Abstract
BACKGROUND: Proton pump inhibitors (PPIs) are often prescribed to manage corticosteroid-induced gastrointestinal toxicity during glioblastoma (GBM) treatment, but were recently identified as strong inducers of aldehyde dehydrogenase-1A1 (ALDH1A1). ALDH1A1 is a primary metabolic enzyme impacting the outcome of chemotherapy, including temozolomide. High expression of ALDH1A1 is associated with poor prognosis in multiple cancers, suggesting PPIs may have a negative impact on survival. METHODS: Real-world data on GBM patients was annotated from electronic medical records (EMR) according to the prospective observational study, XCELSIOR (NCT03793088). Patients with known IDH1/2 mutations were excluded. Causal effects on survival were analyzed using a multivariate, time-varying Cox Proportional Hazard (CPH) model with stratifications including MGMT methylation status, age, sex, duration of corticosteroid use, extent of resection, starting standard-of-care, and PPI use. RESULTS: EMR data from 554 GBM patients across 225 cancer centers was collected, with 72% of patients receiving care from academic medical centers. Patients treated with PPIs (51%) had numerically lower median overall survival (mOS) and 2-year OS rates in the total population and across most strata, with the greatest difference for MGMT-methylated patients (mOS 29.2 vs. 40.1 months). In a time-varying multivariate CPH analysis of the above strata, PPIs caused an adverse effect on survival (HR 1.67 [95% CI: 1.15-2.44], P = .007). CONCLUSIONS: Evidence from a nationwide cancer registry has suggested PPIs have a negative impact on OS for GBM patients, particularly those with MGMT promoter methylation. This suggests PPIs should be avoided for prophylactic management of gastrointestinal toxicity in patients with GBM receiving chemoradiotherapy.