Background
Pooled analysis of 4 phase II/III trials (DIRECTOR, EORTC 1410, 1608, 26034) in first recurrence glioblastoma; examined association of proton-pump inhibitor use (PA-PPI: strong aldehyde dehydrogenase 1A1 activators) with outcome.
Interventions and follow up
Arm A: Patients on PA-PPI at baseline, week 8, week 16
Arm B: No PA-PPI use during assessment window
Primary endpoint: Overall survival, progression-free survival
mFollow up: Median per trial (range 10–30 months)
Arm B: No PA-PPI use during assessment window
Primary endpoint: Overall survival, progression-free survival
mFollow up: Median per trial (range 10–30 months)
Results
Overall survival: PA-PPI use associated with inferior OS on univariate analysis; not significant on multivariable adjustment
Progression-free survival: No significant association with PA-PPI
Confounding: Steroid use correlated with PA-PPI prescription; steroid use rendered PA-PPI insignificant in multivariable models
Progression-free survival: No significant association with PA-PPI
Confounding: Steroid use correlated with PA-PPI prescription; steroid use rendered PA-PPI insignificant in multivariable models
Adverse events
Main adverse events: Not outcome-related; PA-PPI safety profile standard (GI tolerability); steroid-associated toxicity expected.
Conclusions
No evidence that PA-PPI use compromises outcome in recurrent glioblastoma. Prior first-line MGMT status and steroid requirements are primary confounders. Disconnect from first-line setting (where PA-PPI associated with inferior OS) suggests changing prescription patterns or differential biology in recurrent vs newly diagnosed disease.
Key Limitations
Key Limitations: Observational, non-randomized; steroid indication and dosing not standardized; MGMT promoter methylation not uniformly available; recurrent setting may have inherent selection bias vs first-line cohorts.
Clinical Context
PA-PPI use should not be avoided in recurrent glioblastoma based on this analysis; however, alternatives should be considered where possible in newly diagnosed disease. Reinforces importance of minimizing unnecessary medications in GBM patients.
References
References: Le Rhun et al, Eur J Cancer 2026