control NKG2D-CAR NKAE cells (200, 3,000, 5,000 and 1,200 pg/ml secreted by 

NKG2D-CAR CIML-NKAE cells against U-266, respectively) (Fig. 43B). Most CIML-

NK cells studies measured the percentage of IFN-γ+ cells within NK cells instead of 

analyzing the levels of cytolytic proteins that they release. However, two publications 

reported a higher concentration of IFN-γ regarding CIML-NK cells, compared to 

conventional IL-15-activated NK cells. Zhuang et al. showed a median concentration of 

1,000 pg/ml of IFN-γ in supernatants from CIML-NK cells in co-culture with liver cancer 

cell lines [448]. In line with this, Lusty et al. studied the release of IFN-γ at different time 

points and interleukin combinations to reprogram NK cells. They detected the maximum 

IFN-γ concentration 24 hours after interleukins exposure, obtaining similar and the 

highest levels in IL-12 plus IL-18 condition (around 80,000 pg/ml) and the CIML 

combination, IL-12, IL-15 and IL-18 (around 60,000 pg/ml). Then, IFN-γ concentration 

decreased with time (60 pg/ml after 72 hours) [435]. Nevertheless, we found high levels 

of this cytokine 6 days after memory induction (800 pg/ml without target and up to 5,000 

after co-culture with K562-mb21-41BBL). Huber et al. detected a higher percentage of 

perforin+ NK cells [438], as well as Romee et al, who also described an increased 

production of granzyme B on CIML-NK cells [442]. These results correlate with our 

findings regarding perforin, but we did not observe significant differences in granzyme B 

release. Moreover, NKG2D-CAR CIML-NKAE cells did not release CRS-associated 



DISCUSSION 

187 
 

cytokines as IL-6 and IL-10 (data not shown). Of note, a phase I trial with ML NK cells 

did not report CRS in AML patients [418].  

Not only NKG2D-CAR CIML-NKAE cells displayed a great efficacy against MM cells 

but also, they did not affect CD34+ cells derived from CB (Fig. 47B) or from MM patients 

(Fig. 47A). We demonstrated that the CAR CIML population increases its cytotoxic 

capacity against PC from MM patients without compromising the CD34+ cells present in 

those samples, suggesting the absence of hematotoxicity of this therapy regarding 

hematopoietic progenitors. In this sense, no hematological adverse effects were described 

in the CIML-NK cell NCT02782546 clinical trial, where most patients achieved 

neutrophil and platelet engraftment within the first month of treatment [417]. In line with 

this, NCT03068819 study reported hematological grade 3-4 side effects in only 1 out of 

9 patients treated [418]. However, Saphiro et al. observed pancytopenia in 4 out of 6 

patients with myeloid malignancies, 2 of them had prolonged pancytopenias and CD34+ 

cells from the same patients were infused, recovering neutrophils within the first 14 days 

[459]. Point-of-care all these trials infused CIML-NK cells after hematopoietic cell 

transplant, suggesting the hematological toxicities reported may be a consequence of the 

allogeneic transplant and not of the CIML therapy. 

CIML-NK cells have been also reported to have an increased metabolic activity. Our 

results showed that NKG2D-CAR CIML-NKAE cells have higher glycolysis compared 

to NKG2D-CAR NKAE cells, although no differences were observed in OXPHOS 

analysis (Fig. 44C, D). These results correlate with Terren et al., where CIML-NK cells 

had a higher ECAR tendency without significant differences in the OCR test [453]. 

However, other studies suggested that CIML-NK cells not only have a higher glycolytic 

capacity but also an augmented OXPHOS [451, 452]. Regarding metabolic receptors, 

Terren et al. reported a higher expression of nutrient transporters such as the glucose 

transporters GLUT1 and GLUT3, the transferrin receptor CD71 and the amino acid 

transporter CD98 [453]. Although we corroborated the higher CD71 expression on our 

NKG2D-CAR CIML-NKAE population, we observed a significantly lower GLUT1 

expression, compared to NKG2D-CAR NKAE cells (Fig. 46). 

Different results were also obtained compared to other studies regarding diverse surface 

markers, except from CD25 (Fig. 46). CD25, the γ-chain of the IL-2 receptor, was 

significantly increased on NKG2D-CAR CIML-NKAE cells as well as in most CIML 

studies [434, 436, 438, 439, 441, 451, 453, 503], suggesting that the high expression of 



DISCUSSION 

188 
 

the γ-chain IL-2 receptor may be involved in the higher proliferation of CIML-NK cells 

in response to low doses of IL-2. In addition, our results indicated a lower expression of 

NCRs, NKp30, NKp44 and NKp46, as well as of the maturation antigen CD57, while a 

significantly higher expression of CD2 was observed (Fig. 46). However, previously 

published studies reported other differences in markers expression, in comparison to 

conventional NK cells. CD69, NKG2A, CD56 [437], Sca-1 [438] and SEMA7A [440] 

are upregulated in CIML-NK cells, while a decreased expression of CD62L, CD16 [438], 

CXCR4 [439], TGFβR and KIRs [443] has been reported. 

Regarding RNA-seq analysis, we detected 162 DEG between NKG2D-CAR CIML-

NKAE and NKAE cells (Fig. 45A), indicating that the induction of memory through the 

exposure of IL-12, IL-15, and IL-18 is able to modify gene expression program. It is 

worth noting that the transferrin receptor, CD71, not only was overexpressed on NKG2D-

CAR CIML-NKAE in the RNA-seq study, but also we could corroborate it by flow 

cytometry. Since CD71 is also a target of cMyc, this receptor is involved in NK cell 

proliferation [504] and could be related to the higher Ki67 expression and the augmented 

percentage of S-G2-M+ NK cells we found within the NKG2D-CAR CIML-NKAE 

population. By the same token, the increased transcription of the CCND2, RANK and 

MANF genes, mainly related to cell growth [471-473], may be involved in this 

incremented proliferation. However, some studies have discovered that the expression of 

RANK on NK cells diminishes their antitumor efficacy when binding to RANKL but, 

remarkably, the treatment with denosumab, an antibody that targets RANKL, counteract 

this effect [505, 506]. Therefore, in MM, a concomitant use of denosumab with CAR 

CIML-NKAE cells not only would decrease osteoclast activity, but also may augment 

this NK therapy efficacy. Another transcription factor involved in cell growth is 

CABLES1. Our RNA-seq data indicate a lower transcription of this gene in the NKG2D-

CAR CIML-NKAE cells, which decreases CDK2 kinase activity and negatively impact 

on cell proliferation and/or differentiation, also in line with our Ki67 and cell cycle 

analysis [474, 507]. Moreover, relevant inhibitory receptors such as TIGIT [303] and 

KLRG1 were found underexpressed on the memory population, suggesting a lower 

sensitivity to be inhibited. Since KLRG1 has been also described as a marker of 

differentiation or senescence [308], our NKG2D-CAR CIML-NKAE cells seem to 

display a less mature and senescent phenotype. In addition, NKG2D-CAR CIML-NKAE 

cells have an increased expression of chemokines receptors (CCR4 and CCR7), which 



DISCUSSION 

189 
 

may enhance their ability to respond to chemokines and increase their homing to tumor 

sites [508, 509]. Of note, NKG2D-CAR CIML-NKAE cells were enriched for pathways 

involved in migration, regulation of IL-12 production and chemotaxis. Furthermore, a 

high expression of CCR7 has been also associated with a less mature NK cell subset 

(CD56brightCD16-) [510], matching with the lower KLRG1 transcription. Altogether, these 

modifications might be also responsible for the higher anti-MM activity of the NKG2D-

CAR CIML-NKAE cells. Although this is the first study comparing RNA expression in 

CAR CIML-NKAE cells and CAR NKAE cells, other groups have also reported 

differences regarding CIML-NK cell populations versus conventional NK cells. Kerbauy 

et al. found an enrichment of genes involved in IFN-γ, TNF, IL-2/STAT5 and IL-

6/JAK/STAT3 signaling and mTOR signaling [455], while Berrien-Elliott et al. found a 

higher enrichment in GATA3 target genes [445]. Of note, Becker-Hapak identified an 

increased induction of the IFNG gene [451]. Despite we have not identified this gene 

overexpressed in our NKG2D-CAR CIML-NKAE cells, their intracellular IFN-γ 

concentration (Fig. 42C), as well as their IFN-γ release (Fig. 43B), are significantly high 

compared to NKG2D-CAR NKAE cells. 

Furthermore, we have observed 115 CpG sites differentially methylated between 

NKG2D-CAR CIML-NKAE cells and NKG2D-CAR NKAE cells, related to cytokine 

signaling, cytoskeletal remodeling and cell-extracellular matrix interactions (Fig. 45B). 

Most common demethylated gene observed between CIML-NK cells and conventional 

NK cells is the CNS-1 region of the IFN-γ locus [451]. Differences in RNA-seq and 

epigenomic studies compared to other published reports in CIML-NK cells may probably 

lie in the different control populations. 

The most relevant induced memory property in CIML-NK cells is persistence in vivo. 

Several studies have reported that CIML-NK cells have a higher proliferation compared 

to control NK cells [432, 439, 441]. Our results correlate with these published outcomes 

in vitro. NKG2D-CAR CIML-NKAE cells had a higher percentage of Ki67+ cells as well 

as a higher percentage of cells entering mitosis (S-G2-M+), 6 days after memory 

reprogramming (Fig. 44A, B). Remarkably, NKG2D-CAR CIML-NKAE cells had a 

significantly higher percentage of NK cells at day 7 in mice, and CAR expression was 

superior at day 14 and 29 (Fig. 48C). Thus, in vivo results corroborate our in vitro results, 

showing that the exposure of NKG2D-CAR NKAE cells to the interleukins IL-12, IL-15 

and IL-18 enhances the percentage of NKG2D-CAR CIML-NKAE cells in mouse PB, 



DISCUSSION 

190 
 

probably due to a higher proliferation in vivo. Moreover, other studies have reported the 

longer persistence or proliferation of CIML-NK cells in mice models, compared to 

conventional NK cells [434, 436, 437, 441]. Notably, He et al. also demonstrated a higher 

relative percentage of CAR CIML-NK cells (0.5% of NK cells) in mouse PB, compared 

to CAR NK cells (nearly 0%) at day 7 [457]. Of note, our NKG2D-CAR CIML-NKAE 

cells represented a median of 18% at day 7, much superior to what this study obtained. 

Nevertheless, we have not observed, at any kinetic time point, the presence of NKG2D-

CAR CIML-NKAE cells, once NKG2D-CAR NKAE cells cannot be detected. Thus, we 

cannot assure a higher persistence of the CIML population. 

The superior percentage of NKG2D-CAR CIML-NKAE cells at day 7 and 14 may be one 

of the main responsible for the higher antitumor responses in vivo. NKG2D-CAR CIML-

NKAE cells rescued 50% of mice from death, using a 1:8 T:E ratio (Fig. 49). Notably, 

these mice were MRD- at necropsy after 154 days from U-266 ffLucGFP infusion. This 

result correlates with another CAR CIML-NK cell publication. He et al. also observed a 

higher in vivo efficacy in the CAR CIML-NK cell group against a lymphoma cell line, 

compared to conventional CAR NK cells [457]. Other studies support the superior in vivo 

activity of CIML-NK cells compared to conventional NK cells, without CAR [441, 442, 

448-451, 511]. 

CIML-NK cell efficacy could be increased by different strategies [512]. Although our 

NKG2D-CAR CIML-NKAE cells have demonstrated great anti-MM activity in vivo, 

after enhancing their potential including genetic modifications (CAR transduction) and 

improving their expansion (co-culture with K562-mb21-41BBL), we tested the effect of 

inhibiting the NKG2A receptor with a neutralizing α-NKG2A Ab. Berrien-Elliott et al. 

reported that NKG2A was transcriptionally induced in CIML-NK cells and demonstrated 

that the KO of the NKG2A-encoding gene KLRC1 increases their anti-AML responses 

[445]. Taking into account our results blocking the HLA-E/NKG2A axis in NKG2D- and 

BCMA-CAR NKAE cells and the high expression of the NKG2A inhibitory receptor on 

NKG2D-CAR CIML-NKAE cells, we also pretreated these cells with an α-NKG2A Ab. 

Analysis against three different MM cell lines confirmed the synergistic effect of the 

induction of the memory phenotype and the blockade of NKG2A eradicating MM cells 

in vitro (Fig. 50). Subsequent in vivo studies are ongoing to fully corroborate this 

enhanced anti-MM efficacy. 



DISCUSSION 

191 
 

This study has certain limitations. Although the blockade of NKG2A on our CAR NKAE 

populations has exhibited an enhanced efficacy against MM cells, we have not studied 

the HLA-E peptidome. Depending on the nature or the sequence of the peptide, the 

stability of HLA-E on the cell membrane and its binding to NKG2A can be modified, 

modulating NK cell response (eg. Hsp60 or CMV peptides) [513, 514]. Battin et al. 

recently reported that the blockade of this inhibitory receptor also depends on the peptides 

presented by HLA-E. In the mentioned study, they demonstrated that monalizumab was 

more effective when HLA-E loaded HLA-A, HLA-B and HLA-C peptides compared to 

HLA-G peptides presentation [402]. Further analysis would be needed to specifically 

confirm the absence of efficacy of BMS-986315 Ab in monotherapy against BMMCs 

from MM patients cultured in a native context. In line with this, we have studied the in 

vivo activity of NKG2D- and BCMA-CAR NKAE cells in combination with BMS-

986315 Ab in NSG mice that constitutively express human IL-15. However, a humanized 

mouse model engrafting not only primary MM cells, but also a complete immune system 

from the same patient, to study the efficacy of these combinations in the human context 

would have reported more accurate results. The difficulty engrafting primary MM cells 

as well as a complete TME [515] reflect the lack of PDX-His MM mouse models to 

properly evaluate treatment efficacy and safety. Regarding our in vivo studies, we have 

also found other limitations. The first study was conducted by combining NKG2D- and 

BCMA-CAR NKAE cells and BMS-986315 Ab, in pretreatment and intraperitoneally 

infused in NSG-Tg(Hu-IL15) mice bearing RPMI-8226 ffLucGFP MM cells. The lower 

percentage of CAR NKAE cells in BMS-986315 groups compared to the isotypes 

suggested a fratricide effect that we did not detect in the subsequent studies, after 

eliminating the intraperitoneally Ab administration. However, the pretreatment with the 

human α-NKG2A only blocked NKG2A expression on CAR NKAE cells for 48 hours. 

In this regard, other strategies to permanently abolish or reduce NKG2A expression on 

CAR NKAE cells, such as siRNA or PEBL, will be further tested. In our lab, the knock 

out of the KLRC1 gene, which encodes NKG2A, is being studied in BCMA-CAR NKAE 

cells and preliminary results have not reported a fratricide effect after eliminating NKG2A 

expression in vitro. CRISPR-Cas9 technology can completely and permanently eliminate 

NKG2A expression compared to Ab blockade, but this technique has a higher cost in 

GMP-setting because nucleofection procedures are needed. Furthermore, the total 

elimination of this receptor could affect NK cell licensing, suggesting that a partial 

deletion could have better results [103]. In addition to this, although the objectives of this 



DISCUSSION 

192 
 

thesis are focused on the clinical translation of enhanced CAR NKAE effectors, we have 

slightly investigated the molecular mechanisms underlying NKG2A signaling after BMS-

986315 pretreatment. In this context, the study of the phosphoproteome of other proteins 

involved in this balance signaling, such as VAV-1, SHP-2 and FYN, or proteins that 

participate in CAR NKAE cell synapsis, would arise a better knowledge about the effect 

of NKG2A blockade on CAR NKAE cells.  

Regarding NKG2D-CAR CIML-NKAE cells, massive analysis and flow cytometry 

studies revealed a multitude of DEG, methylations and differentially expressed markers 

on the cell surface, but unfortunately, we have not identified a set of candidate genes, 

overlapping between our NKG2D-CAR CIML-NKAE cells and conventional CIML-NK 

cells compared to NKG2D-CAR NKAE cells or conventional NK cells, respectively. 

Thus, common candidates associated with universal memory-induced properties, which 

could be targeted to redirect CAR NK cells towards a memory-like phenotype, remain 

unveiled. 

In continuation with this line of work, the efficacy of NKG2D- or BCMA-CAR NKAE 

cells as well as NKG2D-CAR CIML-NKAE cells can be improved by redirecting these 

effector cells to other tumor antigens using BiKEs. As our NKG2D-CAR CIML-NKAE 

have a high expression of CD16, in contrast to conventional CIML-NK cells, we are 

currently designing a BiKE molecule that specifically targets CD16 on the CAR NKAE 

cell and CD44v6 on the MM cell. This strategy would not only enhance tumor cell 

recognition through CAR binding but also trigger CAR NK cell activity through other 

endogenous signaling molecules (CD16). 



DISCUSSION 

193 
 

 

Figure 51. The blockade of NKG2A enhances NKG2D- and BCMA-CAR NKAE cells against MM 

cells. The use of neutralizing α-NKG2A Abs impairs NKG2A binding to HLA-E in MM cells. CAR 

signaling, together with native activating receptors, augment NK cell cytotoxicity against tumor cells and 

IFN-γ production, probably mediated by a higher phosphorylation of intracellular kinases as ZAP70 and 

Syk. NKG2A: natural killer group 2 member A; HLA: human leukocyte antigen; NKG2D: natural killer 

group 2 member D; CAR: chimeric antigen receptor; ZAP-70: zeta-chain-associated protein kinase 70; 

SYK: spleen tyrosine kinase; MM: multiple myeloma; IFN-γ: interferon-γ; BCMA: B-cell maturation 

antigen. Created with BioRender.com. 

 

In summary, we propose two models to enhance CAR NKAE therapies for the treatment 

of RRMM. We have studied the antitumoral activity of NKG2A blockade in combination 

with NKG2D- and BCMA-CAR NKAE cells (Fig. 51). First preclinical data combining 

the BMS-986315 α-NKG2A Ab and two novel CAR NK cell adoptive therapies, targeting 

NKG2D-L and BCMA on MM cells demonstrate an enhanced efficacy against MM cells 



DISCUSSION 

194 
 

in vitro and in vivo. Furthermore, we have cytokine-induced a memory phenotype in 

NKG2D-CAR NKAE cells to target, for the first time, MM, demonstrating an improved 

antitumor ability in vitro and in vivo. The brief exposure to IL-12, IL-15, and IL-18 is 

able to reprogram NKG2D-CAR CIML-NKAE cells towards a more proliferative and 

glycolytic cell through transcriptomic and epigenetic changes (Fig. 52). Thus, the 

blockade of NKG2A or the induction of memory properties on CAR NKAE cells entail 

two new strategies to treat not only MM patients, but also other patients with HLA-E+ 

tumors. 

