DOP47 Gut microbiome contributes to the development of immune checkpoint inhibitor-related colitis

Abstract

Abstract Background The gut microbiota is implicated in shaping immune checkpoint inhibitors (ICIs) efficacy, but relatively little is known about how the microbiota influences immune-related adverse events (irAEs). Our study aimed to explore the association and causal relationship between gut microbiome and immune checkpoint inhibitor-related colitis. Methods We collected faecal samples from 32 patients with advanced malignant tumor patients developing irAE after anti-PD-1 therapy from May 2020 to Aug 2021. Faecal samples were analyzed by 16s rDNA amplicon sequencing method on Illumina MiSeq (PE300) sequencing platform. We also performed faecal microbiota transplantation (FMT) for antibiotic-treated mice with samples from patients with colitic irAE and without irAE. Results We compared the microbiota between 8 colitic irAE and 24 non-colitic irAE patients. Principal coordinates analysis (PCoA) plot showed significant difference of microbiota composition by Adonis analysis (P=0.003) (Fig. 1A). LEfSe analysis detected a lower abundance of Bacteroidota and a higher abundance of Proteobacteria at the phylum level and a lower abundance of Bacteroides and Bifidobacterium and higher abundance of Enterococcus at the genus level in colitic irAE (Fig. 1B). The total abundance of major butyrate-producing bacteria was lower in colitic irAE compared with non-colitic irAE patients (P=0.018) (Fig. 1C). Functional prediction analyses revealed different abundances of environmental information processing, genetic information processing, and metabolic pathways between colitic and non-colitic irAE patients (Fig. 1D). By FMT experiments in mice, we sought to explore the causal relationship between microbiota immune-related colitis. After injection of anti-PD-1 and anti-CTLA-4, three of nine colitic-irAE-FMT mice developed fatal severe colitis, while all nine non-irAE-FMT mice survived without colitis (P=0.052). The three colitic mice showed more rapid weight loss and higher disease activity index (DAI) scores compared with other mice (Fig. 2A and B). The histological colitis score in the irAE-FMT-Sac. mouse (14 points) was higher than those in the non-irAE-FMT mice (3.89 ± 1.17 points) and colitic-irAE-FMT-survival mice (3.67 ± 1.03 points) (Fig. 2C). The mean IHC scores of irAE-FMT-Sac. mouse from high to low were CD8 3.8 ± 1.3 points, CD4 2.0 ± 1.9 points, and CD20 1.1 ± 1.3 points (Fig. 2D), which also supported the occurrence of immune-related colitis. Figure 1. The gut microbiotas of patients with colitic and non-colitic type irAEs. Figure 2. Comparison of colitis in colitic-irAE-FMT and non-irAE-FMT mice. Conclusion Gut microbiota might play important role in the development of immune-related colitis, possibly by modulating metabolic pathways.