Reproducible stool metagenomic biomarkers linked to the melanoma immunotherapy positive outcome

Abstract

The human gut microbiome plays an important role both in humans health and disease. Recent studies have shown the undeniable influence of gut microbiota composition on cancer immunotherapy efficacy. However, these researches show a lack of consensus in defining reproducible metagenomic markers for a positive immunotherapy outcome. Accordingly, extended published data re-analysis may help reveal clearer associations between the composition of the gut microbiota and treatment response. In this study, we analyzed 358 stool metagenomes from 5 studies published earlier: 210 metagenomes from melanoma patients with positive immunotherapy outcome, 148 metagenomes from melanoma patients with negative immunotherapy outcome. The biomarkers were selected by the group comparison of patients stool samples with different treatment responses (47 responders vs 55 non-responders, 102 metagenomes). Selected biomarkers were verified using the available data describing the influence of the fecal microbiota transplantation on melanoma immunotherapy outcomes (9 donors, 6 responders, 19 non-responders, 256 metagenomes). According to our analysis, the resulting cross-study reproducible taxonomic biomarkers correspond to 12 Firmicutes, 4 Bacteroidetes, and 3 Actinobacteria. 140 gene groups were identified as reproducible functional biomarkers, including those potentially involved in production of immune-stimulating molecules and metabolites. In addition, we ranked taxonomic biomarkers by the number of functional biomarkers found in their metagenomic context. In other words, we predicted a list of the potential "most beneficial" bacteria for a positive response to melanoma immunotherapy. The obtained results can be used to make recommendations for the gut microbiota correction in cancer immunotherapy, and the resulting list of biomarkers can be considered for potential diagnostic ways for predicting melanoma immunotherapy outcome. Another important point is the functional biomarkers of positive immunotherapy outcome are distributed in different bacterial species that can explain the lack of consensus of defining melanoma immunotherapy beneficial species between different studies.