 

 

 

Figure 52. The induction of memory properties improves NKG2D-CAR NKAE cells cytotoxicity 

against MM cells. The exposure to IL-12, IL-15, and IL-18 is able to induce epigenetic and transcriptomic 

changes on NKG2D-CAR NKAE cells. The new NKG2D-CAR CIML-NKAE population exerts a higher 

cytotoxicity against MM cells through an augmented degranulation and IFN-γ, perforin and granzyme A 

production, as well as a higher proliferation. Furthermore, these enhance cells have a high expression of 

CD25, CD2 and CD71, together with a superior glycolysis, and reduced CD57. CIML: cytokine-induced 

memory-like; TGF-β: transforming growth factor-β; NKG2D: natural killer group 2 member D; CAR: 

chimeric antigen receptor; IFN-γ: interferon-γ; MM: multiple myeloma; MICA/B: major histocompatibility 

complex class I chain-related protein A/B; ULBP1-6: unique long 16 binding protein 1-6. Created with 

BioRender.com.

 

 

 



 

195 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

6. CONCLUSIONES 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

196 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



CONCLUSIONES 

197 
 

Las principales conclusiones extraídas de este trabajo se enumeran a continuación: 

1. La expresión de HLA-E es mayor en las células plasmáticas de pacientes de MM 

comparada con la de las células plasmáticas de donantes sanos. HLA-E está sobre-

expresado en las células plasmáticas patológicas de pacientes de MM con respecto 

a las células plasmáticas normales y las células CD34+ del mismo paciente, 

especialmente en progresión.  

 

2. Tanto el IFN-γ como el BTZ modulan la expresión de HLA-E en líneas celulares 

y células primarias de MM. 

 

3. Aunque las células NK citotóxicas presentes en la médula ósea de pacientes de 

MM muestran una alta expresión de NKG2A y las células plasmáticas sobre-

expresan HLA-E, el tratamiento en monoterapia con dos anticuerpos de bloqueo 

α-NKG2A (Z199 o BMS-986315) no restaura la capacidad anti-MM de las células 

NK in vitro. 

 

4. La generación de células alogénicas NKG2D- o BCMA-CAR NKAE a partir de 

sangre periférica de donantes sanos es factible. Las señales que participan en el 

proceso de expansión y activación de estas células inducen un aumento de la 

expresión de NKG2A en ambas poblaciones. 

 

5. El pretratamiento de las células NKG2D- o BCMA-CAR NKAE con el anticuerpo 

de bloqueo α-NKG2A BMS-986315 aumenta la eficacia de las CAR NKAE frente 

a células de MM in vitro e in vivo. 

 

6. El aumento de eficacia in vitro de las células NKG2D- o BCMA-CAR NKAE en 

combinación con los anticuerpos de bloqueo de NKG2A se asocia con una mayor 

función de la célula NK mediada por un incremento de la degranulación y de la 

secreción de IFN-γ, sin alteraciones importantes en el inmunofenotipo. 

 

7. El tratamiento con células NKG2D- o BCMA-CAR NKAE en combinación con 

anticuerpos α-NKG2A no produce toxicidad relevante sobre células CD34+ de 

pacientes de MM ni sobre células maduras de sangre periférica de donantes sanos. 

 



CONCLUSIONES 

198 
 

8. Es factible la generación de células NKG2D-CAR CIML-NKAE a partir de 

volúmenes reducidos de sangre periférica de donantes sanos. 

 

9. La exposición de las células NKG2D-CAR NKAE a las interleuquinas IL-12, IL-

15 e IL-18 induce cambios a nivel epigenético, transcriptómico e 

inmunofenotípico, que solapan parcialmente con los descritos en las células 

CIML-NK convencionales. 

 

10. Las células NKG2D-CAR CIML-NKAE se caracterizan por un aumento de la 

degranulación, producción de proteínas citolíticas, metabolismo glucolítico y 

proliferación in vitro, así como una mayor actividad anti-MM in vitro e in vivo, 

comparadas con las células NKG2D-CAR NKAE. 

 

11. El aumento de la eficacia anti-MM in vivo de las células NKG2D-CAR CIML-

NKAE no está relacionado con una mayor persistencia respecto a las células 

NKG2D-CAR NKAE, si no que se asocia con un aumento en la proporción de las 

células NKG2D-CAR CIML-NKAE en los ratones. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

199 
 

 

 

 

 

 

 

 

 

 

 

 

 

7. CONCLUSIONS 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

200 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



CONCLUSIONS 

201 
 

Main conclusions drawn from this work are listed below: 

1. HLA-E expression is higher on PC from MM patients compared to PC from HD. 

HLA-E is overexpressed on pPC with respect to nPC and CD34+ cells from the 

same MM patients, especially at progression. 

 

2. Both IFN-γ and BTZ modulate HLA-E expression on cell lines and primary MM 

cells. 

 

3. Although cytotoxic NK cells from MM patient BM show a high expression of 

NKG2A and PC overexpress HLA-E, monotherapy treatment with two blocking 

Abs (Z199 or BMS-986315) does not restore the anti-MM NK cell capacity in 

vitro. 

 

4. Allogeneic NKG2D- or BCMA-CAR NKAE cell generation from HD PB is 

feasible. The signals that participate in NK cell expansion and activation process 

induce an increase of NKG2A expression in both populations. 

 

5. NKG2D- or BCMA-CAR NKAE cell pretreatment with BMS-986315 α-NKG2A 

Ab augments CAR NKAE cell efficacy against MM cells in vitro and in vivo. 

 

6. The increase of in vitro efficacy of NKG2D- or BCMA-CAR NKAE cells in 

combination with NKG2A blocking Abs is associated with a higher NK function 

mediated by an augmented degranulation and IFN-γ secretion, without major 

immunophenotype alterations. 

 

7. NKG2D- or BCMA-CAR NKAE treatment in combination with α-NKG2A Abs 

do not produce a relevant toxicity against either CD34+ cells from MM patients 

or mature PB cells from HD. 

 

8. NKG2D-CAR CIML-NKAE cell generation is feasible starting from reduced HD 

PB volumes. 

 

9. The exposure of NKG2D-CAR NKAE cells to IL-12, IL-15, and IL-18 

interleukins induces epigenetic, transcriptomic, and immunophenotypic changes 

that partially overlaps with the ones described in conventional CIML-NK cells. 



CONCLUSIONS 

202 
 

 

10. NKG2D-CAR CIML-NKAE cells are characterized by an increased 

degranulation, cytolytic protein production, glycolytic metabolism, and 

proliferation in vitro, as well as by an augmented anti-MM activity in vitro and in 

vivo, compared to NKG2D-CAR NKAE cells. 

 

11. The increase of the NKG2D-CAR CIML-NKAE cell in vivo anti-MM efficacy is 

not related to a higher persistence with respect to NKG2D-CAR NKAE cells, but 

it is associated with an augmented NKG2D-CAR CIML-NKAE frequency in 

mice. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

203 
 

 

 

 

 

 

 

 

 

 

 

 

 

8. ABBREVIATIONS 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

204 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



ABBREVIATIONS 

205 
 

2-DG: 2-desoxy-D-glucose 

21spe-MLL5: splice variant mixed-

lineage leukemia-5 

4-1BBL: 4-1BB ligand 

 

A 

A2AR: adenosine A2a receptor 

Ab: antibody 

ADAM10/17: a disintegrin and 

metallopeptidase 10/17 

ADC: antibody-drug conjugate 

ADCC: antibody-dependent cellular 

cytotoxicity 

ADCP: antibody-dependent cellular 

phagocytosis 

ADO: adenosine 

ADP: adenosine diphosphate 

AKT: Ak strain transforming 

ALL: acute lymphocytic leukemia 

AML: acute myeloid leukemia 

ANLL: acute nonlymphocytic leukemia 

ANOVA: analysis of variance 

APC: allophycocyanin 

APC/Cy7: allophycocyanin-cyanine7 

APRIL: a proliferation-inducing ligand 

APVAL: adjusted p-value 

ARM: antibody recruiting molecule 

ASCT: autologous stem cell transplant 

ASCT-1/2: alanine/serine/cysteine 

transporter 1/2 

ATP: adenosine triphosphate 

 

B 

Β2-m: β-2 microglobulin 

B7-H6: B7 homolog 6 

BaEv: Baboon envelope 

BAFF: B-cell activating factor  

BAG6: B-cell lymphoma 2 associated 

athanogene 6 

BCL-2: B-cell lymphoma 2 

BCMA: B-cell maturation antigen 

BCR: B-cell receptor 

BFU-E: burst forming unit erythroid 

BiAbs: bispecific antibodies 

BiKE: bispecific killer cell engager 

BIRC2/3: baculoviral IAP repeat 

containing 2/3 

BiTE: bispecific T-cell engager 

BLI: bioluminescent imaging 

BM: bone marrow 

BMMCs: bone marrow mononuclear 

cells 

BMS: Bristol Myers Squibb 

BMSCs: bone marrow mesenchymal 

stromal cells 

BPDCN: blastic plasmacytoid dendritic 

cell neoplasm 

BRCA2: breast cancer gene 2 

Bregs: regulatory B cells 

BSA: bovine serum albumin 

BTZ: bortezomib 

 

 

 



ABBREVIATIONS 

206 
 

C 

CABLES1: cyclin-dependent kinase-5 

and Abelson murine leukemia 1 enzyme 

substrate 1 

Calcein-AM: calcein-

acetoxymethylester 

CAR: chimeric antigen receptor 

Cas9: clustered regularly interspaced 

short palindromic repeats associated 

protein 9 

CB: cord blood 

CB-NK: cord blood natural killer cells 

CCL: CC chemokine ligand 

CCR: CC chemokine receptor 

CCND2: cyclin D2 

CD44v6: CD44 variant domain 6 

CDC: complement-dependent 

cytotoxicity 

CDKN1B: cyclin dependent kinase 

inhibitor 1B 

CDKN2C: cyclin-dependent kinase 4 

inhibitor C 

CFU-GM: colony-forming unit 

granulocyte-macrophage 

Cilta-cel: ciltacabtagene autoleucel 

CIML: cytokine-induced memory-like 

CLEC2D: C-type lectin domain family 

2 member D 

CLL: chronic lymphocytic leukemia 

CMV: cytomegalovirus 

CR: complete response 

CRAB: hypercalcemia, renal failure, 

anemia and bone lesions 

CRISPR: clustered regularly 

interspaced short palindromic repeats 

CRS: cytokine release syndrome 

CRTAM: MHC class I–restricted T 

cell–associated molecule 

CT: computed tomography 

CTLA-4: cytotoxic T-lymphocyte 

antigen 4 

CXCL: CXC chemokine ligand 

CXCR: CXC chemokine receptor 

CyTOF: cytometry by time of flight 

 

D 

DAP10/12: DNAX-activating protein 

10/12 

DAPI: 4',6-diamidino-2-phenylindole 

DC: dendritic cell 

DEG: differentially expressed genes 

DLBCL: diffuse large B cell lymphoma 

DKK1: Dickkopf 1 

DMSO: dimethyl sulfoxide 

DNA: deoxyribonucleic acid 

DNAM-1: DNAX accessory molecule-

1 

 

E 

EBV: Epstein-Barr virus 

ECAR: extracellular acidification rate 

ECL: enhanced chemiluminescence 

EDTA: ethylenediaminetetraacetic acid 

EMA: European Medicines Agency 

EPO: erythropoietin 



ABBREVIATIONS 

207 
 

ER: endoplasmic reticulum 

ERK: extracellular signal-regulated 

kinase 

 

F 

FAM46C: family with sequence 

similarity 46 member C 

FasL: Fas ligand 

FBS: fetal bovine serum 

FCCP: fluoro-carbonyl cyanide 

phenylhydrazone 

FcεRIγ: γ-chain high-affinity IgE 

receptor 

FcRγ: Fc receptor γ-chain 

FcRH5: Fc receptor-homolog 5 

FDA: Food and Drug Administration 

FGFBP2: fibroblast growth factor 

binding protein 2 

FGFR3: fibroblast growth factor 

receptor 3 

FISH: fluorescence in situ hybridization 

FITC: fluorescein isothiocyanate 

FLC: free light chain 

FMO: fluorescence minus one 

FSC: forward side channel  

 

G 

GFP: green fluorescent protein 

GFR: glomerular filtration rate 

GLUT1/3: glucose transporter 1/3 

GM-CSF: granulocyte macrophage 

colony-stimulating factor 

GO BP: gene ontology biological 

processes 

GPRC5D: G protein-coupled receptor 

class C group 5 member D 

Grb2: growth factor receptor-bound 

protein 2 

GvHD: graft versus host disease 

 

H 

h-MM: hyperdiploid multiple myeloma 

H12O: Hospital 12 de Octubre 

HAMA: human anti-mouse antibody 

HCV: hepatitis C virus 

HD: healthy donor 

HDAC: histone deacetylase 

HER2: human epidermal growth factor 

receptor 2 

hESC: human embryonic stem cells 

HLA: human leukocyte antigen 

HNSCC: head and neck squamous cell 

carcinoma 

HR: hazard ratio 

HSPCs: hematopoietic stem/progenitor 

cells 

HRP: horseradish peroxidase 

 

I 

i.p: intraperitoneal 

i.v: intravenous 

ICANS: immune effector cell-

associated neurotoxicity syndrome 

ICI: immune checkpoint inhibitors 



ABBREVIATIONS 

208 
 

ICOS: inducible costimulatory 

Ide-cel: idecabtagene vicleucel 

IDH3A: isocitrate dehydrogenase 

subunit alpha 

IDO: indoleamine-pyrrole 2,3-

dioxygenase 

IFN-γ: interferon-γ 

Ig: immunoglobulin 

IgH: immunoglobulin heavy chain 

IgL: immunoglobulin light chain 

IKZF1/3: Ikaros family zinc finger 1/3 

IL: interleukin 

IMDM: Iscove´s Modified Dulbecco´s 

Medium 

IMiD: immunomodulatory agents 

iPSCs: induced pluripotent stem cells 

IRAK3: interleukin 1 receptor 

associated kinase 3 

IRS2: insulin receptor substrate 2 

ISS: international staging system 

ITAM: immunoreceptor tyrosine-based 

activation motif 

ITIM: immunoreceptor tyrosine-based 

inhibitory motif 

ITT: Ig tyrosine tail 

 

J 

JAK/STAT: Janus kinase/signal 

transducers and activators of 

transcription 

 

K 

KI: knock in 

KIR: killer-cell immunoglobulin-like 

receptors 

KLRC1: killer cell lectin-like receptor 

C1 

KLRG1: killer cell lectin-like receptor 

G1 

KO: knock out 

 

L 

LAG3: lymphocyte activation gene 3 

LCK: lymphocyte-specific protein 

tyrosine kinase 

LDH: lactate dehydrogenase 

LDL-R: low density lipoprotein 

receptor 

 

M 

mAb: monoclonal antibody 

MAC: membrane attack complex 

MANF3: mesencephalic astrocyte-

derived neurotrophic factor 3 

MAPK: mitogen-activated protein 

kinase 

MCP-1: monocyte chemoattractant 

protein-1 

MDS: myelodysplastic syndrome 

MDSCs: myeloid-derived suppressor 

cells 

MFI: median fluorescence intensity 

MGUS: monoclonal gammopathy of 

undetermined significance 

MHC: major histocompatibility 

complex 



ABBREVIATIONS 

209 
 

MICA/B: major histocompatibility 

complex class I chain-related protein 

A/B 

MIL: marrow-infiltrating lymphocyte 

MIP-1α: macrophage inflammatory 

protein-1α 

MM: multiple myeloma 

MMP: matrix metalloproteinase 

MOI: multiplicity of infection 

MR: minimal response 

MRD: minimal residual disease 

MRI: magnetic resonance imaging 

mRNA: messenger RNA 

MUC1: mucin 1 

 

N 

NAADP: nicotinic acid adenine 

dinucleotide phosphate 

NAD: nicotinamide adenine 

dinucleotide 

NADP: nicotinamide adenine 

dinucleotide phosphate 

NCR: natural cytotoxicity receptors 

NCT: national clinical trial 

NDMM: newly diagnosed multiple 

myeloma 

NF-κB: nuclear factor κ B 

nh-MM: non-hyperdiploid multiple 

myeloma 

NHL: non-Hodgkin lymphoma 

NK: natural killer 

NKG2A/B/C/D/E: natural killer group 

2 member A/B/C/D/E 

NKG2D-L: natural killer group 2 

member D ligands 

nPC: normal plasma cell 

NPM1: nucleophosmin 1 

NRROS: negative regulator of reactive 

oxygen species 

NSCLC: non-small cell lung cancer 

 

O 

OCR: oxygen consumption rate 

OPG: osteoprotegerin 

ORR: overall response rate 

OS: overall survival 

OXPHOS: oxidative phosphorylation 

 

P 

P/S: penicillin streptomycin 

PB: peripheral blood 

PBMCs: peripheral blood mononuclear 

cells 

PBS: phosphate-buffered saline 

PC: plasma cell 

PCNA: proliferating cell nuclear 

antigen 

PD-1: programmed cell death 1 

PD-L1/2: programmed cell death ligand 

1/2 

PDAC: pancreatic ductal 

adenocarcinoma 

PE: phycoerythrin 

PE/Cy7: phycoerythrin-cyanine7 

PEBL: protein expression blocker 



ABBREVIATIONS 

210 
 

PerCP/Cy5.5: peridinin chlorophyll 

protein-Cyanine5.5 

PET-CT: positron emission 

tomography and computed tomography 

PFS: progression-free survival 

PGE2: prostaglandin E2 

PI: proteosome inhibitor 

PI3K: phosphoinositide 3-kinase 

pPC: pathologic plasma cell 

PR: partial response 

PVR: poliovirus receptor 

PYHIN1: pyrin and hematopoietic 

expression, interferon-inducible nature, 

and nuclear localization domain family 

member 1 

 

Q 

q-PCR: quantitative polymerase chain 

reaction 

 

R 

R-ISS: revised international staging 

system 

RANK: receptor activator for nuclear 

factor κ B 

RANKL: receptor activator for nuclear 

factor κ B ligand 

RAS: rat sarcoma 

RB1: retinoblastoma protein 

RHAMM: receptor for hyaluronan-

mediated motility 

RNA: ribonucleic acid 

RPMI: Roswell Park Memorial 

Institute 

RRMM: relapsed/refractory multiple 

myeloma 

RT: room temperature 

 

S 

SEM: standard error of the mean 

SCF: stem cell factor 

scFv: single-chain variable fragment 

SCLC: small-cell lung cancer 

SDS-PAGE: sodium dodecyl sulfate 

polyacrylamide gel electrophoresis 

SEMA: semaphoring 

SHP-1: Src homology region 2 domain-

containing phosphatase-1 

SIRPα: signal regulatory protein α 

SLAMF7: signaling lymphocyte 

activation molecule family member 7 

SMM: smoldering multiple myeloma 

SNAP: self-labeling protein tag 

SSC: side scatter channel 

STAT1/3: signal transducer and 

activator of transcription 1/3 

SYK: spleen tyrosine kinase 

SYNE1: synaptic nuclear envelope 

protein 1 

synNotch: synthetic Notch 

 

T 

T:E: target: effector 

TACI: transmembrane activator and 

calcium modulator and cyclophilin 

ligand interactor 

TBS: tris-buffered saline 



ABBREVIATIONS 

211 
 

TBS-T: tris-buffered saline with 0.1% 

Tween 20 

TCF7: transcription factor 7 

TCR: T cell receptor 

TGF-β: transforming growth factor-β 

TGFBRII: transforming growth factor-

β receptor 2 

TFRC: transferrin receptor 

TIGIT: T cell immunoreceptor with Ig 

and ITIM domains 

TIM-3: T cell immunoglobulin domain 

and mucin domain-3 

TIML-NK: tumor-induced memory-

like NK cell 

TLR1: toll-like receptor 1 

TME: tumor microenvironment 

TNF-α: tumor necrosis factor α 

TNFR11A: tumor necrosis factor 

receptor superfamily member 11a 

TP53: tumor protein p53 

TRAIL: tumor necrosis factor related 

apoptosis-inducing ligand 

TRAIL-R: tumor necrosis factor related 

apoptosis-inducing ligand receptor 

Tregs: regulatory T cells 

 

U 

ULBP: unique long 16 binding protein 

 

 

V 

VCAM-1: vascular cell adhesion 

protein 1 

VCN: vector copy number 

VEGF: vascular endothelial growth 

factor 

VGPR: very good partial response 

VRd: Bortezomib, lenalidomide, and 

dexamethasone  

VSV-G: vesicular stomatitis virus G 

 

W 

WB: western blot 

WHSC1: Wolf-Hirschhorn syndrome 

candidate 1 

WNT: wingless-related integration site 

WWOX: WW domain containing 

oxidoreductase 

 

X 

XCL: chemokine (C motif) ligand 

XPO1: exportin 1 

 

Z 

ZAP-70: zeta-chain-associated protein 

kinase 70 

 

 

 

 

 



 

212 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

213 
 

 

 

 

 

 

 

 

 

 

 

 

 

9. BIBLIOGRAPHY 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



 

214 
 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 



BIBLIOGRAPHY 

215 
 

1. Rajkumar, S.V., Multiple myeloma: 2022 update on diagnosis, risk stratification, and 
management. Am J Hematol, 2022. 97(8): p. 1086-1107. 

