Gut microbiota dynamics of antibiotic fecal microbiota transplantation in esophageal and gastric cancer patients treated with immune checkpoint inhibitors: Preliminary results from the Biorich2 study.

Abstract

e16030 Background: Despite advances in immune checkpoint inhibitor (ICI)–based therapy, outcomes for patients with unresectable or recurrent esophageal and gastric cancers remain poor. The gut microbiome has emerged as a determinant of response to ICI, and modulation of the gut microbiome through fecal microbiota transplantation (FMT) has shown potential to improve ICI efficacy. We conducted a phase I/II study to evaluate the safety, feasibility, and efficacy of FMT with antibiotic pretreatment (A-FMT) combined with ICI-based regimens in patients with esophageal squamous cell carcinoma (ESCC) and gastric/gastroesophageal junction (G/GEJ) adenocarcinoma. Methods: The study consists of a safety part and an expansion part. Patients receive a 1-week course of oral antibiotics (amoxicillin, fosfomycin, and metronidazole), followed by a single donor-derived FMT via colonoscopy. ICI-based therapy is initiated the day after A-FMT. For ESCC, regimens included nivolumab (3 mg/kg) plus ipilimumab (1 mg/kg), or fluorouracil (800 mg/m 2 ) plus cisplatin (80 mg/m 2 ) [CF] combined with pembrolizumab (200 mg/body). For G/GEJ cancer, the regimen consisted of oxaliplatin (130 mg/m 2 ) plus S-1 (40 mg/m 2 ) [SOX] combined with nivolumab (360 mg/body). The primary endpoint was the incidence of dose-limiting toxicities (DLTs) within 28 days after FMT. DLTs were defined as any grade 3 or higher non-hematological toxicity related to FMT. Secondary endpoints included efficacy outcomes and incidence of adverse events. Exploratory translational analyses using stool, blood, and tumor samples evaluated immune modulation and gut microbiome dynamics. Results: Since June 2024, seven patients have been enrolled in the safety part. Six patients received A-FMT; one patient with G/GEJ cancer discontinued the protocol prior to FMT due to rapid disease progression. The ICI-based regimens administered ipilimumab plus nivolumab (n = 3), CF plus pembrolizumab (n = 1), and SOX plus nivolumab (n = 2). Among these six patients, no DLTs were observed within 28 days. A grade 3 AST elevation (n = 1) and grade 3 pneumonitis (n = 1) were observed during the DLT evaluation period; however, these were considered unrelated to FMT. An objective response was achieved in 4 of 6 patients, including two patients with ESCC and two with G/GEJ cancer; stable disease was observed in two patients with ESCC. Notably, one patient with G/GEJ cancer achieved complete resolution of peritoneal metastasis and underwent conversion surgery. Longitudinal fecal microbiome analyses demonstrated decreased Bray–Curtis dissimilarity between recipient and donor microbiota after A-FMT. Conclusions: A-FMT demonstrated an acceptable safety profile and feasibility, allowing the study to proceed to the expansion part for further efficacy evaluation. Clinical trial information: jRCTs031240170.