The use of proton pump inhibitors and antibiotics appears to attenuate the benefit of immunotherapy and adversely impact survival in patients with unresectable stage III NSCLC.
Findings from a study published in Lancet Oncology [link here] suggest “microbiome–immunotherapy interactions” need to be further investigated.
“The future work should focus on pragmatic antibiotic stewardship and on identifying patients in whom alternative management strategies might be reasonable,” it said.
“Microbiome-based classifiers and mitigation approaches, including dietary modulation, probiotics, and faecal microbiota transplantation, remain investigational and should be tested prospectively.”
The study was a post-hoc analysis on data from the phase 3 PACIFIC RCT of durvalumab versus placebo as consolidation therapy in patients with unresectable stage III NSCLC who had not progressed after completion of concurrent chemoradiotherapy.
Included in the analysis was data from 660 patients who had received durvalumab 10 mg/kg or matched placebo as a 60-min intravenous infusion every 2 weeks for up to 12 months or until disease progression.
About 40% of patients had been exposed to PPIs at baseline and 10% to antibiotics.
After a median follow-up of 62.4 months, the PACIFIC trial found durvalumab delivered improved PFS (14.4 v 8.2 months) and OS (47.7 v 27.8 months) compared to placebo.
However, baseline exposure to PPIs was associated with significantly shorter PFS than no PPI exposure in the durvalumab arm (9·4 v 17·2 months; HR 1·57; p<0·0001) and similarly for OS (33.0 v 57.9 months; HR 1.66; p<0.0001).
“Conversely, in the placebo group, no significant differences in survival outcomes were observed according to proton pump inhibitor exposure,” it said.
Similarly, baseline exposure to antibiotics in the durvalumab arm was associated with shorter PFS than no antibiotic exposure (9·2 v 15.6 months; HR 1·50; p=0·016). However there was no significant change in the OS based on antibiotic exposure (37.7 v 49.2 months; HR 1.33; p=0.16).
There was also no significant change in PFS or OS in the placebo arm based on antibiotic exposure.
A multivariable sensitivity analysis confirmed that exposure to PPIs was independently associated with an increased risk of disease progression (HR 1·65) and death (HR 1.73) compared to patients with no exposure in the durvalumab group, while no effect was reported among patients in the placebo group.
“In the second-tier multivariable sensitivity analysis, exposure to antibiotics lost the significant association with progression-free survival (HR 1·40 [95% CI 0·98–2·00]; p=0·061), with no impact on overall survival,” the study said.
The study also found that patients with exposure to both PPIs and antibiotics had the shortest PFS and OS, compared to those who had no exposure to either drug class and those who only had exposure to a single drug class.
“To our knowledge, this is the first analysis within a randomised trial population to describe this pattern in patients with unresectable stage III NSCLC treated with consolidation durvalumab,” the investigators said.
“This finding is clinically relevant because proton pump inhibitors and antibiotics are frequently prescribed in the peri-concurrent chemoradiotherapy setting.”
The study also identified baseline dysbiosis, as measured by Toposcore, was associated with inferior PFS and was directionally enriched in patients with antibiotic exposure.
The investigators said the exploratory findings provide some preliminary mechanistic support for a biological framework of “reduced microbial diversity, depletion of immune-modulating taxa, and impaired antitumour immune priming in the setting of dysbiosis.”
“Overall, these results contribute to growing evidence that concomitant medications might modulate the efficacy of immune checkpoint inhibitors…” they concluded.
“Prospective validation with harmonised exposure definitions, clearly defined temporal windows, and integrated microbiome profiling will be needed before translating these findings into clinical decision making.”
Independent comment
A comment article in the journal [link here] said the findings expand on the growing literature linking microbiome-disrupting medications to impaired outcomes from immune checkpoint inhibitors and strengthen the case for altering co-medication use in patients.
“These data support a pragmatic shift: judicious use of proton pump inhibitors, active antibiotic stewardship at each decision point, and minimising avoidable polypharmacy in the absence of robust data on how most drugs affect the gut microbiome,” it said.
However it was possible the association could be explained by indication bias – the underlying reasons for using the medications – rather than a direct effect of the medications.
Regardless, it said the findings and magnitude of the effects warrant further interventional studies.
“The present study by Brunetti and colleagues, as well as extensive related literature, tell us that we urgently need a clear clinical research roadmap to translate these associations into actionable practice,” it said.
“Prospective studies with standardised exposure definitions and pre-specified timing windows should validate whether proton pump inhibitors and antibiotics attenuate the benefit of immunotherapy or mark higher-risk patients.”