2. Kyle, R.A., Multiple myeloma: an odyssey of discovery. Br J Haematol, 2000. 111(4): p. 
1035-44. 

3. Kyle, R.A., et al., Clinical course of light-chain smouldering multiple myeloma (idiopathic 
Bence Jones proteinuria): a retrospective cohort study. Lancet Haematol, 2014. 1(1): p. 
e28-e36. 

4. Fernandez de Larrea, C., et al., Primary plasma cell leukemia: consensus definition by 
the International Myeloma Working Group according to peripheral blood plasma cell 
percentage. Blood Cancer J, 2021. 11(12): p. 192. 

5. Kumar, S.K., et al., Multiple myeloma. Nat Rev Dis Primers, 2017. 3: p. 17046. 
6. Siegel, R.L., et al., Cancer statistics, 2023. CA Cancer J Clin, 2023. 73(1): p. 17-48. 
7. (ACS), A.C.S. Cancer Stat Facts: Myeloma. 2023; Available from: 

https://seer.cancer.gov/statfacts/html/mulmy.html. 
8. (AECC), A.E.C.e.C. Mieloma Múltiple. 2021; Available from: 

https://www.contraelcancer.es/es/todo-sobre-cancer/tipos-cancer/mieloma-
multiple/epidemiolog%C3%ADa#:~:text=De%20acuerdo%20con%20los%20datos,1.354
%20casos%20nuevos%20respectivamente%20en. 

9. Kyle, R.A. and S.V. Rajkumar, Multiple myeloma. Blood, 2008. 111(6): p. 2962-72. 
10. Landgren, O., et al., Risk of monoclonal gammopathy of undetermined significance 

(MGUS) and subsequent multiple myeloma among African American and white 
veterans in the United States. Blood, 2006. 107(3): p. 904-6. 

11. Korde, N., S.Y. Kristinsson, and O. Landgren, Monoclonal gammopathy of undetermined 
significance (MGUS) and smoldering multiple myeloma (SMM): novel biological insights 
and development of early treatment strategies. Blood, 2011. 117(21): p. 5573-81. 

12. Boffetta, P., et al., Exposure to ultraviolet radiation and risk of malignant lymphoma 
and multiple myeloma--a multicentre European case-control study. Int J Epidemiol, 
2008. 37(5): p. 1080-94. 

13. Friedman, G.D., Multiple myeloma: relation to propoxyphene and other drugs, 
radiation and occupation. Int J Epidemiol, 1986. 15(3): p. 424-6. 

14. Ichimaru, M., et al., Multiple myeloma among atomic bomb survivors in Hiroshima and 
Nagasaki, 1950-76: relationship to radiation dose absorbed by marrow. J Natl Cancer 
Inst, 1982. 69(2): p. 323-8. 

15. Landgren, O., et al., Pesticide exposure and risk of monoclonal gammopathy of 
undetermined significance in the Agricultural Health Study. Blood, 2009. 113(25): p. 
6386-91. 

16. Khuder, S.A. and A.B. Mutgi, Meta-analyses of multiple myeloma and farming. Am J Ind 
Med, 1997. 32(5): p. 510-6. 

17. Altekruse, S.F., S.J. Henley, and M.J. Thun, Deaths from hematopoietic and other 
cancers in relation to permanent hair dye use in a large prospective study (United 
States). Cancer Causes Control, 1999. 10(6): p. 617-25. 

18. Preston, D.L., et al., Cancer incidence in atomic bomb survivors. Part III. Leukemia, 
lymphoma and multiple myeloma, 1950-1987. Radiat Res, 1994. 137(2 Suppl): p. S68-
97. 

19. Marques-Mourlet, C., et al., Obesity and myeloma: Clinical and mechanistic 
contributions to disease progression. Front Endocrinol (Lausanne), 2023. 14: p. 
1118691. 

20. Rodriguez-Garcia, A., et al., Efficacy of Antiviral Treatment in Hepatitis C Virus (HCV)-
Driven Monoclonal Gammopathies Including Myeloma. Front Immunol, 2021. 12: p. 
797209. 

21. Nair, S., et al., Antigen-mediated regulation in monoclonal gammopathies and 
myeloma. JCI Insight, 2018. 3(8). 

https://seer.cancer.gov/statfacts/html/mulmy.html
https://www.contraelcancer.es/es/todo-sobre-cancer/tipos-cancer/mieloma-multiple/epidemiolog%C3%ADa#:~:text=De%20acuerdo%20con%20los%20datos,1.354%20casos%20nuevos%20respectivamente%20en
https://www.contraelcancer.es/es/todo-sobre-cancer/tipos-cancer/mieloma-multiple/epidemiolog%C3%ADa#:~:text=De%20acuerdo%20con%20los%20datos,1.354%20casos%20nuevos%20respectivamente%20en
https://www.contraelcancer.es/es/todo-sobre-cancer/tipos-cancer/mieloma-multiple/epidemiolog%C3%ADa#:~:text=De%20acuerdo%20con%20los%20datos,1.354%20casos%20nuevos%20respectivamente%20en


BIBLIOGRAPHY 

216 
 

22. Nair, S., et al., Clonal Immunoglobulin against Lysolipids in the Origin of Myeloma. N 
Engl J Med, 2016. 374(6): p. 555-61. 

23. Maldonado, J.E. and R.A. Kyle, Familial myeloma. Report of eight families and a study 
of serum proteins in their relatives. Am J Med, 1974. 57(6): p. 875-84. 

24. Altieri, A., et al., Familial risks and temporal incidence trends of multiple myeloma. Eur J 
Cancer, 2006. 42(11): p. 1661-70. 

25. Vachon, C.M., et al., Increased risk of monoclonal gammopathy in first-degree relatives 
of patients with multiple myeloma or monoclonal gammopathy of undetermined 
significance. Blood, 2009. 114(4): p. 785-90. 

26. Morgan, G.J., B.A. Walker, and F.E. Davies, The genetic architecture of multiple 
myeloma. Nat Rev Cancer, 2012. 12(5): p. 335-48. 

27. Pinto, V., et al., Multiple Myeloma: Available Therapies and Causes of Drug Resistance. 
Cancers (Basel), 2020. 12(2). 

28. Amodio, N., et al., Epigenetic modifications in multiple myeloma: recent advances on 
the role of DNA and histone methylation. Expert Opin Ther Targets, 2017. 21(1): p. 91-
101. 

29. Hideshima, T., et al., Understanding multiple myeloma pathogenesis in the bone 
marrow to identify new therapeutic targets. Nat Rev Cancer, 2007. 7(8): p. 585-98. 

30. Barwick, B.G., et al., Cell of Origin and Genetic Alterations in the Pathogenesis of 
Multiple Myeloma. Front Immunol, 2019. 10: p. 1121. 

31. Papadhimitriou, S.I., et al., The Cytogenetic Profile of Primary and Secondary Plasma 
Cell Leukemia: Etiopathogenetic Perspectives, Prognostic Impact and Clinical Relevance 
to Newly Diagnosed Multiple Myeloma with Differential Circulating Clonal Plasma Cells. 
Biomedicines, 2022. 10(2). 

32. Dimopoulos, K., P. Gimsing, and K. Gronbaek, The role of epigenetics in the biology of 
multiple myeloma. Blood Cancer J, 2014. 4(5): p. e207. 

33. Garcia-Ortiz, A., et al., The Role of Tumor Microenvironment in Multiple Myeloma 
Development and Progression. Cancers (Basel), 2021. 13(2). 

34. Manier, S., et al., Bone marrow microenvironment in multiple myeloma progression. J 
Biomed Biotechnol, 2012. 2012: p. 157496. 

35. Palumbo, A. and K. Anderson, Multiple myeloma. N Engl J Med, 2011. 364(11): p. 1046-
60. 

36. Maiso, P., et al., Bone Marrow Mesenchymal Stromal Cells in Multiple Myeloma: Their 
Role as Active Contributors to Myeloma Progression. Cancers (Basel), 2021. 13(11). 

37. Oyajobi, B.O., Multiple myeloma/hypercalcemia. Arthritis Res Ther, 2007. 9 Suppl 
1(Suppl 1): p. S4. 

38. Chatziravdeli, V., et al., A systematic review and meta-analysis of interventional studies 
of bisphosphonates and denosumab in multiple myeloma and future perspectives. J 
Musculoskelet Neuronal Interact, 2022. 22(4): p. 596-621. 

39. Borrello, I., Can we change the disease biology of multiple myeloma? Leuk Res, 2012. 
36 Suppl 1(0 1): p. S3-12. 

40. Panaroni, C., A.J. Yee, and N.S. Raje, Myeloma and Bone Disease. Curr Osteoporos Rep, 
2017. 15(5): p. 483-498. 

41. Dimopoulos, M.A., et al., Pathogenesis and treatment of renal failure in multiple 
myeloma. Leukemia, 2008. 22(8): p. 1485-93. 

42. Garcia-Sanz, R., M.V. Mateos, and J.F. San Miguel, [Multiple myeloma]. Med Clin (Barc), 
2007. 129(3): p. 104-15. 

43. Cowan, A.J., et al., Diagnosis and Management of Multiple Myeloma: A Review. JAMA, 
2022. 327(5): p. 464-477. 

44. Liu, L., et al., Multiple myeloma hinders erythropoiesis and causes anaemia owing to 
high levels of CCL3 in the bone marrow microenvironment. Sci Rep, 2020. 10(1): p. 
20508. 



BIBLIOGRAPHY 

217 
 

45. Ludwig, H., G. Pohl, and A. Osterborg, Anemia in multiple myeloma. Clin Adv Hematol 
Oncol, 2004. 2(4): p. 233-41. 

46. Sorrig, R., et al., Immunoparesis in newly diagnosed Multiple Myeloma patients: Effects 
on overall survival and progression free survival in the Danish population. PLoS One, 
2017. 12(12): p. e0188988. 

47. Nucci, M. and E. Anaissie, Infections in patients with multiple myeloma in the era of 
high-dose therapy and novel agents. Clin Infect Dis, 2009. 49(8): p. 1211-25. 

48. Schutt, P., et al., Immune parameters in multiple myeloma patients: influence of 
treatment and correlation with opportunistic infections. Leuk Lymphoma, 2006. 47(8): 
p. 1570-82. 

49. Rajkumar, S.V., et al., International Myeloma Working Group updated criteria for the 
diagnosis of multiple myeloma. Lancet Oncol, 2014. 15(12): p. e538-48. 

50. Rajkumar, S.V. and S. Kumar, Multiple Myeloma: Diagnosis and Treatment. Mayo Clin 
Proc, 2016. 91(1): p. 101-19. 

51. Hillengass, J., et al., International myeloma working group consensus recommendations 
on imaging in monoclonal plasma cell disorders. Lancet Oncol, 2019. 20(6): p. e302-
e312. 

52. Johnsen, H.E., et al., Multiparametric flow cytometry profiling of neoplastic plasma 
cells in multiple myeloma. Cytometry B Clin Cytom, 2010. 78(5): p. 338-47. 

53. Radzevicius, M., et al., Multiple Myeloma Immunophenotype Related to Chromosomal 
Abnormalities Used in Risk Assessment. Diagnostics (Basel), 2022. 12(9). 

54. Bataille, R., et al., CD117 (c-kit) is aberrantly expressed in a subset of MGUS and 
multiple myeloma with unexpectedly good prognosis. Leuk Res, 2008. 32(3): p. 379-82. 

55. Paiva, B., et al., Clinical significance of CD81 expression by clonal plasma cells in high-
risk smoldering and symptomatic multiple myeloma patients. Leukemia, 2012. 26(8): p. 
1862-9. 

56. Palumbo, A., et al., Revised International Staging System for Multiple Myeloma: A 
Report From International Myeloma Working Group. J Clin Oncol, 2015. 33(26): p. 
2863-9. 

57. D'Agostino, M., et al., Second Revision of the International Staging System (R2-ISS) for 
Overall Survival in Multiple Myeloma: A European Myeloma Network (EMN) Report 
Within the HARMONY Project. J Clin Oncol, 2022. 40(29): p. 3406-3418. 

58. Tan, J.L.C., et al., The second revision of the International Staging System (R2-ISS) 
stratifies progression-free and overall survival in multiple myeloma: Real world data 
results in an Australian and New Zealand Population. Br J Haematol, 2023. 200(2): p. 
e17-e21. 

59. Blade, J., et al., Treatment of alkylating resistant multiple myeloma with vincristine, 
BCNU, doxorubicin and prednisone (VBAP). Eur J Cancer Clin Oncol, 1986. 22(10): p. 
1193-7. 

60. Durie, B.G.M., et al., Bortezomib with lenalidomide and dexamethasone versus 
lenalidomide and dexamethasone alone in patients with newly diagnosed myeloma 
without intent for immediate autologous stem-cell transplant (SWOG S0777): a 
randomised, open-label, phase 3 trial. Lancet, 2017. 389(10068): p. 519-527. 

61. Aguilar Franco, C., et al., Hematoguía Mieloma. 2023. 
62. Moreau, P., et al., Maintenance with daratumumab or observation following treatment 

with bortezomib, thalidomide, and dexamethasone with or without daratumumab and 
autologous stem-cell transplant in patients with newly diagnosed multiple myeloma 
(CASSIOPEIA): an open-label, randomised, phase 3 trial. Lancet Oncol, 2021. 22(10): p. 
1378-1390. 

63. Mateos, M.V., et al., Daratumumab Plus Bortezomib, Melphalan, and Prednisone Versus 
Bortezomib, Melphalan, and Prednisone in Transplant-Ineligible Newly Diagnosed 



BIBLIOGRAPHY 

218 
 

Multiple Myeloma: Frailty Subgroup Analysis of ALCYONE. Clin Lymphoma Myeloma 
Leuk, 2021. 21(11): p. 785-798. 

64. Parrondo, R.D., et al., Autologous Stem-Cell Transplantation for Multiple Myeloma in 
the Era of Novel Therapies. JCO Oncol Pract, 2020. 16(2): p. 56-66. 

65. Michel Attal, V.L.-C., Cyrille Hulin, Thierry Facon, Denis Caillot, Martine Escoffre, 
Bertrand Arnulf, Margaret MACRO, Karim Belhadj, Laurent Garderet, Murielle Roussel, 
Claire Mathiot, Herve Avet-Loiseau, Nikhil C. Munshi, Paul G. Richardson, Kenneth C. 
Anderson, jean Luc Harousseau, Philippe Moreau, Autologous Transplantation for 
Multiple Myeloma in the Era of New Drugs: A Phase III Study of the Intergroupe 
Francophone Du Myelome (IFM/DFCI 2009 Trial). ASH, 2015. 

66. Ntanasis-Stathopoulos, I., et al., Multiple myeloma: Role of autologous transplantation. 
Cancer Treat Rev, 2020. 82: p. 101929. 

67. Bazarbachi, A.H., et al., Induction therapy prior to autologous stem cell transplantation 
(ASCT) in newly diagnosed multiple myeloma: an update. Blood Cancer J, 2022. 12(3): 
p. 47. 

68. More, S., et al., Autologous Stem Cell Transplantation in Multiple Myeloma: Where Are 
We and Where Do We Want to Go? Cells, 2022. 11(4). 

69. Gagelmann, N., et al., Tandem Autologous Stem Cell Transplantation Improves 
Outcomes in Newly Diagnosed Multiple Myeloma with Extramedullary Disease and 
High-Risk Cytogenetics: A Study from the Chronic Malignancies Working Party of the 
European Society for Blood and Marrow Transplantation. Biol Blood Marrow 
Transplant, 2019. 25(11): p. 2134-2142. 

70. Minnie, S.A. and G.R. Hill, Autologous Stem Cell Transplantation for Myeloma: 
Cytoreduction or an Immunotherapy? Front Immunol, 2021. 12: p. 651288. 

71. Orrantia, A., et al., NK Cell Reconstitution After Autologous Hematopoietic Stem Cell 
Transplantation: Association Between NK Cell Maturation Stage and Outcome in 
Multiple Myeloma. Front Immunol, 2021. 12: p. 748207. 

72. Ito, T., et al., Identification of a primary target of thalidomide teratogenicity. Science, 
2010. 327(5971): p. 1345-50. 

73. Quach, H., et al., Mechanism of action of immunomodulatory drugs (IMiDS) in multiple 
myeloma. Leukemia, 2010. 24(1): p. 22-32. 

74. Heim, C. and M.D. Hartmann, High-resolution structures of the bound effectors 
avadomide (CC-122) and iberdomide (CC-220) highlight advantages and limitations of 
the MsCI4 soaking system. Acta Crystallogr D Struct Biol, 2022. 78(Pt 3): p. 290-298. 

75. Fionda, C., et al., The IMiDs targets IKZF-1/3 and IRF4 as novel negative regulators of 
NK cell-activating ligands expression in multiple myeloma. Oncotarget, 2015. 6(27): p. 
23609-30. 

76. Richardson, P.G., et al., Mezigdomide plus Dexamethasone in Relapsed and Refractory 
Multiple Myeloma. N Engl J Med, 2023. 

