Effects of antibiotics and proton pump inhibitors in NSCLC patients treated with atezolizumab and docetaxel: Pooled analysis of the OAK and POPLAR trials

Background: Preclinical data show that antibiotics (ATB) and proton pump inhibitors (PPI) may modulate the microbiome. Recent data from retrospective analyses suggest reduced benefit of immune checkpoint inhibitors (ICI) in patients treated with ATB. Here we explore ATB and PPI use in NSCLC patients who received atezolizumab or chemotherapy. Methods: This retrospective analysis was performed with pooled data from the phase III OAK (NCT02008227) and phase II POPLAR (NCT01903993) studies. In both studies, patients with previously treated metastatic NSCLC were randomized to receive either atezolizumab or docetaxel. All patients from these studies were included in the analysis (n = 1512, 757 in the atezolizumab and 755 in the docetaxel group). The primary objective of this study was to assess the impact of ATB or PPI use within 30 days before and after beginning treatment on overall survival (OS). Hazard ratios (HR) were estimated using univariate and multivariate cox regression models adjusting for baseline variables, such as treatment arm, histology, ECOG number of metastatic sites and age. Results: A total of 169 (22.3%) patients in the atezolizumab and 202 (26.8%) patients in the docetaxel group received ATB. PPI was used in 234 (30.9%) patients in the atezolizumab group and 260 (34.4%) patients in the docetaxel group. The univariate analysis showed that ATB use was associated with shorter OS (HR 1.21 [95%CI 1.05-1.39]), as was PPI use (HR 1.29 [95% CI 1.13-1.48]). This impact remained significant at multivariate analysis in the overall population. While univariate analysis of interactions between ATB or PPI use and treatment showed a shorter OS in the atezolizumab treated population (HR 1.32 [95%CI 1.06-1.63] and HR 1.45 [95% CI 1.20-1.75], resp.), the multivariate analysis did not confirm the relationship to be statistically significant. Conclusions: Together with previous published data, this retrospective analysis suggest that ATB or PPI use 30 days before and after beginning treatment in patients with metastatic NSCLC may be associated with lower efficacy of ICI. Future research on cancer immunotherapy should elucidate the effects of concomitant medications and the role of microbiome. Clinical trial identification: NCT0190399; NCT02008227. Legal entity responsible for the study: F. Hoffmann- La Roche. Funding: Funding support provided by the imCORE Network on behalf of F.Hoffmann- La Roche. Disclosure: M. Chalabi, M. Kok: Advisory board: Bristol-Myers Squibb; Research funding to institute: Roche-Genentech, Bristol-Myers Squibb. A. Cardona, A.M. Dhawahir Scala: Full time employee: Hoffmann-La Roche. D. Nagarkar: Full time employee and stocks: Genentech. M. Albert: Full time employee: Genentech. T.B. Powles: Honoraria: BMS, Roche, AZ, Exelexis, MSD, Pfizer, Novartis, Ipsen, Esai, Astellas, Seattle Genetics, Incyte; Research funding: BMS, Roche, AZ, MSD, Pfizer. All other authors have declared no conflicts of interest.