Untargeted metabolomics approaches challenge the nutri-score FOPNL system in soluble cocoa products

Abstract

Front-of-pack nutrition labels (FOPNL) like the Nutri-Score aim to guide healthier food choices, but their focus on limited nutritional parameters may overlook key health-promoting compounds. This study analyzed 54 commercial soluble cocoa powders from the Spanish market—covering 19 brands and Nutri-Score categories A to D—using untargeted metabolomics (HPLC-ESI-TOF-MS) and univariate and multivariate statistical analysis. PCA and PLS2 analyses showed no clear clustering by Nutri-Score categories, while ANOVA and post-hoc tests revealed limited metabolic differences, highlighting a mismatch between Nutri-Score classification and the actual chemical diversity of soluble cocoa products. Nutri-Score ratings primarily reflect macronutrient balance but fail to account for health-promoting metabolites. In several cases, products richer in bioactive compounds were penalized with lower Nutri-Scores, underscoring a limitation of this system when applied to complex food matrices. These findings highlight the need for more comprehensive labeling approaches that integrate metabolomic insights.