77. Kotla, V., et al., Mechanism of action of lenalidomide in hematological malignancies. J 
Hematol Oncol, 2009. 2: p. 36. 

78. Jungkunz-Stier, I., et al., Modulation of natural killer cell effector functions through 
lenalidomide/dasatinib and their combined effects against multiple myeloma cells. 
Leuk Lymphoma, 2014. 55(1): p. 168-76. 

79. Lagrue, K., et al., Lenalidomide augments actin remodeling and lowers NK-cell 
activation thresholds. Blood, 2015. 126(1): p. 50-60. 

80. Davies, F.E., et al., Thalidomide and immunomodulatory derivatives augment natural 
killer cell cytotoxicity in multiple myeloma. Blood, 2001. 98(1): p. 210-6. 

81. Corral, L.G., et al., Differential cytokine modulation and T cell activation by two distinct 
classes of thalidomide analogues that are potent inhibitors of TNF-alpha. J Immunol, 
1999. 163(1): p. 380-6. 



BIBLIOGRAPHY 

219 
 

82. Delforge, M. and H. Ludwig, How I manage the toxicities of myeloma drugs. Blood, 
2017. 129(17): p. 2359-2367. 

83. Manasanch, E.E. and R.Z. Orlowski, Proteasome inhibitors in cancer therapy. Nat Rev 
Clin Oncol, 2017. 14(7): p. 417-433. 

84. Besse, A., et al., Proteasome Inhibition in Multiple Myeloma: Head-to-Head 
Comparison of Currently Available Proteasome Inhibitors. Cell Chem Biol, 2019. 26(3): 
p. 340-351 e3. 

85. Carlsten, M., et al., Bortezomib sensitizes multiple myeloma to NK cells via ER-stress-
induced suppression of HLA-E and upregulation of DR5. Oncoimmunology, 2019. 8(2): 
p. e1534664. 

86. Nencioni, A., et al., Proteasome inhibitors: antitumor effects and beyond. Leukemia, 
2007. 21(1): p. 30-6. 

87. Mattingly, L.H., R.A. Gault, and W.J. Murphy, Use of systemic proteasome inhibition as 
an immune-modulating agent in disease. Endocr Metab Immune Disord Drug Targets, 
2007. 7(1): p. 29-34. 

88. Wieten, L., et al., Clinical and immunological significance of HLA-E in stem cell 
transplantation and cancer. Tissue Antigens, 2014. 84(6): p. 523-35. 

89. Shi, J., et al., Bortezomib down-regulates the cell-surface expression of HLA class I and 
enhances natural killer cell-mediated lysis of myeloma. Blood, 2008. 111(3): p. 1309-
17. 

90. Niu, C., et al., Low-dose bortezomib increases the expression of NKG2D and DNAM-1 
ligands and enhances induced NK and gammadelta T cell-mediated lysis in multiple 
myeloma. Oncotarget, 2017. 8(4): p. 5954-5964. 

91. Hallett, W.H., et al., Sensitization of tumor cells to NK cell-mediated killing by 
proteasome inhibition. J Immunol, 2008. 180(1): p. 163-70. 

92. Zheng, Y., et al., Cardiovascular Toxicity of Proteasome Inhibitors in Multiple Myeloma 
Therapy. Curr Probl Cardiol, 2023. 48(3): p. 101536. 

93. Poczta, A., A. Rogalska, and A. Marczak, Treatment of Multiple Myeloma and the Role 
of Melphalan in the Era of Modern Therapies-Current Research and Clinical 
Approaches. J Clin Med, 2021. 10(9). 

94. Samur, M.K., et al., High-dose melphalan treatment significantly increases mutational 
burden at relapse in multiple myeloma. Blood, 2023. 141(14): p. 1724-1736. 

95. Palumbo, A., et al., Daratumumab, Bortezomib, and Dexamethasone for Multiple 
Myeloma. N Engl J Med, 2016. 375(8): p. 754-66. 

96. Dimopoulos, M.A., et al., Daratumumab plus lenalidomide and dexamethasone versus 
lenalidomide and dexamethasone in relapsed or refractory multiple myeloma: updated 
analysis of POLLUX. Haematologica, 2018. 103(12): p. 2088-2096. 

97. Rajkumar, S.V., et al., Consensus recommendations for the uniform reporting of clinical 
trials: report of the International Myeloma Workshop Consensus Panel 1. Blood, 2011. 
117(18): p. 4691-5. 

98. Gandhi, U.H., et al., Outcomes of patients with multiple myeloma refractory to CD38-
targeted monoclonal antibody therapy. Leukemia, 2019. 33(9): p. 2266-2275. 

99. Podar, K., et al., Selinexor for the treatment of multiple myeloma. Expert Opin 
Pharmacother, 2020. 21(4): p. 399-408. 

100. Moreau, P., et al., Treatment of relapsed and refractory multiple myeloma: 
recommendations from the International Myeloma Working Group. Lancet Oncol, 
2021. 22(3): p. e105-e118. 

101. Fisher, J.G., et al., Selinexor Enhances NK Cell Activation Against Malignant B Cells via 
Downregulation of HLA-E. Front Oncol, 2021. 11: p. 785635. 

102. Hu, F., et al., Drug resistance biomarker ABCC4 of selinexor and immune feature in 
multiple myeloma. Int Immunopharmacol, 2022. 108: p. 108722. 



BIBLIOGRAPHY 

220 
 

103. Fisher, J.G., et al., Disruption of the NKG2A:HLA-E Immune Checkpoint Axis to Enhance 
NK Cell Activation against Cancer. Vaccines (Basel), 2022. 10(12). 

104. Ray, A., et al., A novel alkylating agent Melflufen induces irreversible DNA damage and 
cytotoxicity in multiple myeloma cells. Br J Haematol, 2016. 174(3): p. 397-409. 

105. Pour, L., et al., Efficacy and safety of melflufen plus daratumumab and dexamethasone 
in relapsed/refractory multiple myeloma: results from the randomized, open-label, 
phase III LIGHTHOUSE study. Haematologica, 2023. 

106. (EMA), E.M.A., 2023. 
107. Foundation, I.M., Current FDA-Approved Medications: Multiple Myeloma Drugs. 2023. 
108. Rodriguez-Otero, P., et al., Ide-cel or Standard Regimens in Relapsed and Refractory 

Multiple Myeloma. N Engl J Med, 2023. 388(11): p. 1002-1014. 
109. San-Miguel, J., et al., Cilta-cel or Standard Care in Lenalidomide-Refractory Multiple 

Myeloma. N Engl J Med, 2023. 
110. Moreau, P., et al., Teclistamab in Relapsed or Refractory Multiple Myeloma. N Engl J 

Med, 2022. 387(6): p. 495-505. 
111. Lesokhin, A.M., et al., Elranatamab in relapsed or refractory multiple myeloma: phase 2 

MagnetisMM-3 trial results. Nat Med, 2023. 29(9): p. 2259-2267. 
112. Chari, A., et al., Talquetamab, a T-Cell-Redirecting GPRC5D Bispecific Antibody for 

Multiple Myeloma. N Engl J Med, 2022. 387(24): p. 2232-2244. 
113. Jimenez-Zepeda, V.H., et al., Second autologous stem cell transplantation as salvage 

therapy for multiple myeloma: impact on progression-free and overall survival. Biol 
Blood Marrow Transplant, 2012. 18(5): p. 773-9. 

114. Dhakal, B., et al., Salvage second transplantation in relapsed multiple myeloma. 
Leukemia, 2021. 35(4): p. 1214-1217. 

115. Kumar, S., et al., Efficacy of venetoclax as targeted therapy for relapsed/refractory 
t(11;14) multiple myeloma. Blood, 2017. 130(22): p. 2401-2409. 

116. Sidiqi, M.H., et al., Venetoclax for the treatment of multiple myeloma: Outcomes 
outside of clinical trials. Am J Hematol, 2021. 96(9): p. 1131-1136. 

117. Kumar, S.K., et al., Venetoclax or placebo in combination with bortezomib and 
dexamethasone in patients with relapsed or refractory multiple myeloma (BELLINI): a 
randomised, double-blind, multicentre, phase 3 trial. Lancet Oncol, 2020. 21(12): p. 
1630-1642. 

118. Pan, D., et al., Outcomes with panobinostat in heavily pretreated multiple myeloma 
patients. Semin Oncol, 2023. 50(1-2): p. 40-48. 

119. Maouche, N., et al., Panobinostat in combination with bortezomib and dexamethasone 
in multiply relapsed and refractory myeloma; UK routine care cohort. PLoS One, 2022. 
17(7): p. e0270854. 

120. Minnie, S.A. and G.R. Hill, Immunotherapy of multiple myeloma. J Clin Invest, 2020. 
130(4): p. 1565-1575. 

121. Visram, A., et al., Relapsed multiple myeloma demonstrates distinct patterns of immune 
microenvironment and malignant cell-mediated immunosuppression. Blood Cancer J, 
2021. 11(3): p. 45. 

122. Pessoa de Magalhaes, R.J., et al., Analysis of the immune system of multiple myeloma 
patients achieving long-term disease control by multidimensional flow cytometry. 
Haematologica, 2013. 98(1): p. 79-86. 

123. El-Sherbiny, Y.M., et al., The requirement for DNAM-1, NKG2D, and NKp46 in the 
natural killer cell-mediated killing of myeloma cells. Cancer Res, 2007. 67(18): p. 8444-
9. 

124. Carbone, E., et al., HLA class I, NKG2D, and natural cytotoxicity receptors regulate 
multiple myeloma cell recognition by natural killer cells. Blood, 2005. 105(1): p. 251-8. 

125. Yang, Y., et al., HLA-E Binding Peptide as a Potential Therapeutic Candidate for High-
Risk Multiple Myeloma. Front Oncol, 2021. 11: p. 670673. 



BIBLIOGRAPHY 

221 
 

126. Racanelli, V., et al., Alterations in the antigen processing-presenting machinery of 
transformed plasma cells are associated with reduced recognition by CD8+ T cells and 
characterize the progression of MGUS to multiple myeloma. Blood, 2010. 115(6): p. 
1185-93. 

127. Konjevic, G., et al., Decreased CD161 activating and increased CD158a inhibitory 
receptor expression on NK cells underlies impaired NK cell cytotoxicity in patients with 
multiple myeloma. J Clin Pathol, 2016. 

128. Rosenblatt, J., et al., PD-1 blockade by CT-011, anti-PD-1 antibody, enhances ex vivo T-
cell responses to autologous dendritic cell/myeloma fusion vaccine. J Immunother, 
2011. 34(5): p. 409-18. 

129. Liu, J., et al., Plasma cells from multiple myeloma patients express B7-H1 (PD-L1) and 
increase expression after stimulation with IFN-gamma and TLR ligands via a MyD88-, 
TRAF6-, and MEK-dependent pathway. Blood, 2007. 110(1): p. 296-304. 

130. Perez-Andres, M., et al., Characterization of bone marrow T cells in monoclonal 
gammopathy of undetermined significance, multiple myeloma, and plasma cell 
leukemia demonstrates increased infiltration by cytotoxic/Th1 T cells demonstrating a 
squed TCR-Vbeta repertoire. Cancer, 2006. 106(6): p. 1296-305. 

131. Jinushi, M., et al., MHC class I chain-related protein A antibodies and shedding are 
associated with the progression of multiple myeloma. Proc Natl Acad Sci U S A, 2008. 
105(4): p. 1285-90. 

132. Uhl, C., et al., Natural killer cells activity against multiple myeloma cells is modulated 
by osteoblast-induced IL-6 and IL-10 production. Heliyon, 2022. 8(3): p. e09167. 

133. Ratta, M., et al., Dendritic cells are functionally defective in multiple myeloma: the role 
of interleukin-6. Blood, 2002. 100(1): p. 230-7. 

134. D'Andrea, A., et al., Interleukin 10 (IL-10) inhibits human lymphocyte interferon 
gamma-production by suppressing natural killer cell stimulatory factor/IL-12 synthesis 
in accessory cells. J Exp Med, 1993. 178(3): p. 1041-8. 

135. Rana, P.S., et al., Targeting TGF-beta signaling in the multiple myeloma 
microenvironment: Steering CARs and T cells in the right direction. Front Cell Dev Biol, 
2022. 10: p. 1059715. 

136. Strobl, H. and W. Knapp, TGF-beta1 regulation of dendritic cells. Microbes Infect, 1999. 
1(15): p. 1283-90. 

137. Iwasa, M., et al., PD-L1 upregulation in myeloma cells by panobinostat in combination 
with interferon-gamma. Oncotarget, 2019. 10(20): p. 1903-1917. 

138. Ray, A., et al., Targeting tryptophan catabolic kynurenine pathway enhances antitumor 
immunity and cytotoxicity in multiple myeloma. Leukemia, 2020. 34(2): p. 567-577. 

139. Katz, J.B., A.J. Muller, and G.C. Prendergast, Indoleamine 2,3-dioxygenase in T-cell 
tolerance and tumoral immune escape. Immunol Rev, 2008. 222: p. 206-21. 

140. Li, W., et al., CD38: An important regulator of T cell function. Biomed Pharmacother, 
2022. 153: p. 113395. 

141. Rebmann, V., et al., Soluble MICA as an independent prognostic factor for the overall 
survival and progression-free survival of multiple myeloma patients. Clin Immunol, 
2007. 123(1): p. 114-20. 

142. Diaz-Tejedor, A., et al., Immune System Alterations in Multiple Myeloma: Molecular 
Mechanisms and Therapeutic Strategies to Reverse Immunosuppression. Cancers 
(Basel), 2021. 13(6). 

143. Dosani, T., et al., The cellular immune system in myelomagenesis: NK cells and T cells in 
the development of MM and their uses in immunotherapies. Blood Cancer J, 2015. 5(7): 
p. e321. 

144. Richardson, P.G., et al., Isatuximab plus pomalidomide and low-dose dexamethasone 
versus pomalidomide and low-dose dexamethasone in patients with relapsed and 



BIBLIOGRAPHY 

222 
 

refractory multiple myeloma (ICARIA-MM): follow-up analysis of a randomised, phase 3 
study. Lancet Oncol, 2022. 23(3): p. 416-427. 

145. Shah, N., et al., B-cell maturation antigen (BCMA) in multiple myeloma: rationale for 
targeting and current therapeutic approaches. Leukemia, 2020. 34(4): p. 985-1005. 

146. Van Oekelen, O., et al., Neurocognitive and hypokinetic movement disorder with 
features of parkinsonism after BCMA-targeting CAR-T cell therapy. Nat Med, 2021. 
27(12): p. 2099-2103. 

147. Cho, S.F., et al., Promising Antigens for the New Frontier of Targeted Immunotherapy in 
Multiple Myeloma. Cancers (Basel), 2021. 13(23). 

148. Giordano, D., et al., B cell-activating factor (BAFF) from dendritic cells, monocytes and 
neutrophils is required for B cell maturation and autoantibody production in SLE-like 
autoimmune disease. Front Immunol, 2023. 14: p. 1050528. 

149. Hahne, M., et al., APRIL, a new ligand of the tumor necrosis factor family, stimulates 
tumor cell growth. J Exp Med, 1998. 188(6): p. 1185-90. 

150. Moreaux, J., et al., BAFF and APRIL protect myeloma cells from apoptosis induced by 
interleukin 6 deprivation and dexamethasone. Blood, 2004. 103(8): p. 3148-57. 

151. Ndacayisaba, L.J., et al., Characterization of BCMA Expression in Circulating Rare Single 
Cells of Patients with Plasma Cell Neoplasms. Int J Mol Sci, 2022. 23(21). 

152. Tai, Y.T., et al., APRIL and BCMA promote human multiple myeloma growth and 
immunosuppression in the bone marrow microenvironment. Blood, 2016. 127(25): p. 
3225-36. 

153. Wiedemann, A., et al., Soluble B-cell maturation antigen as a monitoring marker for 
multiple myeloma. Pathol Oncol Res, 2023. 29: p. 1611171. 

154. Pont, M.J., et al., gamma-Secretase inhibition increases efficacy of BCMA-specific 
chimeric antigen receptor T cells in multiple myeloma. Blood, 2019. 134(19): p. 1585-
1597. 

155. Leivas, A., et al., NKG2D-CAR-transduced natural killer cells efficiently target multiple 
myeloma. Blood Cancer J, 2021. 11(8): p. 146. 

156. Allez, M., et al., CD4+NKG2D+ T cells in Crohn's disease mediate inflammatory and 
cytotoxic responses through MICA interactions. Gastroenterology, 2007. 132(7): p. 
2346-58. 

157. Saez-Borderias, A., et al., Expression and function of NKG2D in CD4+ T cells specific for 
human cytomegalovirus. Eur J Immunol, 2006. 36(12): p. 3198-206. 

158. Upshaw, J.L., et al., NKG2D-mediated signaling requires a DAP10-bound Grb2-Vav1 
intermediate and phosphatidylinositol-3-kinase in human natural killer cells. Nat 
Immunol, 2006. 7(5): p. 524-32. 

159. Roda-Navarro, P. and H.T. Reyburn, The traffic of the NKG2D/Dap10 receptor complex 
during natural killer (NK) cell activation. J Biol Chem, 2009. 284(24): p. 16463-16472. 

160. Sallman, D.A., et al., CYAD-01, an autologous NKG2D-based CAR T-cell therapy, in 
relapsed or refractory acute myeloid leukaemia and myelodysplastic syndromes or 
multiple myeloma (THINK): haematological cohorts of the dose escalation segment of a 
phase 1 trial. Lancet Haematol, 2023. 10(3): p. e191-e202. 

161. Gasser, S., et al., The DNA damage pathway regulates innate immune system ligands of 
the NKG2D receptor. Nature, 2005. 436(7054): p. 1186-90. 

162. Ward, J., et al., HIV modulates the expression of ligands important in triggering natural 
killer cell cytotoxic responses on infected primary T-cell blasts. Blood, 2007. 110(4): p. 
1207-14. 

163. Pende, D., et al., Major histocompatibility complex class I-related chain A and UL16-
binding protein expression on tumor cell lines of different histotypes: analysis of tumor 
susceptibility to NKG2D-dependent natural killer cell cytotoxicity. Cancer Res, 2002. 
62(21): p. 6178-86. 



BIBLIOGRAPHY 

223 
 

164. Groh, V., et al., Cell stress-regulated human major histocompatibility complex class I 
gene expressed in gastrointestinal epithelium. Proc Natl Acad Sci U S A, 1996. 93(22): p. 
12445-50. 

165. Raulet, D.H., et al., Regulation of ligands for the NKG2D activating receptor. Annu Rev 
Immunol, 2013. 31: p. 413-41. 

166. Fionda, C., et al., NKG2D and DNAM-1 Ligands: Molecular Targets for NK Cell-Mediated 
Immunotherapeutic Intervention in Multiple Myeloma. Biomed Res Int, 2015. 2015: p. 
178698. 

