Gut Microbiome Transplant Plus Immunotherapy Gets Tested in Esophageal and Gastric Cancer

Esophageal and gastric cancers remain difficult to treat. Even with immune checkpoint inhibitors (ICIs) added to chemotherapy, median overall survival after first-line therapy is estimated at only 13–17 months. Evidence suggests the gut microbiota shapes systemic and antitumor immunity. In metastatic melanoma, a single fecal microbiota transplant (FMT) from healthy donors before anti-PD-1 therapy was linked to a 65% response rate, versus roughly 30% typically reported. A 2024 phase Ib trial also found that FMT from responders restored benefit in anti-PD-1-refractory solid tumors.

This paper is a protocol description, not a results report, for a single-institution phase I-II trial at the National Cancer Center Hospital in Japan (jRCTs031240170). Patients take oral amoxicillin (1,500 mg/day), fosfomycin (3,000 mg/day), and metronidazole (750 mg/day) for one week (minimum five days) to reduce resident bacteria. At least two days later, they receive one 200 mL dose of donor microbiota solution by total colonoscopy, and ICI-containing therapy starts the next day. The solution is prepared by Juntendo University and Metagen Therapeutics. The authors cite ulcerative colitis data showing remission of 82% with antibiotic pretreatment versus 58% without.

A safety part enrolls 3–6 patients per cohort (6–12 total). Cohort 1 receives chemotherapy plus nivolumab or pembrolizumab, and cohort 2 (esophageal squamous cell carcinoma only) receives nivolumab plus ipilimumab. If safety criteria are met, an expansion part enrolls 10 patients per cohort: two esophageal cohorts and one gastric cohort. Eligible patients are adults with ECOG performance status 0–1, measurable disease, and adequate organ function. Exclusions include HER2-positive tumors, inflammatory bowel disease, active autoimmune disease, recent steroids or immunosuppressants, active infection, and HIV or hepatitis B/C. Donors undergo extensive pathogen, blood, and stool screening, including drug-resistant bacteria and SARS-CoV-2.

The primary endpoint is dose-limiting toxicity (DLT) from day 1 to day 28 after FMT. DLT means grade 3–4 non-hematologic toxicity possibly related to FMT, or a delay of 14 days or more in chemotherapy. Secondary endpoints are response rate, disease control rate, progression-free survival, overall survival, and adverse events, with CT every 8 weeks and then every 12 weeks after week 24. Efficacy uses the first stage of Simon's two-stage futility design. Threshold response rates are 45% (fluorouracil/cisplatin plus PD-1 antibody), 30% (nivolumab plus ipilimumab), and 50% (gastric chemotherapy plus nivolumab). A cohort stops if responders number ≤5/10, ≤4/10, or ≤6/11.

Translational work is extensive. Stool is collected after antibiotics and at 4 and 8 weeks, blood at 8 weeks, and tumor and intestinal epithelial biopsies before and after treatment. Flow cytometry will profile CD4+, CD8+, and regulatory T cells, NK cells, MAIT cells, and myeloid-derived suppressor cells. Immunohistochemistry will assess changes in the tumor immune microenvironment. Multi-omics analyses (metabolomics, proteomics, transcriptomics, epigenomics, genomics) aim to identify predictive biomarkers.

Enrollment began in June 2024, and seven patients had enrolled by August 2025, but no safety or efficacy data are reported. The design is single-arm with small cohorts, so it cannot separate the effect of FMT from antibiotics or from standard ICI therapy. Results will be preliminary and hypothesis-generating, and larger randomized studies would be needed to confirm any benefit.