Harnessing the Microbiome in Cancer Immunotherapy: Regulation, Prediction, and Therapeutic Targeting

Humans are metaorganisms, composed of both host (human) cells and a roughly equal number of commensal microorganisms-collectively known as the microbiome-residing primarily at epithelial barrier surfaces. This review considers human cancer as a disease of the metaorganism, to which the microbiome contributes by influencing genome stability, tissue organization, inflammation, immunity, tumor initiation and promotion, metastasis formation, and therapeutic response. We summarize evidence demonstrating that machine learning models trained on patients' microbiome features moderately predict clinical response to immunotherapy and the development of immune-related adverse events. We review results from single-arm and randomized clinical trials wherein fecal microbiome transplantation from therapy-responsive patients or healthy donors, when combined with therapy targeting programmed cell death 1 (PD-1), improved outcomes in PD-1-refractory patients or served as an effective first-line intervention. We conclude by highlighting the emerging opportunities and ongoing challenges in leveraging the microbiome to enhance the efficacy and safety of cancer immunotherapy.