167. Zhang, T., A. Barber, and C.L. Sentman, Generation of antitumor responses by genetic 
modification of primary human T cells with a chimeric NKG2D receptor. Cancer Res, 
2006. 66(11): p. 5927-33. 

168. Atamaniuk, J., et al., Overexpression of G protein-coupled receptor 5D in the bone 
marrow is associated with poor prognosis in patients with multiple myeloma. Eur J Clin 
Invest, 2012. 42(9): p. 953-60. 

169. Pillarisetti, K., et al., A T-cell-redirecting bispecific G-protein-coupled receptor class 5 
member D x CD3 antibody to treat multiple myeloma. Blood, 2020. 135(15): p. 1232-
1243. 

170. Hasselbalch Riley, C., et al., RG6234, a novel GPRC5D T-cell engaging bispecific 
antibody, induces rapid responses in patients with relapsed/refractory multiple 
myeloma: preliminary results from a first-in-human trial. EHA, 2022. 

171. Smith, E.L., et al., GPRC5D is a target for the immunotherapy of multiple myeloma with 
rationally designed CAR T cells. Sci Transl Med, 2019. 11(485). 

172. Elkins, K., et al., FcRL5 as a target of antibody-drug conjugates for the treatment of 
multiple myeloma. Mol Cancer Ther, 2012. 11(10): p. 2222-32. 

173. Jiang, D., et al., Chimeric antigen receptor T cells targeting FcRH5 provide robust 
tumour-specific responses in murine xenograft models of multiple myeloma. Nat 
Commun, 2023. 14(1): p. 3642. 

174. Stewart, A.K., et al., Phase I study of the anti-FcRH5 antibody-drug conjugate 
DFRF4539A in relapsed or refractory multiple myeloma. Blood Cancer J, 2019. 9(2): p. 
17. 

175. Piedra-Quintero, Z.L., et al., CD38: An Immunomodulatory Molecule in Inflammation 
and Autoimmunity. Front Immunol, 2020. 11: p. 597959. 

176. Ogiya, D., et al., The JAK-STAT pathway regulates CD38 on myeloma cells in the bone 
marrow microenvironment: therapeutic implications. Blood, 2020. 136(20): p. 2334-
2345. 

177. Saltarella, I., et al., Mechanisms of Resistance to Anti-CD38 Daratumumab in Multiple 
Myeloma. Cells, 2020. 9(1). 

178. van de Donk, N. and S.Z. Usmani, CD38 Antibodies in Multiple Myeloma: Mechanisms 
of Action and Modes of Resistance. Front Immunol, 2018. 9: p. 2134. 

179. Zambello, R., et al., NK cells and CD38: Implication for (Immuno)Therapy in Plasma Cell 
Dyscrasias. Cells, 2020. 9(3). 

180. Wang, Y., et al., Fratricide of NK Cells in Daratumumab Therapy for Multiple Myeloma 
Overcome by Ex Vivo-Expanded Autologous NK Cells. Clin Cancer Res, 2018. 24(16): p. 
4006-4017. 

181. Kind, S., et al., Prevalence of Syndecan-1 (CD138) Expression in Different Kinds of 
Human Tumors and Normal Tissues. Dis Markers, 2019. 2019: p. 4928315. 

182. Akhmetzyanova, I., et al., Dynamic CD138 surface expression regulates switch between 
myeloma growth and dissemination. Leukemia, 2020. 34(1): p. 245-256. 

183. Kawano, Y., et al., Multiple myeloma cells expressing low levels of CD138 have an 
immature phenotype and reduced sensitivity to lenalidomide. Int J Oncol, 2012. 41(3): 
p. 876-84. 



BIBLIOGRAPHY 

224 
 

184. Beauvais, D.M., et al., Syndecan-1 (CD138) Suppresses Apoptosis in Multiple Myeloma 
by Activating IGF1 Receptor: Prevention by SynstatinIGF1R Inhibits Tumor Growth. 
Cancer Res, 2016. 76(17): p. 4981-93. 

185. Lovell, R., et al., Soluble syndecan-1 level at diagnosis is an independent prognostic 
factor in multiple myeloma and the extent of fall from diagnosis to plateau predicts for 
overall survival. Br J Haematol, 2005. 130(4): p. 542-8. 

186. Rapraeger, A.C., Syndecan-regulated receptor signaling. J Cell Biol, 2000. 149(5): p. 995-
8. 

187. Coombe, D.R., Biological implications of glycosaminoglycan interactions with 
haemopoietic cytokines. Immunol Cell Biol, 2008. 86(7): p. 598-607. 

188. Casu, B., A. Naggi, and G. Torri, Heparin-derived heparan sulfate mimics to modulate 
heparan sulfate-protein interaction in inflammation and cancer. Matrix Biol, 2010. 
29(6): p. 442-52. 

189. Moreaux, J., et al., APRIL and TACI interact with syndecan-1 on the surface of multiple 
myeloma cells to form an essential survival loop. Eur J Haematol, 2009. 83(2): p. 119-
29. 

190. Heider, K.H., et al., CD44v6: a target for antibody-based cancer therapy. Cancer 
Immunol Immunother, 2004. 53(7): p. 567-79. 

191. Casucci, M., et al., CD44v6-targeted T cells mediate potent antitumor effects against 
acute myeloid leukemia and multiple myeloma. Blood, 2013. 122(20): p. 3461-72. 

192. Todaro, M., et al., CD44v6 is a marker of constitutive and reprogrammed cancer stem 
cells driving colon cancer metastasis. Cell Stem Cell, 2014. 14(3): p. 342-56. 

193. Jung, T., W. Gross, and M. Zoller, CD44v6 coordinates tumor matrix-triggered motility 
and apoptosis resistance. J Biol Chem, 2011. 286(18): p. 15862-74. 

194. Ponta, H., L. Sherman, and P.A. Herrlich, CD44: from adhesion molecules to signalling 
regulators. Nat Rev Mol Cell Biol, 2003. 4(1): p. 33-45. 

195. Shirure, V.S., et al., CD44 variant isoforms expressed by breast cancer cells are 
functional E-selectin ligands under flow conditions. Am J Physiol Cell Physiol, 2015. 
308(1): p. C68-78. 

196. Jijiwa, M., et al., CD44v6 regulates growth of brain tumor stem cells partially through 
the AKT-mediated pathway. PLoS One, 2011. 6(9): p. e24217. 

197. Khan, F., et al., Identification of novel CD44v6-binding peptides that block CD44v6 and 
deliver a pro-apoptotic peptide to tumors to inhibit tumor growth and metastasis in 
mice. Theranostics, 2021. 11(3): p. 1326-1344. 

198. Wang, Z., et al., CD44/CD44v6 a Reliable Companion in Cancer-Initiating Cell 
Maintenance and Tumor Progression. Front Cell Dev Biol, 2018. 6: p. 97. 

199. Riechelmann, H., et al., Phase I trial with the CD44v6-targeting immunoconjugate 
bivatuzumab mertansine in head and neck squamous cell carcinoma. Oral Oncol, 2008. 
44(9): p. 823-9. 

200. Awuah, D., et al., Developing a Safer Anti-CD44v6 Chimeric Antigen Receptor T Cell 
Against Hematological Cancers By Mitigating on-Target Off-Tumor Toxicity. Blood, 
2021. 138: p. 2796. 

201. Braumuller, H., et al., CD44v6 cell surface expression is a common feature of 
macrophages and macrophage-like cells - implication for a natural macrophage 
extravasation mechanism mimicked by tumor cells. FEBS Lett, 2000. 476(3): p. 240-7. 

202. Bruno, B., et al., European Myeloma Network perspective on CAR T-Cell therapies for 
multiple myeloma. Haematologica, 2021. 106(8): p. 2054-2065. 

203. Kumaresan, P.R., et al., CS1, a novel member of the CD2 family, is homophilic and 
regulates NK cell function. Mol Immunol, 2002. 39(1-2): p. 1-8. 

204. Chan, W.K., et al., A CS1-NKG2D Bispecific Antibody Collectively Activates Cytolytic 
Immune Cells against Multiple Myeloma. Cancer Immunol Res, 2018. 6(7): p. 776-787. 



BIBLIOGRAPHY 

225 
 

205. Malaer, J.D. and P.A. Mathew, CS1 (SLAMF7, CD319) is an effective immunotherapeutic 
target for multiple myeloma. Am J Cancer Res, 2017. 7(8): p. 1637-1641. 

206. Chu, J., et al., CS1-specific chimeric antigen receptor (CAR)-engineered natural killer 
cells enhance in vitro and in vivo antitumor activity against human multiple myeloma. 
Leukemia, 2014. 28(4): p. 917-27. 

207. O'Neal, J., et al., CS1 CAR-T targeting the distal domain of CS1 (SLAMF7) shows efficacy 
in high tumor burden myeloma model despite fratricide of CD8+CS1 expressing CAR-T 
cells. Leukemia, 2022. 36(6): p. 1625-1634. 

208. Li, X., et al., CD19, from bench to bedside. Immunol Lett, 2017. 183: p. 86-95. 
209. Barsan, V., et al., Tisagenlecleucel utilisation and outcomes across refractory, first 

relapse and multiply relapsed B-cell acute lymphoblastic leukemia: a retrospective 
analysis of real-world patterns. EClinicalMedicine, 2023. 65: p. 102268. 

210. Schuster, S.J., et al., Tisagenlecleucel in Adult Relapsed or Refractory Diffuse Large B-
Cell Lymphoma. N Engl J Med, 2019. 380(1): p. 45-56. 

211. Wang, Y., et al., Long-Term Follow-Up of Combination of B-Cell Maturation Antigen and 
CD19 Chimeric Antigen Receptor T Cells in Multiple Myeloma. J Clin Oncol, 2022. 
40(20): p. 2246-2256. 

212. Garfall, A.L., et al., Anti-CD19 CAR T cells with high-dose melphalan and autologous 
stem cell transplantation for refractory multiple myeloma. JCI Insight, 2018. 3(8). 

213. Wang, Y., et al., BCMA-targeting Bispecific Antibody That Simultaneously Stimulates 
NKG2D-enhanced Efficacy Against Multiple Myeloma. J Immunother, 2020. 43(6): p. 
175-188. 

214. Shi, X., et al., Anti-CD19 and anti-BCMA CAR T cell therapy followed by lenalidomide 
maintenance after autologous stem-cell transplantation for high-risk newly diagnosed 
multiple myeloma. Am J Hematol, 2022. 97(5): p. 537-547. 

215. Radhakrishnan, S.V., et al., CD229 CAR T cells eliminate multiple myeloma and tumor 
propagating cells without fratricide. Nat Commun, 2020. 11(1): p. 798. 

216. Hosen, N., et al., The activated conformation of integrin beta(7) is a novel multiple 
myeloma-specific target for CAR T cell therapy. Nat Med, 2017. 23(12): p. 1436-1443. 

217. Larson, R.C., et al., Anti-TACI single and dual-targeting CAR T cells overcome BCMA 
antigen loss in multiple myeloma. Nat Commun, 2023. 14(1): p. 7509. 

218. Anderson, G.S.F., et al., Unbiased cell surface proteomics identifies SEMA4A as an 
effective immunotherapy target for myeloma. Blood, 2022. 139(16): p. 2471-2482. 

219. Soekojo, C.Y., et al., Immunotherapy in Multiple Myeloma. Cells, 2020. 9(3). 
220. Krejcik, J., et al., Daratumumab depletes CD38+ immune regulatory cells, promotes T-

cell expansion, and skews T-cell repertoire in multiple myeloma. Blood, 2016. 128(3): p. 
384-94. 

221. Dimopoulos, M.A., et al., Overall Survival With Daratumumab, Lenalidomide, and 
Dexamethasone in Previously Treated Multiple Myeloma (POLLUX): A Randomized, 
Open-Label, Phase III Trial. J Clin Oncol, 2023. 41(8): p. 1590-1599. 

222. Sonneveld, P., et al., Overall Survival With Daratumumab, Bortezomib, and 
Dexamethasone in Previously Treated Multiple Myeloma (CASTOR): A Randomized, 
Open-Label, Phase III Trial. J Clin Oncol, 2023. 41(8): p. 1600-1609. 

223. Nijhof, I.S., et al., CD38 expression and complement inhibitors affect response and 
resistance to daratumumab therapy in myeloma. Blood, 2016. 128(7): p. 959-70. 

224. Naeimi Kararoudi, M., et al., CD38 deletion of human primary NK cells eliminates 
daratumumab-induced fratricide and boosts their effector activity. Blood, 2020. 
136(21): p. 2416-2427. 

225. Verkleij, C.P.M., et al., NK Cell Phenotype Is Associated With Response and Resistance to 
Daratumumab in Relapsed/Refractory Multiple Myeloma. Hemasphere, 2023. 7(5): p. 
e881. 



BIBLIOGRAPHY 

226 
 

226. De Luca, F., et al., Monoclonal Antibodies: The Greatest Resource to Treat Multiple 
Myeloma. Int J Mol Sci, 2023. 24(4). 

227. Dhillon, S., Isatuximab: First Approval. Drugs, 2020. 80(9): p. 905-912. 
228. Martin, T., et al., Isatuximab, carfilzomib, and dexamethasone in patients with relapsed 

multiple myeloma: updated results from IKEMA, a randomized Phase 3 study. Blood 
Cancer J, 2023. 13(1): p. 72. 

229. Moreau, P., et al., Isatuximab, carfilzomib, and dexamethasone in relapsed multiple 
myeloma (IKEMA): a multicentre, open-label, randomised phase 3 trial. Lancet, 2021. 
397(10292): p. 2361-2371. 

230. Moore, D.C., et al., Hepatitis B reactivation in patients with multiple myeloma treated 
with anti-CD38 monoclonal antibody-based therapies: a pharmacovigilance analysis. 
Int J Clin Pharm, 2023. 

231. Gozzetti, A., et al., Anti CD38 monoclonal antibodies for multiple myeloma treatment. 
Hum Vaccin Immunother, 2022. 18(5): p. 2052658. 

232. Hiemstra, I.H., et al., Preclinical anti-tumour activity of HexaBody-CD38, a next-
generation CD38 antibody with superior complement-dependent cytotoxic activity. 
EBioMedicine, 2023. 93: p. 104663. 

233. Magen, H. and E. Muchtar, Elotuzumab: the first approved monoclonal antibody for 
multiple myeloma treatment. Ther Adv Hematol, 2016. 7(4): p. 187-95. 

234. Tai, Y.T., et al., Anti-CS1 humanized monoclonal antibody HuLuc63 inhibits myeloma cell 
adhesion and induces antibody-dependent cellular cytotoxicity in the bone marrow 
milieu. Blood, 2008. 112(4): p. 1329-37. 

235. Ritchie, D. and M. Colonna, Mechanisms of Action and Clinical Development of 
Elotuzumab. Clin Transl Sci, 2018. 11(3): p. 261-266. 

236. Liu, Y.C., S. Szmania, and F. van Rhee, Profile of elotuzumab and its potential in the 
treatment of multiple myeloma. Blood Lymphat Cancer, 2014. 2014(4): p. 15-27. 

237. Dimopoulos, M.A., et al., Elotuzumab Plus Pomalidomide and Dexamethasone for 
Relapsed/Refractory Multiple Myeloma: Final Overall Survival Analysis From the 
Randomized Phase II ELOQUENT-3 Trial. J Clin Oncol, 2023. 41(3): p. 568-578. 

238. Ray, U. and R.Z. Orlowski, Antibody-Drug Conjugates for Multiple Myeloma: Just the 
Beginning, or the Beginning of the End? Pharmaceuticals (Basel), 2023. 16(4). 

239. Markham, A., Belantamab Mafodotin: First Approval. Drugs, 2020. 80(15): p. 1607-
1613. 

240. Lonial, S., et al., Belantamab mafodotin for relapsed or refractory multiple myeloma 
(DREAMM-2): a two-arm, randomised, open-label, phase 2 study. Lancet Oncol, 2020. 
21(2): p. 207-221. 

241. Katja Weisel, V.T.H., Atanas Radinoff, Sosana Delimpasi, Gabor Mikala, Tamas Masszi,, 
M.C. Jian Li, Morio Matsumoto, Neal Sule, Mary Li, Astrid McKeown, Wei He, Shelley 
Bright,, and J.B. Brooke Currie, Joanna Opalinska, Meletios A. Dimopoulos, A phase 3, 
open-label, randomized study to evaluate the efficacy and safety of singleagent 
belantamab mafodotin (belamaf) compared to pomalidomide plus low-dose 
dexamethasone (Pd) in patients (pts) with relapsed/refractory multiple myeloma 
(RRMM): DREAMM-3. Journal of Clinical Oncology, 2023. 

242. Martin, T.G., et al., Cilta-cel, a BCMA-targeting CAR-T therapy for heavily pretreated 
patients with relapsed/refractory multiple myeloma. Future Oncol, 2023. 

243. de la Rubia, J., et al., Belantamab Mafodotin in Patients with Relapsed/Refractory 
Multiple Myeloma: Results of the Compassionate Use or the Expanded Access Program 
in Spain. Cancers (Basel), 2023. 15(11). 

244. Abramson, H.N., Immunotherapy of Multiple Myeloma: Current Status as Prologue to 
the Future. Int J Mol Sci, 2023. 24(21). 

245. Cho, S.F., et al., Bispecific antibodies in multiple myeloma treatment: A journey in 
progress. Front Oncol, 2022. 12: p. 1032775. 



BIBLIOGRAPHY 

227 
 

246. Pillarisetti, K., et al., Teclistamab is an active T cell-redirecting bispecific antibody 
against B-cell maturation antigen for multiple myeloma. Blood Adv, 2020. 4(18): p. 
4538-4549. 

247. Verkleij, C.P.M., et al., Preclinical activity and determinants of response of the 
GPRC5DxCD3 bispecific antibody talquetamab in multiple myeloma. Blood Adv, 2021. 
5(8): p. 2196-2215. 

248. Zhao, J., et al., Bispecific antibodies targeting BCMA, GPRC5D, and FcRH5 for multiple 
myeloma therapy: latest updates from ASCO 2023 Annual Meeting. J Hematol Oncol, 
2023. 16(1): p. 92. 

249. Li, J., et al., Membrane-Proximal Epitope Facilitates Efficient T Cell Synapse Formation 
by Anti-FcRH5/CD3 and Is a Requirement for Myeloma Cell Killing. Cancer Cell, 2017. 
31(3): p. 383-395. 

250. Suzanne Trudel, A.D.C., Amrita Y. Krishnan, Rafael Fonseca, Andrew Spencer, Jesús G. 
Berdeja, Alexander Lesokhin, Peter A Forsberg, Jacob P. Laubach, Luciano J. Costa, 
Paula Rodriguez-Otero, Rayan Kaedbey, Joshua Richter, Maria-Victoria Mateos, Sheeba 
K Thomas, Chihunt Wong, Mengsong Li, Voleak Choeurng, Anjali Vaze, Divya Samineni, 
Teiko Sumiyoshi, James Cooper, Simon J Harrison, Cevostamab Monotherapy Continues 
to Show Clinically Meaningful Activity and Manageable Safety in Patients with Heavily 
Pre-Treated Relapsed/Refractory Multiple Myeloma (RRMM): Updated Results from an 
Ongoing Phase I Study. ASH 2021, 2021. 

