Abstract 2803: Identification of microbiome and predictive immune biomarkers in patients with advanced non-small lung cancer treated with pembrolizumab with or without chemotherapy

Abstract Background: Recent studies have shown that gut microbiota modulates antitumor immunity, and affect clinical efficacy of immune checkpoint inhibitors, including anti-programmed death-1 (PD-1) and programmed death-ligand 1 (PD-L1) inhibitors. We aimed to explore the metagenomic profiles of previously untreated, treatment naïve, stage IV non-small cell lung cancer (NSCLC) treated with pembrolizumab, an anti-PD1 agent with or without platinum-based chemotherapy and identify distinct ecological niche between responder and non-responder groups. Methods: We performed metagenome sequencing, whole exome sequencing (WES), and RNA sequencing (RNA-seq) on 50 patients with stage IV NSCLC, including adenocarcinoma (N=31) and squamous cell carcinoma (N=22). We employed Dirichlet Multinomial Mixtures (DMM) community typing to cluster microbial community profiling data at the genus level and analyzed the changes in microbial composition and diversity between enterotype 1 (E1, N=27) and enterotype 2 (E2, N=23). Mutation landscape including genomic alterations in relation to immune response was analyzed between responders and non-responders using WES data. Differentially expressed genes (DEGs), Gene Set Enrichment Analysis (GSEA), and immune deconvolution analysis were carried out using RNA-seq data to analyze molecular profiling of two enterotypes between the responder (N=15) and non-responder (N=35) groups. Results: The two enterotypes exhibited distinct microbial abundances and alpha diversity indices. Abundances of members of Bacteroidaceae (P=0.011) and Oscillospiraceae (P<0.001) were higher in E2 members, and enrichment of Lactobacillaceae (P=0.006) and Bifidobacteriaceae (P=0.041) were higher in E1. Compared to E1, the number of observed bacterial species (P<0.001) and Shannon index (P<0.001) was significantly higher in E2, indicating a healthier gut environment. The proportion of responders was higher in E2 than in E1 cluster. Comparison between responders and non-responder groups within each enterotype revealed that Lactobacillaceae showed an association with the response in E2 cluster. Mutation landscape revealed TP53 mutation as the most common mutation, more frequently observed in the E2 (100%) than in the E1 (56%) cluster. Conclusion: We explored the association between the gut microbiome and clinical response to pembrolizumab with or without combination with chemotherapy and found that E2 cluster enriched with Lactobacillaceae is associated with treatment response. Further analysis with larger cohort is ongoing. Citation Format: Jii Bum (Joy) Lee, Su-Jin Choi, Gang-Taik Lee, Junwon Yang, Hyun-Seok Oh, Arim Min, Bo-Eun Kwon, Seong-san Kang, Sujeong Baek, Dong Kwon Kim, Jae Hwan Kim, Sun Min Lim, Min Hee Hong, Kyoung-Ho Pyo, Byoung Chul Cho. Identification of microbiome and predictive immune biomarkers in patients with advanced non-small lung cancer treated with pembrolizumab with or without chemotherapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(6_Suppl):Abstract nr 2803.