What impact do the proton pump inhibitors have on the efficacy of immune check point inhibitors in metastatic malignant melanoma?

Abstract

e21040 Background: Acidic microenvironment facilitates the proliferation, progression, and metastasis of the tumors. Mouse models have shown that the proton pump inhibitors (PPI) cause neutralization of the acidic tumor microenvironment (TME) and inhibit the tumor growth. PPI also sensitize the resistant tumors to chemotherapeutic and immunotherapeutic effects. Moreover, PPI’s modify the gut microbiota, that impact the efficacy of immune checkpoint inhibitors (ICI). Clinical studies have shown variable impact of PPI on efficacy of ICI, ranging from no benefit at all to a negative impact. Since, PPI are one of the most commonly used medications, it is important to evaluate its impact on the efficacy of ICI. Methods: We conducted a retrospective analysis on 120 metastatic malignant melanoma patients treated with anti- CTLA-4 (ipilimumab), anti PD-1(pembrolizumab) alone or in combination. Cohort A included patients who were taking PPI at the time of ICI initiation. Cohort B included non-PPI users. Objective response rate (ORR), disease control rate (DCR), overall survival (OS), and progression free survival (PFS) were calculated. ORR and PFS were the primary outcomes. Cox regression univariate and multivariate analyses were performed. The proportions of grade 3 or 4 immune related adverse events (IrAE) were compared between PPI users and nonusers using the Fisher’s exact test. Results: Cohort A included 29 (24.2%) patients, of which 68.7% patients were male. Median age was 65 years. Majority of the patients were taking omeprazole (58.6%). Proportion of the patients taking ipilimumab and pembrolizumab or nivolumab were almost similar (48.3% vs. 44.8%, respectively). Significantly higher proportion of patients achieved ORR and DCR in cohort A [76%, OR = 3.8(1.39 – 10.5), P = 0.009) and 84%, OR = 4.3(1.37 – 13.7), P = 0.013, respectively). The median PFS was significantly higher in cohort A [A = 27.6 vs. B = 4.4 months, HR = 0.3, P = 0.005(0.1 – 0.7)]. Although, median OS was considerably higher in cohort A, the difference was not statistically significant [A = 39.3 vs. B = 28.0 months, HR = 1.01, P = 0.9(0.4-2.0). PPI did not significantly increase the Odds of having grade 3 or 4 IrAE. Conclusions: In this retrospective review, we have observed favorable outcomes in patients receiving PPI when starting on ICI contrary to some of the studies suggesting negative impact of PPI on ICI efficacy. As PPI is one of the commonly used chronic medications, prospective studies are needed to evaluate the real impact of PPI on ICI when taking concomitantly.