251. Rahman, M.M. and G. McFadden, Oncolytic Viruses: Newest Frontier for Cancer 
Immunotherapy. Cancers (Basel), 2021. 13(21). 

252. Cook, J., et al., Oncolytic virotherapy - Forging its place in the immunomodulatory 
paradigm for Multiple Myeloma. Cancer Treat Res Commun, 2021. 29: p. 100473. 

253. Stewart, G., A. Chantry, and M. Lawson, The Use of Oncolytic Viruses in the Treatment 
of Multiple Myeloma. Cancers (Basel), 2021. 13(22). 

254. Boussi, L.S., Z.M. Avigan, and J. Rosenblatt, Immunotherapy for the treatment of 
multiple myeloma. Front Immunol, 2022. 13: p. 1027385. 

255. Leivas, A., et al., Novel treatment strategy with autologous activated and expanded 
natural killer cells plus anti-myeloma drugs for multiple myeloma. Oncoimmunology, 
2016. 5(12): p. e1250051. 

256. Verheye, E., et al., Dendritic Cell-Based Immunotherapy in Multiple Myeloma: 
Challenges, Opportunities, and Future Directions. Int J Mol Sci, 2022. 23(2). 

257. Daher, M. and K. Rezvani, Outlook for New CAR-Based Therapies with a Focus on CAR 
NK Cells: What Lies Beyond CAR-Engineered T Cells in the Race against Cancer. Cancer 
Discov, 2021. 11(1): p. 45-58. 

258. Tokarew, N., et al., Teaching an old dog new tricks: next-generation CAR T cells. Br J 
Cancer, 2019. 120(1): p. 26-37. 

259. Mullard, A., FDA approves first CAR T therapy. Nat Rev Drug Discov, 2017. 16(10): p. 
669. 

260. Maude, S.L., et al., Tisagenlecleucel in Children and Young Adults with B-Cell 
Lymphoblastic Leukemia. N Engl J Med, 2018. 378(5): p. 439-448. 

261. Chen, Y.J., B. Abila, and Y. Mostafa Kamel, CAR-T: What Is Next? Cancers (Basel), 2023. 
15(3). 

262. Berdeja, J.G., et al., Ciltacabtagene autoleucel, a B-cell maturation antigen-directed 
chimeric antigen receptor T-cell therapy in patients with relapsed or refractory multiple 
myeloma (CARTITUDE-1): a phase 1b/2 open-label study. Lancet, 2021. 398(10297): p. 
314-324. 

263. Mailankody, S. and O. Landgren, T-Cell Engagers - Modern Immune-Based Therapies for 
Multiple Myeloma. N Engl J Med, 2022. 387(6): p. 558-561. 

264. Barber, A., et al., Chimeric NKG2D receptor-expressing T cells as an immunotherapy for 
multiple myeloma. Exp Hematol, 2008. 36(10): p. 1318-28. 



BIBLIOGRAPHY 

228 
 

265. Lonez, C., et al., Study protocol for THINK: a multinational open-label phase I study to 
assess the safety and clinical activity of multiple administrations of NKR-2 in patients 
with different metastatic tumour types. BMJ Open, 2017. 7(11): p. e017075. 

266. Baumeister, S.H., et al., Phase I Trial of Autologous CAR T Cells Targeting NKG2D 
Ligands in Patients with AML/MDS and Multiple Myeloma. Cancer Immunol Res, 2019. 
7(1): p. 100-112. 

267. Rodriguez-Otero, P. and J.F. San-Miguel, Cellular therapy for multiple myeloma: what's 
now and what's next. Hematology Am Soc Hematol Educ Program, 2022. 2022(1): p. 
180-189. 

268. Khan, A.N., et al., Immunogenicity of CAR-T Cell Therapeutics: Evidence, Mechanism 
and Mitigation. Front Immunol, 2022. 13: p. 886546. 

269. Zhu, X., Q. Li, and X. Zhu, Mechanisms of CAR T cell exhaustion and current 
counteraction strategies. Front Cell Dev Biol, 2022. 10: p. 1034257. 

270. Gumber, D. and L.D. Wang, Improving CAR-T immunotherapy: Overcoming the 
challenges of T cell exhaustion. EBioMedicine, 2022. 77: p. 103941. 

271. Roeven, M.W., et al., Immunotherapeutic approaches to treat multiple myeloma. Hum 
Vaccin Immunother, 2014. 10(4): p. 896-910. 

272. Fernandez de Larrea, C., et al., Defining an Optimal Dual-Targeted CAR T-cell Therapy 
Approach Simultaneously Targeting BCMA and GPRC5D to Prevent BCMA Escape-
Driven Relapse in Multiple Myeloma. Blood Cancer Discov, 2020. 1(2): p. 146-154. 

273. Ruffo, E., et al., Post-translational covalent assembly of CAR and synNotch receptors for 
programmable antigen targeting. Nat Commun, 2023. 14(1): p. 2463. 

274. Martin, T., et al., Ciltacabtagene Autoleucel, an Anti-B-cell Maturation Antigen Chimeric 
Antigen Receptor T-Cell Therapy, for Relapsed/Refractory Multiple Myeloma: 
CARTITUDE-1 2-Year Follow-Up. J Clin Oncol, 2023. 41(6): p. 1265-1274. 

275. Zhang, X., et al., CAR-T cell therapy in multiple myeloma: Current limitations and 
potential strategies. Front Immunol, 2023. 14: p. 1101495. 

276. Chen, D., et al., Subsequent anti-myeloma therapy after maturation antigen (BCMA) 
chimeric antigen receptor (CAR)-T cell (HDS269B) treatment in patients with 
relapsed/refractory multiple myeloma. Am J Hematol, 2022. 97(12): p. E478-E481. 

277. Morris, E.C., et al., Cytokine release syndrome and associated neurotoxicity in cancer 
immunotherapy. Nat Rev Immunol, 2022. 22(2): p. 85-96. 

278. Hines, M.R., et al., Hemophagocytic lymphohistiocytosis-like toxicity (carHLH) after 
CD19-specific CAR T-cell therapy. Br J Haematol, 2021. 194(4): p. 701-707. 

279. Jain, T., T.S. Olson, and F.L. Locke, How I treat cytopenias after CAR T-cell therapy. 
Blood, 2023. 141(20): p. 2460-2469. 

280. Valeri, A., et al., Overcoming tumor resistance mechanisms in CAR-NK cell therapy. 
Front Immunol, 2022. 13: p. 953849. 

281. Sanber, K., B. Savani, and T. Jain, Graft-versus-host disease risk after chimeric antigen 
receptor T-cell therapy: the diametric opposition of T cells. Br J Haematol, 2021. 195(5): 
p. 660-668. 

282. Sivori, S., et al., NK Cell-Based Immunotherapy for Hematological Malignancies. J Clin 
Med, 2019. 8(10). 

283. Shah, N.N. and T.J. Fry, Mechanisms of resistance to CAR T cell therapy. Nat Rev Clin 
Oncol, 2019. 16(6): p. 372-385. 

284. Marofi, F., et al., Renaissance of armored immune effector cells, CAR-NK cells, brings 
the higher hope for successful cancer therapy. Stem Cell Res Ther, 2021. 12(1): p. 200. 

285. Cozar, B., et al., Tumor-Infiltrating Natural Killer Cells. Cancer Discov, 2021. 11(1): p. 34-
44. 

286. Caruso, S., et al., Safe and effective off-the-shelf immunotherapy based on CAR.CD123-
NK cells for the treatment of acute myeloid leukaemia. J Hematol Oncol, 2022. 15(1): p. 
163. 



BIBLIOGRAPHY 

229 
 

287. Biederstadt, A. and K. Rezvani, Engineering the next generation of CAR-NK 
immunotherapies. Int J Hematol, 2021. 114(5): p. 554-571. 

288. Zhang, Y., et al., In vivo kinetics of human natural killer cells: the effects of ageing and 
acute and chronic viral infection. Immunology, 2007. 121(2): p. 258-65. 

289. Glienke, W., et al., Advantages and applications of CAR-expressing natural killer cells. 
Front Pharmacol, 2015. 6: p. 21. 

290. Burga, R.A., et al., Improving efficacy of cancer immunotherapy by genetic modification 
of natural killer cells. Cytotherapy, 2016. 18(11): p. 1410-1421. 

291. Wlodarczyk, M. and B. Pyrzynska, CAR-NK as a Rapidly Developed and Efficient 
Immunotherapeutic Strategy against Cancer. Cancers (Basel), 2022. 15(1). 

292. Berrien-Elliott, M.M., R. Romee, and T.A. Fehniger, Improving natural killer cell cancer 
immunotherapy. Curr Opin Organ Transplant, 2015. 20(6): p. 671-80. 

293. Gong, Y., et al., Chimeric antigen receptor natural killer (CAR-NK) cell design and 
engineering for cancer therapy. J Hematol Oncol, 2021. 14(1): p. 73. 

294. Rafei, H., M. Daher, and K. Rezvani, Chimeric antigen receptor (CAR) natural killer (NK)-
cell therapy: leveraging the power of innate immunity. Br J Haematol, 2021. 193(2): p. 
216-230. 

295. Cichocki, F., et al., Epigenetic regulation of NK cell differentiation and effector functions. 
Front Immunol, 2013. 4: p. 55. 

296. Berrien-Elliott, M.M., M.T. Jacobs, and T.A. Fehniger, Allogeneic natural killer cell 
therapy. Blood, 2023. 141(8): p. 856-868. 

297. Jonsson, A.H. and W.M. Yokoyama, Natural killer cell tolerance licensing and other 
mechanisms. Adv Immunol, 2009. 101: p. 27-79. 

298. Debska-Zielkowska, J., et al., KIR Receptors as Key Regulators of NK Cells Activity in 
Health and Disease. Cells, 2021. 10(7). 

299. Zhang, Y., et al., KIR-HLA interactions extend human CD8+ T cell lifespan in vivo. J Clin 
Invest, 2023. 133(12). 

300. Zhong, X., et al., PD-L2 expression extends beyond dendritic cells/macrophages to B1 
cells enriched for V(H)11/V(H)12 and phosphatidylcholine binding. Eur J Immunol, 
2007. 37(9): p. 2405-10. 

301. Yearley, J.H., et al., PD-L2 Expression in Human Tumors: Relevance to Anti-PD-1 Therapy 
in Cancer. Clin Cancer Res, 2017. 23(12): p. 3158-3167. 

302. Ghosh, C., G. Luong, and Y. Sun, A snapshot of the PD-1/PD-L1 pathway. J Cancer, 2021. 
12(9): p. 2735-2746. 

303. Harjunpaa, H. and C. Guillerey, TIGIT as an emerging immune checkpoint. Clin Exp 
Immunol, 2020. 200(2): p. 108-119. 

304. Cifaldi, L., et al., DNAM-1 chimeric receptor-engineered NK cells: a new frontier for CAR-
NK cell-based immunotherapy. Front Immunol, 2023. 14: p. 1197053. 

305. Wolf, Y., A.C. Anderson, and V.K. Kuchroo, TIM3 comes of age as an inhibitory receptor. 
Nat Rev Immunol, 2020. 20(3): p. 173-185. 

306. Lamb, M.G., et al., Natural killer cell therapy for hematologic malignancies: successes, 
challenges, and the future. Stem Cell Res Ther, 2021. 12(1): p. 211. 

307. Borys, S.M., et al., The Yin and Yang of Targeting KLRG1(+) Tregs and Effector Cells. 
Front Immunol, 2022. 13: p. 894508. 

308. Henson, S.M. and A.N. Akbar, KLRG1--more than a marker for T cell senescence. Age 
(Dordr), 2009. 31(4): p. 285-91. 

309. Borrego, F., et al., The CD94/NKG2 family of receptors: from molecules and cells to 
clinical relevance. Immunol Res, 2006. 35(3): p. 263-78. 

310. Kaiser, B.K., et al., Interactions between NKG2x immunoreceptors and HLA-E ligands 
display overlapping affinities and thermodynamics. J Immunol, 2005. 174(5): p. 2878-
84. 



BIBLIOGRAPHY 

230 
 

311. Medjouel Khlifi, H., et al., Role of the ITAM-Bearing Receptors Expressed by Natural 
Killer Cells in Cancer. Front Immunol, 2022. 13: p. 898745. 

312. de Andrade, L.F., M.J. Smyth, and L. Martinet, DNAM-1 control of natural killer cells 
functions through nectin and nectin-like proteins. Immunol Cell Biol, 2014. 92(3): p. 
237-44. 

313. Sanchez-Correa, B., et al., DNAM-1 and the TIGIT/PVRIG/TACTILE Axis: Novel Immune 
Checkpoints for Natural Killer Cell-Based Cancer Immunotherapy. Cancers (Basel), 2019. 
11(6). 

314. Rezvani, K., et al., Engineering Natural Killer Cells for Cancer Immunotherapy. Mol Ther, 
2017. 25(8): p. 1769-1781. 

315. Guillerey, C., N.D. Huntington, and M.J. Smyth, Targeting natural killer cells in cancer 
immunotherapy. Nat Immunol, 2016. 17(9): p. 1025-36. 

316. Chiossone, L. and E. Vivier, Bringing natural killer cells to the clinic. J Exp Med, 2022. 
219(10). 

317. Klingemann, H., Are natural killer cells superior CAR drivers? Oncoimmunology, 2014. 3: 
p. e28147. 

318. Kang, L., et al., Characterization and ex vivo Expansion of Human Placenta-Derived 
Natural Killer Cells for Cancer Immunotherapy. Front Immunol, 2013. 4: p. 101. 

319. Wen, W., et al., Enhancing cord blood stem cell-derived NK cell growth and 
differentiation through hyperosmosis. Stem Cell Res Ther, 2023. 14(1): p. 295. 

320. Guo, X., et al., CBLB ablation with CRISPR/Cas9 enhances cytotoxicity of human 
placental stem cell-derived NK cells for cancer immunotherapy. J Immunother Cancer, 
2021. 9(3). 

321. Schmidt, D., et al., Engineering CAR-NK cells: how to tune innate killer cells for cancer 
immunotherapy. Immunother Adv, 2022. 2(1): p. ltac003. 

322. Zhang, L., et al., Natural killer cells: of-the-shelf cytotherapy for cancer 
immunosurveillance. Am J Cancer Res, 2021. 11(4): p. 1770-1791. 

323. Hermanson, D.L. and D.S. Kaufman, Utilizing chimeric antigen receptors to direct 
natural killer cell activity. Front Immunol, 2015. 6: p. 195. 

324. Sutlu, T., et al., Inhibition of intracellular antiviral defense mechanisms augments 
lentiviral transduction of human natural killer cells: implications for gene therapy. Hum 
Gene Ther, 2012. 23(10): p. 1090-100. 

325. Herrera, L., et al., Adult peripheral blood and umbilical cord blood NK cells are good 
sources for effective CAR therapy against CD19 positive leukemic cells. Sci Rep, 2019. 
9(1): p. 18729. 

326. Liu, E., et al., Use of CAR-Transduced Natural Killer Cells in CD19-Positive Lymphoid 
Tumors. N Engl J Med, 2020. 382(6): p. 545-553. 

327. Shin, M.H., et al., NK Cell-Based Immunotherapies in Cancer. Immune Netw, 2020. 
20(2): p. e14. 

328. Zhang, L., et al., CAR-NK cells for cancer immunotherapy: from bench to bedside. 
Biomark Res, 2022. 10(1): p. 12. 

329. Wu, X. and S. Matosevic, Gene-edited and CAR-NK cells: Opportunities and challenges 
with engineering of NK cells for immunotherapy. Mol Ther Oncolytics, 2022. 27: p. 224-
238. 

330. Jiang, H., et al., Transfection of chimeric anti-CD138 gene enhances natural killer cell 
activation and killing of multiple myeloma cells. Mol Oncol, 2014. 8(2): p. 297-310. 

331. Roex, G., et al., Two for one: targeting BCMA and CD19 in B-cell malignancies with off-
the-shelf dual-CAR NK-92 cells. J Transl Med, 2022. 20(1): p. 124. 

332. Luanpitpong, S., et al., Selective Cytotoxicity of Single and Dual Anti-CD19 and Anti-
CD138 Chimeric Antigen Receptor-Natural Killer Cells against Hematologic 
Malignancies. J Immunol Res, 2021. 2021: p. 5562630. 



BIBLIOGRAPHY 

231 
 

333. Motais, B., et al., Anti-BCMA-CAR NK Cells Expressing Soluble TRAIL: Promising 
Therapeutic Approach for Multiple Myeloma in Combination with Bortezomib and γ-
Secretase Inhibitors. ASH, 2022. 

334. Golubovskaya, V., et al., CAR-NK Cells Generated with mRNA-LNPs Kill Tumor Target 
Cells In Vitro and In Vivo. Int J Mol Sci, 2023. 24(17). 

335. Ng, Y.Y., et al., CXCR4 and anti-BCMA CAR co-modified natural killer cells suppress 
multiple myeloma progression in a xenograft mouse model. Cancer Gene Ther, 2022. 
29(5): p. 475-483. 

336. Wang, X., et al., Inducible MyD88/CD40 synergizes with IL-15 to enhance antitumor 
efficacy of CAR-NK cells. Blood Adv, 2020. 4(9): p. 1950-1964. 

337. Karvouni, M., et al., Challenges in alphaCD38-chimeric antigen receptor (CAR)-
expressing natural killer (NK) cell-based immunotherapy in multiple myeloma: 
Harnessing the CD38dim phenotype of cytokine-stimulated NK cells as a strategy to 
prevent fratricide. Cytotherapy, 2023. 25(7): p. 763-772. 

338. Brophy, S., et al., CB derived, optimized affinity CD38 CAR-NK cells with CD38 KO show 
promising in-vivo activity in a Multiple Myeloma model. IMS 2022, 2022. 

339. Reiser, J., et al., FT555: Off-the-Shelf CAR-NK Cell Therapy Co-Targeting GPRC5D and 
CD38 for the Treatment of Multiple Myeloma. ASH, 2022. 

340. Cichocki, F., et al., Quadruple gene-engineered natural killer cells enable multi-antigen 
targeting for durable antitumor activity against multiple myeloma. Nat Commun, 2022. 
13(1): p. 7341. 

341. Dhakal, B., et al., Interim phase I clinical data of FT576 as monotherapy and in 
combination with daratumumab in subjects with relapsed/refractory multiple 
myeloma. ASH, 2022. 

342. Li, Y., et al., KIR-based inhibitory CARs overcome CAR-NK cell trogocytosis-mediated 
fratricide and tumor escape. Nat Med, 2022. 28(10): p. 2133-2144. 

343. Cichocki, F., et al., GSK3 Inhibition Drives Maturation of NK Cells and Enhances Their 
Antitumor Activity. Cancer Res, 2017. 77(20): p. 5664-5675. 

