Abstract
The gut microbiota influences immune checkpoint inhibitors (ICIs) efficacy. Microbiome-modulating strategies (MMSs), including probiotics, synbiotics, and faecal microbiota transplantation (FMT), have emerged as promising adjuncts, but their clinical impact remains uncertain. We systematically reviewed PubMed, Embase, and CENTRAL to February 2025 for clinical cohorts evaluating MMS in cancer patients receiving ICIs. Thirty-six studies (25 trials/cohorts; n = 2,746) were included. Meta-analyses, and subgroup analyses were performed for efficacy along with microbiome shifts and safety. MMS plus ICIs achieved a pooled objective response rate (ORR) of 40% (95% CI: 31%-49%; I² = 63.4%; p = 0.0003; 95% PI: 15%-72%). Descriptive proportions showed ORR of 45% (95% CI: 32%-58%; I² = 72.5%; p = 0.0058) for probiotics and 33% (95% CI: 22%-48%; I² = 60.7%; p = 0.0064) for FMT; however, these findings are non-comparative and confounded by study differences. Exploratory subgroup signals were noted for probiotics in NSCLC (ORR 55%; 95%CI: 45%-64%; I² = 0%; p = 0.3683) and FMT in melanoma (ORR 39%; 95% CI: 15%-69%; I² = 72.5%; p = 0.0262). Dual ICI regimens showed the highest point estimate for ORR (43%; 95% CI: 17%-73%; I² = 68.5%; p = 0.0747) but increased toxicity. Microbiome analyses revealed enrichment of short-chain fatty acid-producing taxa and Bifidobacterium spp. among responders. Based on a limited pooled sample size (n = 143), MMS-related adverse events were mostly grade 1-2 (42%; 95% CI: 14%-77%, I² = 53.8%, p = 0.0210), with rare severe events (1%). Overall, MMS show promising, though preliminary, hypothesis-generating signals for modulating ICI response. Given high heterogeneity and reliance on early-phase, single-arm trials, the findings underscore urgent need for large, biomarker-driven randomized controlled trials to define optimal interventions and cautiously integrate microbiome modulation into immuno-oncology care.