Association of gut microbiota and peri-initiation proton pump inhibitor/antibiotic exposure with outcomes of atezolizumab-bevacizumab-carboplatin-paclitaxel regimen in patients with advanced non-squamous NSCLC: Prospective phase II study (K-TAIL-201).

Abstract

8541 Background: Concomitant medications that alter gut microbiome composition, such as proton pump inhibitors (PPIs) and antibiotics, have been suggested as possible modifiers of immune checkpoint inhibitor outcomes; however, there is a paucity of prospective data with pre-specified exposure windows during atezolizumab-bevacizumab-carboplatin-paclitaxel (ABCP) treatment. This study investigated the associations between peri-initiation PPI/antibiotic exposure and gut microbiota with clinical outcomes. Methods: K-TAIL-201 is a prospective, single-arm, phase II study (jRCT031200088) that enrolled Japanese patients with previously untreated advanced non-squamous non-small cell lung cancer (NSCLC) who received an induction ABCP regimen, followed by maintenance with atezolizumab plus bevacizumab. The primary endpoint was six-month progression-free survival (PFS) rate. PPI/antibiotic exposure was evaluated in two pre-specified windows: pre-treatment (up to 21 days before ABCP) and early on-treatment (the first 21 days of treatment). Longitudinal fecal samples were collected and analyzed using 16S rRNA gene sequencing. Exploratory analyses were conducted to investigate the relationships between exposure/microbiota features and outcomes. Results: Thirty-two patients were enrolled, with a median follow-up period of 20.6 months. The 6-month PFS rate was 59.4% (95% confidence interval (CI), 40.6–76.3), objective response rate was 50.0% (95% CI, 31.9–68.2), median PFS was 7.1 months (95% CI, 5.9–9.6), and the median overall survival (OS) was 24.3 months (95% CI, 18.7–not reached). Early on-treatment PPI exposure was associated with poorer outcomes, including shorter PFS (hazard ratio [HR] 7.07, p < 0.001) and OS (HR 5.66, p = 0.009), whereas pre-treatment PPI exposure was not associated with PFS of at least six months. Early on-treatment antibiotic exposure was associated with a lower 6-month PFS rate (21% versus 89%, p < 0.001), but showed no association with time-to-event PFS (HR 1.62, p = 0.41). Microbiota analyses revealed no significant differences in alpha or beta diversity by outcome or exposure group. However, Bifidobacterium was more frequently detected among responders. Conclusions: Early on-treatment (but not pre-treatment) PPI exposure is strongly associated with inferior PFS and OS in patients with advanced non-squamous NSCLC on ABCP regimen, supporting the clinical relevance of exposure timing. These findings suggest careful PPI use during the induction phase and warrant validation in larger prospective cohorts. Clinical trial information: 031200088.