344. Afolabi, L.O., et al., Synergistic Tumor Cytolysis by NK Cells in Combination With a Pan-
HDAC Inhibitor, Panobinostat. Front Immunol, 2021. 12: p. 701671. 

345. Chen, X., et al., A combinational therapy of EGFR-CAR NK cells and oncolytic herpes 
simplex virus 1 for breast cancer brain metastases. Oncotarget, 2016. 7(19): p. 27764-
77. 

346. Xiao, X., et al., Bispecific NK-cell engager targeting BCMA elicits stronger antitumor 
effects and produces less proinflammatory cytokines than T-cell engager. Front 
Immunol, 2023. 14: p. 1113303. 

347. Zhang, C., et al., Bispecific antibody-mediated redirection of NKG2D-CAR natural killer 
cells facilitates dual targeting and enhances antitumor activity. J Immunother Cancer, 
2021. 9(10). 

348. McWilliams, E.M., et al., Therapeutic CD94/NKG2A blockade improves natural killer cell 
dysfunction in chronic lymphocytic leukemia. Oncoimmunology, 2016. 5(10): p. 
e1226720. 

349. Bexte, T., et al., CRISPR-Cas9 based gene editing of the immune checkpoint NKG2A 
enhances NK cell mediated cytotoxicity against multiple myeloma. Oncoimmunology, 
2022. 11(1): p. 2081415. 

350. Daher, M., et al., Targeting a cytokine checkpoint enhances the fitness of armored cord 
blood CAR-NK cells. Blood, 2021. 137(5): p. 624-636. 

351. Benson, D.M., Jr., et al., IPH2101, a novel anti-inhibitory KIR antibody, and lenalidomide 
combine to enhance the natural killer cell versus multiple myeloma effect. Blood, 2011. 
118(24): p. 6387-91. 



BIBLIOGRAPHY 

232 
 

352. Nijhof, I.S., et al., Daratumumab-mediated lysis of primary multiple myeloma cells is 
enhanced in combination with the human anti-KIR antibody IPH2102 and lenalidomide. 
Haematologica, 2015. 100(2): p. 263-8. 

353. Benson, D.M., Jr., et al., A phase 1 trial of the anti-KIR antibody IPH2101 in patients 
with relapsed/refractory multiple myeloma. Blood, 2012. 120(22): p. 4324-33. 

354. Carlsten, M., et al., Checkpoint Inhibition of KIR2D with the Monoclonal Antibody 
IPH2101 Induces Contraction and Hyporesponsiveness of NK Cells in Patients with 
Myeloma. Clin Cancer Res, 2016. 22(21): p. 5211-5222. 

355. Laskowski, T.J., A. Biederstadt, and K. Rezvani, Natural killer cells in antitumour 
adoptive cell immunotherapy. Nat Rev Cancer, 2022. 22(10): p. 557-575. 

356. Benson, D.M., Jr., et al., A Phase I Trial of the Anti-KIR Antibody IPH2101 and 
Lenalidomide in Patients with Relapsed/Refractory Multiple Myeloma. Clin Cancer Res, 
2015. 21(18): p. 4055-61. 

357. Judge, S.J., et al., Minimal PD-1 expression in mouse and human NK cells under diverse 
conditions. J Clin Invest, 2020. 130(6): p. 3051-3068. 

358. Yang, C., et al., Dual-targeted CAR-NK cell therapy: optimized CAR design to prevent 
antigen escape and elicit a deep and durable response in multiple myeloma. Cancer 
Research, 2023. 

359. Benson, D.M., Jr., et al., The PD-1/PD-L1 axis modulates the natural killer cell versus 
multiple myeloma effect: a therapeutic target for CT-011, a novel monoclonal anti-PD-1 
antibody. Blood, 2010. 116(13): p. 2286-94. 

360. Badros, A.Z., et al., Long-term remissions after stopping pembrolizumab for relapsed or 
refractory multiple myeloma. Blood Adv, 2019. 3(11): p. 1658-1660. 

361. D'Souza, A., et al., A Phase 2 Study of Pembrolizumab during Lymphodepletion after 
Autologous Hematopoietic Cell Transplantation for Multiple Myeloma. Biol Blood 
Marrow Transplant, 2019. 25(8): p. 1492-1497. 

362. Mateos, M.V., et al., Pembrolizumab plus pomalidomide and dexamethasone for 
patients with relapsed or refractory multiple myeloma (KEYNOTE-183): a randomised, 
open-label, phase 3 trial. Lancet Haematol, 2019. 6(9): p. e459-e469. 

363. Liu, Z.Y., et al., CD155/TIGIT signalling plays a vital role in the regulation of bone 
marrow mesenchymal stem cell-induced natural killer-cell exhaustion in multiple 
myeloma. Clin Transl Med, 2022. 12(7): p. e861. 

364. Jo, D.H., et al., Simultaneous engineering of natural killer cells for CAR transgenesis and 
CRISPR-Cas9 knockout using retroviral particles. Mol Ther Methods Clin Dev, 2023. 29: 
p. 173-184. 

365. Brauneck, F., et al., Combined Blockade of TIGIT and CD39 or A2AR Enhances NK-92 
Cell-Mediated Cytotoxicity in AML. Int J Mol Sci, 2021. 22(23). 

366. Maas, R.J., et al., TIGIT blockade enhances functionality of peritoneal NK cells with 
altered expression of DNAM-1/TIGIT/CD96 checkpoint molecules in ovarian cancer. 
Oncoimmunology, 2020. 9(1): p. 1843247. 

367. Zhang, Q., et al., Blockade of the checkpoint receptor TIGIT prevents NK cell exhaustion 
and elicits potent anti-tumor immunity. Nat Immunol, 2018. 19(7): p. 723-732. 

368. Jiang, W., et al., Tim-3 Blockade Elicits Potent Anti-Multiple Myeloma Immunity of 
Natural Killer Cells. Front Oncol, 2022. 12: p. 739976. 

369. Rakova, J., et al., TIM-3 levels correlate with enhanced NK cell cytotoxicity and 
improved clinical outcome in AML patients. Oncoimmunology, 2021. 10(1): p. 1889822. 

370. Gleason, M.K., et al., Tim-3 is an inducible human natural killer cell receptor that 
enhances interferon gamma production in response to galectin-9. Blood, 2012. 
119(13): p. 3064-72. 

371. Ulbrecht, M., et al., Cell surface expression of HLA-E: interaction with human beta2-
microglobulin and allelic differences. Eur J Immunol, 1999. 29(2): p. 537-47. 



BIBLIOGRAPHY 

233 
 

372. Paech, C., et al., HLA-E diversity unfolded: Identification and characterization of 170 
novel HLA-E alleles. HLA, 2021. 97(5): p. 389-398. 

373. Braud, V.M., et al., HLA-E binds to natural killer cell receptors CD94/NKG2A, B and C. 
Nature, 1998. 391(6669): p. 795-9. 

374. Ulbrecht, M., et al., Interaction of HLA-E with peptides and the peptide transporter in 
vitro: implications for its function in antigen presentation. J Immunol, 1998. 160(9): p. 
4375-85. 

375. Borst, L., S.H. van der Burg, and T. van Hall, The NKG2A-HLA-E Axis as a Novel 
Checkpoint in the Tumor Microenvironment. Clin Cancer Res, 2020. 26(21): p. 5549-
5556. 

376. van Hall, T., et al., Monalizumab: inhibiting the novel immune checkpoint NKG2A. J 
Immunother Cancer, 2019. 7(1): p. 263. 

377. Altvater, B., et al., HLA-G and HLA-E Immune Checkpoints Are Widely Expressed in 
Ewing Sarcoma but Have Limited Functional Impact on the Effector Functions of 
Antigen-Specific CAR T Cells. Cancers (Basel), 2021. 13(12). 

378. Sarkar, S., et al., Optimal selection of natural killer cells to kill myeloma: the role of HLA-
E and NKG2A. Cancer Immunol Immunother, 2015. 64(8): p. 951-63. 

379. Gooden, M., et al., HLA-E expression by gynecological cancers restrains tumor-
infiltrating CD8(+) T lymphocytes. Proc Natl Acad Sci U S A, 2011. 108(26): p. 10656-61. 

380. Nguyen, S., et al., NK-cell reconstitution after haploidentical hematopoietic stem-cell 
transplantations: immaturity of NK cells and inhibitory effect of NKG2A override GvL 
effect. Blood, 2005. 105(10): p. 4135-42. 

381. Gross, C., et al., An Hsp70 peptide initiates NK cell killing of leukemic blasts after stem 
cell transplantation. Leuk Res, 2008. 32(4): p. 527-34. 

382. Derre, L., et al., Expression and release of HLA-E by melanoma cells and melanocytes: 
potential impact on the response of cytotoxic effector cells. J Immunol, 2006. 177(5): p. 
3100-7. 

383. Liu, X., et al., Immune checkpoint HLA-E:CD94-NKG2A mediates evasion of circulating 
tumor cells from NK cell surveillance. Cancer Cell, 2023. 41(2): p. 272-287 e9. 

384. Talebian Yazdi, M., et al., The positive prognostic effect of stromal CD8+ tumor-
infiltrating T cells is restrained by the expression of HLA-E in non-small cell lung 
carcinoma. Oncotarget, 2016. 7(3): p. 3477-88. 

385. de Kruijf, E.M., et al., HLA-E and HLA-G expression in classical HLA class I-negative 
tumors is of prognostic value for clinical outcome of early breast cancer patients. J 
Immunol, 2010. 185(12): p. 7452-9. 

386. Andersson, E., et al., Non-classical HLA-class I expression in serous ovarian carcinoma: 
Correlation with the HLA-genotype, tumor infiltrating immune cells and prognosis. 
Oncoimmunology, 2016. 5(1): p. e1052213. 

387. Lagana, A., et al., Increased HLA-E Expression Correlates with Early Relapse in Multiple 
Myeloma. ASH, 2018. 

388. Kren, L., et al., Expression of immune-modulatory molecules HLA-G and HLA-E by tumor 
cells in glioblastomas: an unexpected prognostic significance? Neuropathology, 2011. 
31(2): p. 129-34. 

389. Bossard, C., et al., HLA-E/beta2 microglobulin overexpression in colorectal cancer is 
associated with recruitment of inhibitory immune cells and tumor progression. Int J 
Cancer, 2012. 131(4): p. 855-63. 

390. Thangaraj, J.L., et al., Expanded natural killer cells augment the antimyeloma effect of 
daratumumab, bortezomib, and dexamethasone in a mouse model. Cell Mol Immunol, 
2021. 18(7): p. 1652-1661. 

391. Yun, H.D., et al., Dinaciclib enhances natural killer cell cytotoxicity against acute 
myelogenous leukemia. Blood Adv, 2019. 3(16): p. 2448-2452. 



BIBLIOGRAPHY 

234 
 

392. Enqvist, M., et al., Selenite induces posttranscriptional blockade of HLA-E expression 
and sensitizes tumor cells to CD94/NKG2A-positive NK cells. J Immunol, 2011. 187(7): p. 
3546-54. 

393. Tognarelli, S., et al., Enhancing the Activation and Releasing the Brakes: A Double Hit 
Strategy to Improve NK Cell Cytotoxicity Against Multiple Myeloma. Front Immunol, 
2018. 9: p. 2743. 

394. Kamiya, T., et al., Blocking expression of inhibitory receptor NKG2A overcomes tumor 
resistance to NK cells. J Clin Invest, 2019. 129(5): p. 2094-2106. 

395. Lee, J., et al., Monalizumab efficacy correlates with HLA-E surface expression and NK 
cell activity in head and neck squamous carcinoma cell lines. J Cancer Res Clin Oncol, 
2023. 149(9): p. 5705-5715. 

396. Wang, X., H. Xiong, and Z. Ning, Implications of NKG2A in immunity and immune-
mediated diseases. Front Immunol, 2022. 13: p. 960852. 

397. Plougastel, B., T. Jones, and J. Trowsdale, Genomic structure, chromosome location, and 
alternative splicing of the human NKG2A gene. Immunogenetics, 1996. 44(4): p. 286-
91. 

398. Wagner, J.A., et al., Cytokine-Induced Memory-Like Differentiation Enhances Unlicensed 
Natural Killer Cell Antileukemia and FcgammaRIIIa-Triggered Responses. Biol Blood 
Marrow Transplant, 2017. 23(3): p. 398-404. 

399. Andre, P., et al., Anti-NKG2A mAb Is a Checkpoint Inhibitor that Promotes Anti-tumor 
Immunity by Unleashing Both T and NK Cells. Cell, 2018. 175(7): p. 1731-1743 e13. 

400. Masilamani, M., et al., CD94/NKG2A inhibits NK cell activation by disrupting the actin 
network at the immunological synapse. J Immunol, 2006. 177(6): p. 3590-6. 

401. Das, J. and S.I. Khakoo, NK cells: tuned by peptide? Immunol Rev, 2015. 267(1): p. 214-
27. 

402. Battin, C., et al., NKG2A-checkpoint inhibition and its blockade critically depends on 
peptides presented by its ligand HLA-E. Immunology, 2022. 166(4): p. 507-521. 

403. Figueiredo, C., A. Seltsam, and R. Blasczyk, Permanent silencing of NKG2A expression 
for cell-based therapeutics. J Mol Med (Berl), 2009. 87(2): p. 199-210. 

404. Ruggeri, L., et al., Effects of anti-NKG2A antibody administration on leukemia and 
normal hematopoietic cells. Haematologica, 2016. 101(5): p. 626-33. 

405. Melero, I., et al., Intratumoral co-injection of NK cells and NKG2A-neutralizing 
monoclonal antibodies. EMBO Mol Med, 2023: p. e17804. 

406. Mahaweni, N.M., et al., NKG2A Expression Is Not per se Detrimental for the Anti-
Multiple Myeloma Activity of Activated Natural Killer Cells in an In Vitro System 
Mimicking the Tumor Microenvironment. Front Immunol, 2018. 9: p. 1415. 

407. Tinker, A.V., et al., Dose-Ranging and Cohort-Expansion Study of Monalizumab 
(IPH2201) in Patients with Advanced Gynecologic Malignancies: A Trial of the Canadian 
Cancer Trials Group (CCTG): IND221. Clin Cancer Res, 2019. 25(20): p. 6052-6060. 

408. Galot, R., et al., A phase II study of monalizumab in patients with recurrent/metastatic 
squamous cell carcinoma of the head and neck: The I1 cohort of the EORTC-HNCG-1559 
UPSTREAM trial. Eur J Cancer, 2021. 158: p. 17-26. 

409. Cohen, R.B., et al., Combination of monalizumab and cetuximab in patients with 
recurrent or metastatic head and neck squamous  cell cancer previously treated with 
platinum-based chemotherapy and PD-(L)1 inhibitors: a phase II expansion study. 
ASCO, 2020. 

410. Colevas, D.A., et al., Monalizumab, cetuximab and durvalumab in first-line treatment of 
recurrent or metastatic squamous cell carcinoma of the head and neck (R/M SCCHN): A 
phase II trial. Annals of Oncology, 2023. 

411. Herbst, R.S., et al., COAST: An Open-Label, Phase II, Multidrug Platform Study of 
Durvalumab Alone or in Combination With Oleclumab or Monalizumab in Patients With 



BIBLIOGRAPHY 

235 
 

Unresectable, Stage III Non-Small-Cell Lung Cancer. J Clin Oncol, 2022. 40(29): p. 3383-
3393. 

412. Geurts, V.C.M., et al., Unleashing NK- and CD8 T cells by combining monalizumab and 
trastuzumab for metastatic HER2-positive breast cancer: Results of the MIMOSA trial. 
Breast, 2023. 70: p. 76-81. 

413. Devillier, R., et al., Safety of Anti-NKG2A Blocking Antibody Monalizumab As 
Maintenance Therapy after Allogeneic Hematopoietic 

Stem Cell Transplantation: A Phase I Study. ASH, 2021. 
414. Bezman, N., et al., ANTI-NKG2A ANTIBODIES AND USES THEREOF. 2020: USA. 
415. Grzywacz, B., et al., Natural Killer Cell Homing and Persistence in the Bone Marrow 

After Adoptive Immunotherapy Correlates With Better Leukemia Control. J 
Immunother, 2019. 42(2): p. 65-72. 

416. Curti, A., et al., Successful transfer of alloreactive haploidentical KIR ligand-mismatched 
natural killer cells after infusion in elderly high risk acute myeloid leukemia patients. 
Blood, 2011. 118(12): p. 3273-9. 

417. Berrien-Elliott, M.M., et al., Hematopoietic cell transplantation donor-derived memory-
like NK cells functionally persist after transfer into patients with leukemia. Sci Transl 
Med, 2022. 14(633): p. eabm1375. 

418. Bednarski, J.J., et al., Donor memory-like NK cells persist and induce remissions in 
pediatric patients with relapsed AML after transplant. Blood, 2022. 139(11): p. 1670-
1683. 

419. Liu, E., et al., Cord blood NK cells engineered to express IL-15 and a CD19-targeted CAR 
show long-term persistence and potent antitumor activity. Leukemia, 2018. 32(2): p. 
520-531. 

420. Li, L., et al., Loss of metabolic fitness drives tumor resistance after CAR-NK cell therapy 
and can be overcome by cytokine engineering. Sci Adv, 2023. 9(30): p. eadd6997. 

421. Fernandez, R.A., et al., Improving NK cell function in multiple myeloma with NKTR-255, 
a novel polymer-conjugated human IL-15. Blood Adv, 2023. 7(1): p. 9-19. 

422. Rosario, M., et al., The IL-15-Based ALT-803 Complex Enhances FcgammaRIIIa-Triggered 
NK Cell Responses and In Vivo Clearance of B Cell Lymphomas. Clin Cancer Res, 2016. 
22(3): p. 596-608. 

423. Berrien-Elliott, M.M., J.A. Wagner, and T.A. Fehniger, Human Cytokine-Induced 
Memory-Like Natural Killer Cells. J Innate Immun, 2015. 7(6): p. 563-71. 

424. Pal, M., et al., Tumor-priming converts NK cells to memory-like NK cells. 
Oncoimmunology, 2017. 6(6): p. e1317411. 

425. Kujur, W., et al., Memory like NK cells display stem cell like properties after Zika virus 
infection. PLoS Pathog, 2020. 16(12): p. e1009132. 

426. Foley, B., et al., Cytomegalovirus reactivation after allogeneic transplantation promotes 
a lasting increase in educated NKG2C+ natural killer cells with potent function. Blood, 
2012. 119(11): p. 2665-74. 

427. Schlums, H., et al., Cytomegalovirus infection drives adaptive epigenetic diversification 
of NK cells with altered signaling and effector function. Immunity, 2015. 42(3): p. 443-
56. 

428. Merino, A., et al., Chronic stimulation drives human NK cell dysfunction and epigenetic 
reprograming. J Clin Invest, 2019. 129(9): p. 3770-3785. 

429. Orchard, K., et al., In Vivo Generation of Memory-like NK Cells for the Treatment of AML 
and Myelodysplastic Syndrome; Early Clinical Applications of INKmuneTM. ASH, 2022. 

