The Association of the Microbiome with Melanoma Tumor Response to Immune Checkpoint Inhibitor Treatment and Immune-Related Adverse Events (NCT05102773)

Abstract

Improved understanding of the factors that underlie immune checkpoint inhibitor (ICI) response and toxicity are needed as only half of patients with metastatic melanoma respond, and 10-40% experience immune-related adverse events (irAEs). Modifying the gut microbiome could positively affect response to ICIs and reduce toxicities. Here, we sought to determine if the pre-treatment gut microbiome predicts ICI response or toxicity in the setting of metastatic melanoma. Melanoma patients (n=88) over 18 years of age, planning to receive ICI therapy enrolled in a prospective observational cohort study at The Ohio State University Comprehensive Cancer Center Skin Cancer Clinic. Patients taking corticosteroids for indications other than adrenal physiologic replacement were excluded. Stools were collected at baseline, within 10 days of an irAE as determined by CTCAE v 5.0 criteria, and at 12 weeks. ICI response and progression-free survival (PFS) were evaluated q12 weeks using Response Evaluation Criteria in Solid Tumors (RECIST v1.1). Metagenomic whole-genome shotgun sequencing of the microbiome was classified using MetaPhlAn4/HUMAnN3 and differential abundance analyzed with ANCOM-BC2. Of the 88 patients enrolled, 41 had metastatic disease and complete data. There were 25 participants classified as responders, defined as having complete response or partial response according to RECIST criteria, or stable disease with 6-month PFS. Grade ≥ 1 irAEs were observed in 15/41 participants. The abundance of Intestinimonas butyriciproducens (q-value = 0.002) and Longicatena caecimuris (q-value = 0.003) were enriched in responders, Tenericutes (q-value= 0.001) and Lachnospira sp. NSJ 43 (q-value =0.002) in non-responders. Blautia luti, as well as several other Lachnospiraceae, were associated with response and no irAE (response q-value = 0.02, no irAE q-value = 0.02). The association of response to ICIs with several taxa in the family Lachnospiraceae, a prevalent microbial family in the gut, is consistent with prior research, which has found that this family may influence treatment outcomes through various mechanisms, such as immune regulation, metabolism, and pathogen exclusion. While no statistical relationship was observed between response and irAEs in this cohort, the microbes associated with both could serve as biomarkers. Future studies to assign causal roles for (specific microbes) in response and toxicity could identify mechanisms to improve patient outcomes.

Keywords: immune-related adverse events; immunotherapy; microbiome.

Figure 1.. Summary of the study timeline and sample collection.

A) Visualization of participant recruitment, consent, inclusion, and sample collection showing the final cohort, and the number of samples collected at each time point. B) Study schema depicting trial timeline. Patients with metastatic melanoma were asked to collect blood and microbiome samples within 10 days of ICI cycle 1 day 1 (C1D1), at their 12-week computed tomography (CT) scan, and in the event of any grade irAE within 16 weeks of C1D1. C) A timeline for each patient, showing the variation in microbiome sample collection, CT scan timing, response, and irAE development.

Figure 2.. Differential abundance of baseline gut microbes between R and NR.

A) Relative abundance of the dominant phyla in each patient. B) Enrichment of individual microbes, at all taxonomic levels, in R (right) vs. NR (left) with the false discovery rate-corrected P-values. C) Microbes found only in 1 category or another and present in at least 15% of the samples in that category.

Figure 3.. Differential abundance of baseline gut microbes in the irAE and no-irAE groups.

A) Relative abundance of the dominant phyla in each patient. B) Enrichment of individual microbes, at all taxonomic levels, in the no-irAE (right) versus irAE (left) groups with the false discovery rate corrected P-values. C) Microbes found in a single group (no-irAE or irAE) and present in at least 15% of the samples in that category.