Meta-analysis of the gut microbiota in predicting response to cancer immunotherapy in metastatic melanoma

Abstract

BACKGROUND Identifying factors conferring responses to therapy in cancer is critical to select the best treatment for patients. For immune checkpoint inhibition (ICI) therapy, mounting evidence suggests that the gut microbiome can determine patient treatment outcomes. However, the extent to which gut microbial features are applicable across different patient cohorts has not been extensively explored. METHODS We performed a meta-analysis of 4 published shotgun metagenomic studies ( N tot = 130 patients) investigating differential microbiome composition and imputed metabolic function between responders and nonresponders to ICI. RESULTS Our analysis identified both known microbial features enriched in responders, such as Faecalibacterium as the prevailing taxa, as well as additional features, including overrepresentation of Barnesiella intestinihominis and the components of vitamin B metabolism. A classifier designed to predict responders based on these features identified responders in an independent cohort of 27 patients with the area under the receiver operating characteristic curve of 0.625 (95% CI: 0.348–0.899) and was predictive of prognosis (HR = 0.35, P = 0.081). CONCLUSION These results suggest the existence of a fecal microbiome signature inherent across responders that may be exploited for diagnostic or therapeutic purposes. FUNDING This work was funded by the Knut and Alice Wallenberg Foundation, BioGaia AB, and Cancerfonden.

DeepGeni: Deep generalized interpretable autoencoder elucidates gut microbiota for better cancer immunotherapy

“…Patients' responsiveness to ICI therapy was evaluated with RECIST 1.1 criteria where complete or partial responses are classified as responders and stable or progressive disease states as non-responders [19]. Since Peters et al's data did not have an explicit classification of responsiveness, patients with over 6 months of progression-free survival were regarded as responders and the others as non-responders as suggested by Limeta et al [11]. In total, 130 melanoma patients (66 responders and 64 non-responders) were used (Table 1).…”

“…Many bacteria reported by those studies did not appear in multiple studies at the species level except Faecalibacterium prausnitzii and Bacteroides thetaiotaomicron. Also, previous attempt to train machine learning classifiers on microbiome profiles has shown relatively low accuracy in the prediction of ICI response on unseen data [11]. This suggests the need for curation of massive-scale studies to obtain statistical power to generalize microbial signatures to unseen data.…”

Immunotherapy in Solid Tumors and Gut Microbiota: The Correlation—A Special Reference to Colorectal Cancer

“…This is in agreement with Dublin et al, who demonstrated ICI-associated colitis in 29.4% of melanoma patients under anti-CTLA-4, and their gut microbiota was mainly composed of microorganisms belonging to the phylum of Bacteroidetes , families of Bacteroidaceae , Rikenellaceae and Barnesiellaceae [ 11 ]. A recently published meta-analysis of four shotgun metagenomic studies in metastatic melanoma patients receiving immunotherapy showed differences of microbial composition in the gut of responders and non-responders and examined the predictive value of this composition [ 64 ]. The gut of responders was enriched with unknown Ruminococcaceae and Faecalibacterium spp., Ruminococcus bicirculans and Barnesiella intestinihomini , whereas non-responders were mainly colonized by B. thetaiotaomicron , Adlercreutzia equolifaciens , Bifidobacterium dentium and unknown Mogibacterium spp.…”

“…The gut of responders was enriched with unknown Ruminococcaceae and Faecalibacterium spp., Ruminococcus bicirculans and Barnesiella intestinihomini , whereas non-responders were mainly colonized by B. thetaiotaomicron , Adlercreutzia equolifaciens , Bifidobacterium dentium and unknown Mogibacterium spp. [ 64 ]. Moreover, comparison of gut microbial networks of responders and non-responders revealed that responders’ networks had an abundance of Mogibacterium , Anaerococcus and Eggerthella spp.…”

“…The abundance of unknown Faecalibacterium sp. and of aerobic respiration provided predictive value for the efficacy of immunotherapy [ 64 ].…”