Abstract P3-09-16: Fecal microbiome and association with outcomes among patients (pts) receiving eribulin (E) +/- pembrolizumab (P) for hormone receptor positive (HR+) metastatic breast cancer (MBC)

Abstract

Abstract Background: Recent studies in murine models and humans suggest that modulation of the gut microbiome may enhance response to immune checkpoint blockade in different cancer types. However, little is known about the landscape of gut microbiome in pts with HR+ MBC and its influence on response to chemotherapy and immunotherapy.Methods: A randomized phase 2 trial of E +/- P for pts with HR+ MBC (n= 88) was conducted. The results of this study have been previously presented and found the addition of P did not result in an improvement in progression-free survival (PFS). Fecal samples were prospectively collected (baseline (BL), after 2 cycles of therapy (C2), and end of treatment (EOT)) in a subset of pts (E+P, n = 12; E, n = 11) in order to evaluate if microbiome was associated with response (partial response [PR], stable disease [SD], and progressive disease [PD]), PFS, and overall survival (OS). 16S rRNA gene sequencing was performed to characterize the diversity and composition of fecal microbiome. Results: A total 23 pts provided 23 BL, 22 C2, and 5 EOT fecal specimens for microbiome profiling. 16S(v3-4) rRNA gene data was successfully generated for all samples and the dataset was rarefied to 24,743 reads for analysis. Overall, the variability in the composition and structure of the fecal microbiome was significantly driven by each individual (p<0.001) more than any other variable studied. This effect was observed across all timepoints collected. At BL, subjects randomized to receive E+P and E did not show any differences in composition (p = 0.715) and structure (p = 0.457) of the fecal microbiome. The major genera identified in E+P and E groups were Bacteroides, 37.9% vs. 41.56%; Faecalibacterium 9.9% vs. 4.4%, and Blautia 4.9% vs. 2.9%, respectively. These abundances were observed at a similar level in the C2 sample. Overall, subjects receiving E had a marked increase in Faecalibacterium from 6.4% at BL to 21.2% at C2 and a decrease in Akkermansia from BL (8.5%) to C2 (<1%). Among those pts receiving E who achieved a PR, the abundance of Faecalibacterium increased from BL (4.3%) to C2 (13.9%), while the levels remained unchanged in those achieving SD. These shifts in abundance were not observed in the group receiving E+P. Subjects with PFS below the median had comparable alpha-diversity scores to those above the median in both arms of the study. Specifically, the number of Operational Taxonomic Units (OTUs) observed at BL in pts receiving E+P with PFS above and below the median were 563 and 663, respectively. When looking at the differences between BL and C2, OTUs decreased slightly to 513 and 637, respectively. Similar results were observed in pts receiving E. Among pts with PFS time below the median receiving E, there was a decrease in Akkermansia from BL (5.7%) to C2 (<1%). Pts with OS below the median had comparable alpha-diversity scores at BL to those above the median OS in both arms of the study. Pts with OS above the median receiving E experienced a decrease in the abundance of Akkermansia from BL (5.3%) to C2 (2.4%). Conclusion: The composition and structure of the fecal microbiome identified in this study were found to be associated primarily with the pt rather than with any of the variables studied including type of treatment and outcome. Although the fecal microbiome from all pts randomized were indistinguishable at BL, shifts in the abundance of certain types of bacteria like Faecalibacterium and Akkermansia were observed from BL to C2. These shifts were characteristic of pts receiving E but not of those receiving E+P. Although this study is limited by the small sample size in each arm, these findings warrant further evaluation in a larger population to interrogate if the composition and metabolic potential of the fecal microbiome can be used as a predictor of benefit to treatment. Citation Format: Romualdo Barroso de Sousa, Nadim Ajami, Tanya E Keenan, Chelsea Andrews, Jessica L Pittenger, Gerburg Wulf, Laura Spring, Ian E Krop, Eric P Winer, Elizabeth A Mittendorf, Sara M Tolaney. Fecal microbiome and association with outcomes among patients (pts) receiving eribulin (E) +/- pembrolizumab (P) for hormone receptor positive (HR+) metastatic breast cancer (MBC) [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P3-09-16.