While immune checkpoint inhibitors (ICIs) have revolutionized the treatment of many cancers by producing durable anti-tumor responses, only 10-30% of treated patients respond to the available immunotherapy drugs and the ability to predict response to treatment remains elusive. Preliminary studies suggest that the gut microbiome may be an independent, novel modulator of systemic anti-tumor responses to ICIs, initially through bacterial interaction with the immune system in gut-associated lymphoid tissue. Multiple mouse tumor models have been developed to further elucidate the mechanism(s); specifically, FMT of human stool from ICI responder versus non-responder patients into germ-free mice suggests that responder human microbiota can facilitate tumor and immune responses in murine transplantable tumor models. However, data about how the mouse colon modulates the human microbiota are limited or lacking. We hypothesized that only a subset of specific human microbiota establish in the mouse colon and that analysis of these microbes may provide insight into the specific microbial communities that mediate ICI responses. To test this hypothesis, we first selected two human patients who were a distinct ICI responder (R) and nonresponder (NR) and tested their fecal samples in GF mice using syngeneic transplantable tumor models employing B16F0 and MC38 tumor cell lines. While the human ICI responses were replicated in these models, it was notable that individual mouse tumor responses were highly variable. To identify sources of experimental variability, we performed 16S rRNA amplicon sequencing on the human stool samples, experimental inocula, and mouse fecal samples at multiple time points. Our data show that the inocula (alpha diversity, composition) for each experiment were similar to the pre-treatment human stool. However, only 40-50% of human microbes were able to engraft in the mouse colon, and the relative abundance in the inocula was not the primary indicator for species engraftment. When we compared microbes across multiple experiments using beta diversity metrics, each experiment largely contained a distinct set of bacteria. Further analysis is underway to detect longitudinal microbiome shifts, cage effects, and/or bacteria enriched in small (responding) versus large (non-responding) tumors. Our results show that the mouse colon significantly modulates human microbiota in mouse FMT models and mouse microbiota analyses in germ-free models may yield mechanistic insights into bacteria facilitating ICI responses.

Citation Format: Fyza Y. Shaikh, Joell J. Gills, James R. White, Fuad Mohammad, Courtney M. Stevens, Jarushka Naidoo, Drew M. Pardoll, Cynthia L. Sears. The mouse colon modulates human microbes in transplantable murine tumor models after human fecal microbiota transfer (FMT) [abstract]. In: Abstracts: AACR Virtual Special Conference: Tumor Immunology and Immunotherapy; 2020 Oct 19-20. Philadelphia (PA): AACR; Cancer Immunol Res 2021;9(2 Suppl):Abstract nr PO008.