Gut bacteria may predict and boost cancer immunotherapy response

For cancer patients starting immune checkpoint inhibitor therapy, the drugs work spectacularly well for some and barely at all for others, and doctors still can't reliably predict which group a given patient falls into. A new review pulls together more than a decade of trial data arguing that part of the answer lives in the gut: the bacteria lining the intestines appear to shape both how well these drugs work and how likely they are to cause serious side effects.

Key Takeaways:

• Gut bacteria composition is linked to how well immune checkpoint inhibitors (ICIs) work in cancer patients, and to the risk of serious side effects.
• A randomized kidney cancer trial linked donor fecal microbiota transplants (FMT) to nearly triple the progression-free survival of a placebo transplant (24 vs 9 months).
• Antibiotic use shortly before starting an ICI is consistently linked to worse outcomes across multiple studies.
• This is a review of past trials, not new data, and there is no approved microbiome test or treatment doctors can prescribe today.

Gut Microbiome Composition Shapes Immune Checkpoint Inhibitor Response

A review published in Best Practice & Research Clinical Gastroenterology by researchers at Rome's Università Cattolica del Sacro Cuore concludes the makeup of a cancer patient's gut microbiome (the community of bacteria living in the intestines) is now an established factor in how well immune checkpoint inhibitors (ICIs, a class of drugs including anti-PD-1 and anti-CTLA-4 agents that free the immune system to attack tumors) actually work. Two bacteria groups, Akkermansia muciniphila and Faecalibacterium species, keep showing up in patients who respond well, an association replicated across independent cohorts and cancer types. Antibiotic exposure around treatment time does the opposite: in a 196-patient multicenter cohort of people with lung cancer, melanoma and other solid tumors, antibiotic use before starting an ICI was linked to a median overall survival of 2 months versus 26 months in patients who avoided antibiotics, and to primary treatment failure in 81% of antibiotic-exposed patients versus 44% of others. A meta-analysis of more than 41,000 patients across 123 cohorts confirmed the same pattern.

Melanoma, Lung and Kidney Cancer Patients Have the Most at Stake

The stakes are highest where ICIs are now standard care but response is inconsistent: melanoma, non-small cell lung cancer, kidney cancer and other tumors where objective response rates reach 70% in the most immune-responsive cancers, like melanoma and MSI-high tumors (a marker of genetic instability), but fall as low as 10 to 25% in most other cancer types. The same microbiome patterns linked to poor response also appear linked to who develops immune-related adverse events (irAEs), the sometimes-severe autoimmune-like side effects, including colitis, that checkpoint inhibitors can trigger. A meta-analysis of more than 20,000 patients on anti-PD-1 therapy found severe (grade 3 or higher) irAEs occur in about 14% of patients. For someone weighing whether to start immunotherapy, or already managing its side effects, both response and toxicity may trace back to the same gut ecosystem.

The Review Draws on Named Clinical Trials, Not New Data

This is a synthesis of published trials and cohort studies, not a new experiment. In two early fecal microbiota transplant (FMT) trials, transplanting stool from immunotherapy responders into patients who had stopped responding restored sensitivity to the drugs in a subset of recipients (3 of 10 patients in one trial, 3 of 15 in another). That approach was tested more rigorously in the randomized phase 2 TACITO trial, where kidney cancer patients who received donor FMT (from people with long-term complete responses to checkpoint inhibitors) alongside standard therapy had a median progression-free survival of 24 months, versus 9 months for those given a placebo transplant, and a 52% objective response rate versus 32%. Researchers have also tried to turn microbiome patterns into usable tests: one framework built from 808 patients with lung, kidney and skin cancers predicted response with moderate accuracy (an AUC of about 0.71, where 1.0 is perfect). Another tool, TOPOSCORE, converts a patient's bacterial signature into a roughly 48-hour lab test that matches full genetic sequencing results 78% of the time.

