Antibiotic exposure is associated with inferior outcomes of immune checkpoint inhibitor (ICI) therapy in patients with non-small cell lung cancer (NSCLC), according to a meta-analysis presented at the Society for Immunotherapy of Cancer (SITC) 2021 Annual Meeting.1

Prior studies have suggested that the intestinal microbiota may modulate response to ICIs. For example, tumor-bearing mice responded more favorably to ICI therapy when colonized with feces from malignant melanoma patients who responded to ICI treatment, compared with feces from non-responders.2

In clinical studies, antibiotic exposure resulting in microbiome dysbiosis adversely influenced outcomes of cancer patients treated with ICIs, especially those with NSCLC.3 However, most studies were conducted retrospectively and involved small patient numbers.


Continue Reading

With that in mind, Athena Crespin, of Da Volterra in Paris, and colleagues conducted a meta-analysis to investigate the association between antibiotic use and outcomes in NSCLC patients treated with ICIs.

The researchers previously published their findings in 2020.4 At SITC 2021, Ms Crespin presented updated data encompassing studies published in journals or presented at conferences through September 2021.1

Study Details and Results

The meta-analysis included studies that reported a hazard ratio or Kaplan-Meier curve for overall survival (OS) or progression-free survival (PFS) based on antibiotic exposure and/or data on treatment response, such as objective response rate (ORR) and progressive disease (PD) rate according to antibiotic exposure.

There were 36 studies involving 12,304 patients that included data on OS and 29 studies involving 5425 patients that included data on PFS.

In those studies, patients treated with ICIs who were exposed to antibiotics had significantly worse OS and PFS. The pooled hazard ratios were 1.64 for OS (95% CI, 1.38-1.94) and 1.52 for PFS (95% CI, 1.29-1.80).

About 51% of patients in the OS analysis received antibiotics within 2 months before starting ICI therapy (n=6244), and 64% received antibiotics within 2 months before or after ICI initiation (n=7859).

The hazard ratios for OS were significantly worse for patients who received antibiotics during those 2 time windows, in comparison with patients who did not receive antibiotics at all or patients who received antibiotics outside those time windows.

There were 11 studies involving 2061 patients that included ORR data and 11 studies involving 1341 patients that included PD data. The odds ratios for ORR and PD were 0.60 (95% CI, 0.37-0.95) and 1.99 (95% CI, 1.45-2.7), respectively.

These were both statistically significant, reflecting decreased odds of treatment response and almost a 2-fold increased odds of PD in NSCLC patients treated with ICIs and exposed to antibiotics.