430. Fehniger, T.A., et al., Differential cytokine and chemokine gene expression by human NK 
cells following activation with IL-18 or IL-15 in combination with IL-12: implications for 
the innate immune response. J Immunol, 1999. 162(8): p. 4511-20. 



BIBLIOGRAPHY 

236 
 

431. Cooper, M.A., et al., Cytokine-induced memory-like natural killer cells. Proc Natl Acad 
Sci U S A, 2009. 106(6): p. 1915-9. 

432. Romee, R., et al., Cytokine activation induces human memory-like NK cells. Blood, 2012. 
120(24): p. 4751-60. 

433. Tarannum, M. and R. Romee, Cytokine-induced memory-like natural killer cells for 
cancer immunotherapy. Stem Cell Res Ther, 2021. 12(1): p. 592. 

434. Leong, J.W., et al., Preactivation with IL-12, IL-15, and IL-18 induces CD25 and a 
functional high-affinity IL-2 receptor on human cytokine-induced memory-like natural 
killer cells. Biol Blood Marrow Transplant, 2014. 20(4): p. 463-73. 

435. Lusty, E., et al., IL-18/IL-15/IL-12 synergy induces elevated and prolonged IFN-gamma 
production by ex vivo expanded NK cells which is not due to enhanced STAT4 activation. 
Mol Immunol, 2017. 88: p. 138-147. 

436. Ni, J., et al., Sustained effector function of IL-12/15/18-preactivated NK cells against 
established tumors. J Exp Med, 2012. 209(13): p. 2351-65. 

437. Tanzi, M., et al., Cytokine-Induced Memory-Like NK Cells with High Reactivity against 
Acute Leukemia Blasts and Solid Tumor Cells Suitable for Adoptive Immunotherapy 
Approaches. Cancers (Basel), 2021. 13(7). 

438. Huber, C.M., J.M. Doisne, and F. Colucci, IL-12/15/18-preactivated NK cells suppress 
GvHD in a mouse model of mismatched hematopoietic cell transplantation. Eur J 
Immunol, 2015. 45(6): p. 1727-35. 

439. Terren, I., et al., Implication of Interleukin-12/15/18 and Ruxolitinib in the Phenotype, 
Proliferation, and Polyfunctionality of Human Cytokine-Preactivated Natural Killer Cells. 
Front Immunol, 2018. 9: p. 737. 

440. Ghofrani, J., et al., Semaphorin 7A modulates cytokine-induced memory-like responses 
by human natural killer cells. Eur J Immunol, 2019. 49(8): p. 1153-1166. 

441. Uppendahl, L.D., et al., Cytokine-induced memory-like natural killer cells have enhanced 
function, proliferation, and in vivo expansion against ovarian cancer cells. Gynecol 
Oncol, 2019. 153(1): p. 149-157. 

442. Romee, R., et al., Cytokine-induced memory-like natural killer cells exhibit enhanced 
responses against myeloid leukemia. Sci Transl Med, 2016. 8(357): p. 357ra123. 

443. Pahl, J.H.W., A. Cerwenka, and J. Ni, Memory-Like NK Cells: Remembering a Previous 
Activation by Cytokines and NK Cell Receptors. Front Immunol, 2018. 9: p. 2796. 

444. Orr, M.T. and L.L. Lanier, Natural killer cell education and tolerance. Cell, 2010. 142(6): 
p. 847-56. 

445. Berrien-Elliott, M.M., et al., Multidimensional Analyses of Donor Memory-Like NK Cells 
Reveal New Associations with Response after Adoptive Immunotherapy for Leukemia. 
Cancer Discov, 2020. 10(12): p. 1854-1871. 

446. Vendrame, E., et al., Mass Cytometry Analytical Approaches Reveal Cytokine-Induced 
Changes in Natural Killer Cells. Cytometry B Clin Cytom, 2017. 92(1): p. 57-67. 

447. Boieri, M., et al., IL-12, IL-15, and IL-18 pre-activated NK cells target resistant T cell 
acute lymphoblastic leukemia and delay leukemia development in vivo. 
Oncoimmunology, 2017. 6(3): p. e1274478. 

448. Zhuang, L., et al., Activity of IL-12/15/18 primed natural killer cells against 
hepatocellular carcinoma. Hepatol Int, 2019. 13(1): p. 75-83. 

449. Marin, N.D., et al., Memory-like Differentiation Enhances NK Cell Responses to 
Melanoma. Clin Cancer Res, 2021. 27(17): p. 4859-4869. 

450. Ni, J., et al., Adoptively transferred natural killer cells maintain long-term antitumor 
activity by epigenetic imprinting and CD4(+) T cell help. Oncoimmunology, 2016. 5(9): 
p. e1219009. 

451. Becker-Hapak, M.K., et al., A Fusion Protein Complex that Combines IL-12, IL-15, and IL-
18 Signaling to Induce Memory-Like NK Cells for Cancer Immunotherapy. Cancer 
Immunol Res, 2021. 9(9): p. 1071-1087. 



BIBLIOGRAPHY 

237 
 

452. Kedia-Mehta, N., et al., Cytokine-induced natural killer cell training is dependent on 
cellular metabolism and is defective in obesity. Blood Adv, 2021. 5(21): p. 4447-4455. 

453. Terren, I., et al., Metabolic changes of Interleukin-12/15/18-stimulated human NK cells. 
Sci Rep, 2021. 11(1): p. 6472. 

454. Jacobs, M.T., et al., Memory-like differentiation, tumor targeting monoclonal 
antibodies, and chimeric antigen receptors enhance natural killer cell responses to head 
and neck cancer. Clin Cancer Res, 2023. 

455. Kerbauy, L.N., et al., Combining AFM13, a Bispecific CD30/CD16 Antibody, with 
Cytokine-Activated Blood and Cord Blood-Derived NK Cells Facilitates CAR-like 
Responses Against CD30(+) Malignancies. Clin Cancer Res, 2021. 27(13): p. 3744-3756. 

456. Gang, M., et al., CAR-modified memory-like NK cells exhibit potent responses to NK-
resistant lymphomas. Blood, 2020. 136(20): p. 2308-2318. 

457. He, B., et al., Cytokines induced memory-like NK cells engineered to express CD19 CAR 
exhibit enhanced responses against B cell malignancies. Front Immunol, 2023. 14: p. 
1130442. 

458. Dong, H., et al., Memory-like NK cells armed with a neoepitope-specific CAR exhibit 
potent activity against NPM1 mutated acute myeloid leukemia. Proc Natl Acad Sci U S 
A, 2022. 119(25): p. e2122379119. 

459. Shapiro, R.M., et al., Expansion, persistence, and efficacy of donor memory-like NK cells 
infused for posttransplant relapse. J Clin Invest, 2022. 132(11). 

460. Rutella, S., et al., WU-NK-101: an Enhanced NK Cell Therapy Optimized for Function in 
the Tumor Microenvironment (TME). ESMO, 2022. 

461. Rastelli, L., et al., A first-in-class ex vivo combination between cytokine-induced memory 
like (CIML) NK cells and a CD38 targeting antibody recruiting molecule (ARM) as a novel 
approach to target NK cells without cellular engineering for the treatment of multiple 
myeloma. ASCO, 2020. 

462. Sevilla-Movilla, S., et al., Upregulated expression and function of the alpha4beta1 
integrin in multiple myeloma cells resistant to bortezomib. J Pathol, 2020. 252(1): p. 29-
40. 

463. Denman, C.J., et al., Membrane-bound IL-21 promotes sustained ex vivo proliferation of 
human natural killer cells. PLoS One, 2012. 7(1): p. e30264. 

464. Wang, R.N., et al., Optimized protocols for gammadelta T cell expansion and lentiviral 
transduction. Mol Med Rep, 2019. 19(3): p. 1471-1480. 

465. Song, D.G., et al., Chimeric NKG2D CAR-expressing T cell-mediated attack of human 
ovarian cancer is enhanced by histone deacetylase inhibition. Hum Gene Ther, 2013. 
24(3): p. 295-305. 

466. Amiot, M., et al., Identification and Hierarchy of Clonogenic Factors for Human 
Myeloma Cells: IL-6 for CD45+ Versus IL-6 and Growth Factors for CD45-. ASH, 2005. 

467. Scholzen, T. and J. Gerdes, The Ki-67 protein: from the known and the unknown. J Cell 
Physiol, 2000. 182(3): p. 311-22. 

468. Love, M.I., W. Huber, and S. Anders, Moderated estimation of fold change and 
dispersion for RNA-seq data with DESeq2. Genome Biol, 2014. 15(12): p. 550. 

469. Aryee, K.E., et al., Enhanced development of functional human NK cells in NOD-scid-
IL2rg(null) mice expressing human IL15. FASEB J, 2022. 36(9): p. e22476. 

470. Alter, G., J.M. Malenfant, and M. Altfeld, CD107a as a functional marker for the 
identification of natural killer cell activity. J Immunol Methods, 2004. 294(1-2): p. 15-
22. 

471. Tiedemann, R.E., et al., Identification of kinetin riboside as a repressor of CCND1 and 
CCND2 with preclinical antimyeloma activity. J Clin Invest, 2008. 118(5): p. 1750-64. 

472. Kukita, A. and T. Kukita, Multifunctional properties of RANKL/RANK in cell 
differentiation, proliferation and metastasis. Future Oncol, 2013. 9(11): p. 1609-22. 



BIBLIOGRAPHY 

238 
 

473. Lindahl, M., et al., MANF is indispensable for the proliferation and survival of 
pancreatic beta cells. Cell Rep, 2014. 7(2): p. 366-375. 

474. Shi, Z., et al., Cables1 complex couples survival signaling to the cell death machinery. 
Cancer Res, 2015. 75(1): p. 147-158. 

475. Ito, M., et al., Killer cell lectin-like receptor G1 binds three members of the classical 
cadherin family to inhibit NK cell cytotoxicity. J Exp Med, 2006. 203(2): p. 289-95. 

476. Jelinek, T., B. Paiva, and R. Hajek, Update on PD-1/PD-L1 Inhibitors in Multiple 
Myeloma. Front Immunol, 2018. 9: p. 2431. 

477. Sharma, N., P.M. Reagan, and J.L. Liesveld, Cytopenia after CAR-T Cell Therapy-A Brief 
Review of a Complex Problem. Cancers (Basel), 2022. 14(6). 

478. Zheng, M.M., et al., The systemic cytokine environment is permanently altered in 
multiple myeloma. PLoS One, 2013. 8(3): p. e58504. 

479. Zingone, A., et al., Altered cytokine and chemokine profiles in multiple myeloma and its 
precursor disease. Cytokine, 2014. 69(2): p. 294-7. 

480. Sharma, A., et al., Dysregulation in T helper 1/T helper 2 cytokine ratios in patients with 
multiple myeloma. Leuk Lymphoma, 2010. 51(5): p. 920-7. 

481. Gustafson, K.S. and G.D. Ginder, Interferon-gamma induction of the human leukocyte 
antigen-E gene is mediated through binding of a complex containing STAT1alpha to a 
distinct interferon-gamma-responsive element. J Biol Chem, 1996. 271(33): p. 20035-
46. 

482. Ettari, R., et al., Immunoproteasome-selective and non-selective inhibitors: A promising 
approach for the treatment of multiple myeloma. Pharmacol Ther, 2018. 182: p. 176-
192. 

483. Sun, C., et al., High NKG2A expression contributes to NK cell exhaustion and predicts a 
poor prognosis of patients with liver cancer. Oncoimmunology, 2017. 6(1): p. 
e1264562. 

484. Salome, B., et al., NKG2A and HLA-E define an alternative immune checkpoint axis in 
bladder cancer. Cancer Cell, 2022. 40(9): p. 1027-1043 e9. 

485. Quintarelli, C., et al., Efficacy of third-party chimeric antigen receptor modified 
peripheral blood natural killer cells for adoptive cell therapy of B-cell precursor acute 
lymphoblastic leukemia. Leukemia, 2020. 34(4): p. 1102-1115. 

486. Rasche, L., M. Hudecek, and H. Einsele, CAR T-Cell Therapy in Multiple Myeloma: 
Mission Accomplished? Blood, 2023. 

487. Muller, S., et al., High Cytotoxic Efficiency of Lentivirally and Alpharetrovirally 
Engineered CD19-Specific Chimeric Antigen Receptor Natural Killer Cells Against Acute 
Lymphoblastic Leukemia. Front Immunol, 2019. 10: p. 3123. 

488. Yang, Y., et al., Superior Expansion and Cytotoxicity of Human Primary NK and CAR-NK 
Cells from Various Sources via Enriched Metabolic Pathways. Mol Ther Methods Clin 
Dev, 2020. 18: p. 428-445. 

489. Allan, D.S.J., et al., Expanded NK cells used for adoptive cell therapy maintain diverse 
clonality and contain long-lived memory-like NK cell populations. Mol Ther Oncolytics, 
2023. 28: p. 74-87. 

490. Ciurea, S.O., et al., Phase 1 clinical trial using mbIL21 ex vivo-expanded donor-derived 
NK cells after haploidentical transplantation. Blood, 2017. 130(16): p. 1857-1868. 

491. Vela, M., et al., Haploidentical IL-15/41BBL activated and expanded natural killer cell 
infusion therapy after salvage chemotherapy in children with relapsed and refractory 
leukemia. Cancer Lett, 2018. 422: p. 107-117. 

492. Schmidt, P., M.J. Raftery, and G. Pecher, Engineering NK Cells for CAR Therapy-Recent 
Advances in Gene Transfer Methodology. Front Immunol, 2020. 11: p. 611163. 

493. Dagher, O.K. and A.D. Posey, Jr., Forks in the road for CAR T and CAR NK cell cancer 
therapies. Nat Immunol, 2023. 24(12): p. 1994-2007. 



BIBLIOGRAPHY 

239 
 

494. Colamartino, A.B.L., et al., Efficient and Robust NK-Cell Transduction With Baboon 
Envelope Pseudotyped Lentivector. Front Immunol, 2019. 10: p. 2873. 

495. Mac Donald, A., et al., KLRC1 knockout overcomes HLA-E-mediated inhibition and 
improves NK cell antitumor activity against solid tumors. Front Immunol, 2023. 14: p. 
1231916. 

496. Hromadnikova, I., P. Pirkova, and L. Sedlackova, Influence of in vitro IL-2 or IL-15 alone 
or in combination with Hsp-70-derived 14-mer peptide (TKD) on the expression of NK 
cell activatory and inhibitory receptors. Mediators Inflamm, 2013. 2013: p. 405295. 

497. Nguyen, S., et al., HLA-E upregulation on IFN-gamma-activated AML blasts impairs 
CD94/NKG2A-dependent NK cytolysis after haplo-mismatched hematopoietic SCT. Bone 
Marrow Transplant, 2009. 43(9): p. 693-9. 

498. Zhang, X., et al., Synergized regulation of NK cell education by NKG2A and specific Ly49 
family members. Nat Commun, 2019. 10(1): p. 5010. 

499. Oei, V.Y.S., et al., Intrinsic Functional Potential of NK-Cell Subsets Constrains Retargeting 
Driven by Chimeric Antigen Receptors. Cancer Immunol Res, 2018. 6(4): p. 467-480. 

500. Logue, J.M., et al., Early cytopenias and infections after standard of care idecabtagene 
vicleucel in relapsed or refractory multiple myeloma. Blood Adv, 2022. 6(24): p. 6109-
6119. 

501. Carlsten, M. and R.W. Childs, Genetic Manipulation of NK Cells for Cancer 
Immunotherapy: Techniques and Clinical Implications. Front Immunol, 2015. 6: p. 266. 

502. Romee, R., et al., NK cell CD16 surface expression and function is regulated by a 
disintegrin and metalloprotease-17 (ADAM17). Blood, 2013. 121(18): p. 3599-608. 

503. Carreira-Santos, S., et al., Enhanced expression of natural cytotoxicity receptors on 
cytokine-induced memory-like natural killer cells correlates with effector function. Front 
Immunol, 2023. 14: p. 1256404. 

504. Littwitz-Salomon, E., et al., Metabolic requirements of NK cells during the acute 
response against retroviral infection. Nat Commun, 2021. 12(1): p. 5376. 

505. Schmiedel, B.J., et al., Receptor activator for NF-kappaB ligand in acute myeloid 
leukemia: expression, function, and modulation of NK cell immunosurveillance. J 
Immunol, 2013. 190(2): p. 821-31. 

506. Gomez-Aleza, C., et al., Inhibition of RANK signaling in breast cancer induces an anti-
tumor immune response orchestrated by CD8+ T cells. Nat Commun, 2020. 11(1): p. 
6335. 

507. Zhang, H., et al., Aberrant splicing of cables gene, a CDK regulator, in human cancers. 
Cancer Biol Ther, 2005. 4(11): p. 1211-5. 

508. Schomer, N.T., et al., CCR7 expression in CD19 chimeric antigen receptor-engineered 
natural killer cells improves migration toward CCL19-expressing lymphoma cells and 
increases tumor control in mice with human lymphoma. Cytotherapy, 2022. 24(8): p. 
827-834. 

509. Lachota, M., et al., Mapping the chemotactic landscape in NK cells reveals subset-
specific synergistic migratory responses to dual chemokine receptor ligation. 
EBioMedicine, 2023. 96: p. 104811. 

510. Hong, H.S., et al., Loss of CCR7 expression on CD56(bright) NK cells is associated with a 
CD56(dim)CD16(+) NK cell-like phenotype and correlates with HIV viral load. PLoS One, 
2012. 7(9): p. e44820. 

511. Song, Y., et al., IL-12/IL-18-preactivated donor NK cells enhance GVL effects and 
mitigate GvHD after allogeneic hematopoietic stem cell transplantation. Eur J Immunol, 
2018. 48(4): p. 670-682. 

512. Bakhtiyaridovvombaygi, M., et al., Cytokine-Induced Memory-Like NK Cells: Emerging 
strategy for AML immunotherapy. Biomed Pharmacother, 2023. 168: p. 115718. 

513. Michaelsson, J., et al., A signal peptide derived from hsp60 binds HLA-E and interferes 
with CD94/NKG2A recognition. J Exp Med, 2002. 196(11): p. 1403-14. 



BIBLIOGRAPHY 

240 
 

514. Huisman, B.D., et al., High-throughput characterization of HLA-E-presented 
CD94/NKG2x ligands reveals peptides which modulate NK cell activation. Nat Commun, 
2023. 14(1): p. 4809. 

515. Mehdi, S.H., et al., Animal Models of Multiple Myeloma Bone Disease. Front Genet, 
2021. 12: p. 640954. 

 

 


	Tesis Jessica Encinas Mayoral
	PORTADA
	INDEX
	RESUMEN
	ABSTRACT
	1. INTRODUCTION
	2. HYPOTHESIS AND OBJECTIVES
	3. MATERIALS AND METHODS
	4. RESULTS
	5. DISCUSSION
	6. CONCLUSIONES
	7. CONCLUSIONS
	8. ABBREVIATIONS
	9. BIBLIOGRAPHY