What This Could Mean If You're Facing Immunotherapy

None of this is an approved test or treatment yet, but it suggests specific, low-risk questions worth raising with an oncologist. Because antibiotic exposure shortly before starting an ICI is consistently linked to worse outcomes across these studies, patients who can reasonably delay a course of antibiotics, in consultation with their doctor, may want to ask whether that's an option. Diet has the most encouraging evidence: in an observational cohort of melanoma patients on ICIs, those eating at least 20 grams of fiber a day had longer progression-free survival, with each additional 5 grams per day linked to roughly a 30% lower risk of progression or death, an association the review notes was reproduced in mice and disappeared in germ-free animals, supporting a microbiome-driven mechanism. What doesn't have support: over-the-counter probiotics. A commercial probiotic formulation tested in melanoma patients on ICIs failed to improve outcomes and trended toward worse response, so more bacteria from a bottle is not automatically better.

The Evidence Has a Causality Problem

The review's authors are explicit about its biggest limitation: most human evidence is observational, showing association rather than proof that changing the microbiome causes better or worse outcomes. The antibiotic-outcome link, for instance, could partly reflect that sicker patients receive more antibiotics, rather than antibiotics directly disrupting immune-boosting bacteria. The bacterial species linked to good responses also vary considerably between patient cohorts, part of why researchers are shifting from single species toward broader ecological signatures. Several strategies described, including FMT for both boosting response and treating checkpoint-inhibitor colitis, come from small, uncontrolled or early-phase studies rather than large randomized trials.

Larger Trials and Standardized Testing Are the Next Step

The authors call for adequately powered randomized trials to establish whether deliberately modifying the microbiome, through diet, FMT or engineered bacterial products, actually causes better immunotherapy outcomes rather than just correlating with them. A 2025 international consensus statement already lays out practical guidance for standardizing microbiome testing, a step the authors describe as necessary before these signatures can move from research into routine oncology care.

Common Questions

Can I get a gut microbiome test before starting cancer immunotherapy? Not yet as a standard clinical test tied to a specific treatment decision; the scoring tools described in this review remain research tools that haven't been validated widely enough for routine use.

Should I avoid antibiotics before starting immunotherapy? Talk to your oncologist first. Several studies link antibiotic use shortly before checkpoint inhibitor treatment to worse outcomes, but antibiotics are sometimes medically necessary, and this research doesn't prove that avoiding them causes better results.

Do probiotics help immunotherapy work better? The evidence so far doesn't support it. A commercial probiotic tested in melanoma patients failed to improve response and trended toward worse outcomes, while a high-fiber diet showed a more promising, though still unproven, association.

References

Margarita, E., Saito, K., Bisegna, P., et al. (2026). Gut microbiome in immuno-oncology: potential for clinical exploitation. Best Practice & Research Clinical Gastroenterology. https://www.sciencedirect.com/science/article/pii/S1521691826000533

Routy, B., et al. (2018). Gut microbiome influences efficacy of PD-1-based immunotherapy against epithelial tumors. Science. https://www.science.org/doi/10.1126/science.aan3706

Thomas, A.M., et al. (2023). Gut OncoMicrobiome signatures (GOMS) as next-generation biomarkers for cancer immunotherapy. Nature Reviews Clinical Oncology. https://www.nature.com/articles/s41571-023-00785-8

Derosa, L., et al. (2024). Custom scoring based on ecological topology of gut microbiota associated with cancer immunotherapy outcome. Cell. https://www.cell.com/cell/fulltext/S0092-8674(24)00538-5

Porcari, S., et al. (2026). Fecal microbiota transplantation plus pembrolizumab and axitinib in metastatic renal cell carcinoma: the randomized phase 2 TACITO trial. Nature Medicine. https://www.nature.com/articles/s41591-025-04189-2

Porcari, S., et al. (2025). International consensus statement on microbiome testing in clinical practice. The Lancet Gastroenterology & Hepatology. https://www.thelancet.com/journals/langas/article/PIIS2468-1253(24)00311-X/abstract