Proton pump inhibitor use and overall survival in patients with gastric adenocarcinoma treated with first-line immunotherapy: A real-world pharmacoepidemiologic study.

Abstract

e15693 Background: Proton pump inhibitors (PPIs) are frequently prescribed to patients with gastric cancer for gastroesophageal reflux disease (GERD), ulcer prophylaxis, and symptom control. Emerging evidence suggests that PPIs may adversely affect immune checkpoint inhibitor (ICI) efficacy by disrupting gut microbiota composition, a factor increasingly recognized as a determinant of the antitumor immune response. We evaluated the association between concomitant PPI use and overall survival (OS) in a real-world cohort of patients receiving first-line ICI therapy. Methods: We performed a retrospective study using TriNetX. Adult patients with gastric adenocarcinoma who initiated nivolumab- or pembrolizumab-based first-line treatment between 2016 and 2024 were identified. Patients were classified as active PPI users if they had documented PPI prescription/administration (omeprazole, pantoprazole, esomeprazole, lansoprazole) within 30 days before or after ICI initiation, and non-users otherwise. Cohorts were 1:1 propensity score-matched for age, sex, PPI indication, comorbidities, baseline corticosteroid use, and prior Helicobacter pylori infection. OS was analyzed using Kaplan–Meier and Cox proportional hazards models. Sensitivity analyses exclude patients with recent GI bleeding and restricted exposure to sustained PPI use (> 60 days). Results: Following propensity score matching, 4,372 patients were included (2,186 PPI users and 2,186 non-users). Baseline characteristics were well balanced between cohorts (mean age 64.1 years; 38% female; GERD prevalence 41%; peptic ulcer disease 18%). At a median follow-up of 20.3 months, active PPI use was associated with significantly worse overall survival compared with non-use. Median OS was 11.2 months in PPI users versus 14.9 months in non-users (log-rank p < 0.001). Concomitant PPI exposure was independently associated with inferior OS (HR 1.27, 95% CI 1.16–1.39). Subgroup analyses demonstrated consistent findings across: ICI agents such as nivolumab (HR 1.25) and pembrolizumab (HR 1.29), presence of GERD, HR 1.22 in documented GERD, and absence of ulcer disease: HR 1.31 in patients without prior peptic ulcer disease. Sensitivity analyses yielded similar effect estimates, supporting the robustness of the findings. Conclusions: In this large real-world cohort of patients with gastric adenocarcinoma treated with first-line immunotherapy, concomitant PPI use was associated with significantly worse overall survival, even after rigorous adjustment for PPI indications and comorbidity burden. These findings support the microbiome hypothesis, suggesting that acid-suppressive therapy may impair the effectiveness of immunotherapy. Careful evaluation of the need for PPIs and prospective studies incorporating microbiome-directed interventions are warranted.