Differential impact of proton pump inhibitors and antibiotics on immunotherapy efficacy after chemoradiotherapy in locally advanced non-small-cell lung cancer: a post-hoc analysis of the PACIFIC trial

BACKGROUND: Baseline exposure to antibiotics and proton pump inhibitors has been associated with reduced efficacy of immune checkpoint inhibitors in patients with advanced tumours, possibly through gut microbiome disruption. Whether this outcome extends to those with earlier-stage disease remains unclear. We aimed to assess the association of baseline antibiotics and proton pump inhibitors with progression-free survival and overall survival in patients with unresectable stage III non-small cell lung cancer (NSCLC). METHODS: PACIFIC was a randomised, double-blind, placebo-controlled phase 3 trial done in patients aged 18 years or older with unresectable stage III squamous or non-squamous NSCLC, WHO performance status 0-1, and no progression after two or more cycles of concurrent chemoradiotherapy. Patients were randomly assigned (2:1) to durvalumab 10 mg/kg intravenously every 2 weeks for up to 12 months or placebo, starting 1-42 days after chemoradiotherapy; patients were stratified by age, sex, and smoking history. This post-hoc analysis was based on the final 5-year data cutoff date of the completed trial and included the treated population with consent for exploratory analyses. Co-primary endpoints were progression-free survival and overall survival, assessed according to baseline exposure to proton pump inhibitors and systemic antibiotics. This trial is registered on ClinicalTrials.gov (NCT02125461). FINDINGS: Between May 9, 2014, and April 22, 2016, 713 patients were randomly assigned; 660 were included in this post-hoc analysis, of whom 449 received durvalumab and 211 received placebo; 203 (30·8%) were female and 453 (68·6%) were male. Race was reported as Asian in 153 (23·1%) patients, Black or African American in five (0·7%), White in 424 (64·2%), and unknown in 78 (11·8%). Baseline proton pump inhibitor exposure was recorded in 263 (40%) of 660 patients and antibiotic exposure was recorded in 69 (10%). Median follow-up in the pooled population was 62·4 (IQR 61·9-63·2) months. In the durvalumab group baseline exposure to proton pump inhibitors was associated with shorter progression-free survival (9·4 months [95% CI 7·6-13·7] vs 17·2 months [15·4-23·2]; hazard ratio [HR] 1·57 [95% CI 1·28-1·93]; p<0·0001) and overall survival (33·0 months [95% CI 21·9-46·7] vs 57·9 months [48·7-not computable (NC)]; HR 1·66 [95% CI 1·30-2·13]; p<0·0001) compared to no exposure to proton pump inhibitors, while baseline exposure to antibiotics was associated with shorter progression-free survival (9·2 months [95% CI 4·9-18·1] vs 15·6 months [13·6-17·6]; HR 1·50 [95% CI 1·08-2·10]; p=0·016) compared to no exposure to antibiotics, but there was no significant change in overall survival (37·7 months [95% CI 18·8-NC; 28 events] vs 49·2 months [39·7-57·3]; HR 1·33 [95% CI 0·90-1·97]; p=0·16). In the placebo group, neither proton pump inhibitor exposure nor antibiotic exposure was associated with changes in progression-free survival and overall survival. Interactions between treatment and proton pump inhibitors for progression-free survival (p=0·023) and overall survival (p<0·0001) were significant, but not for antibiotics. INTERPRETATION: Baseline exposure to proton pump inhibitors and antibiotics was associated with inferior outcomes with durvalumab, but not with placebo, consistent with potential attenuation of the benefit of durvalumab with proton pump inhibitors and antibiotics in patients with unresectable stage III NSCLC. FUNDING: None.