Conference Coverage

Antibiotics gut checkpoint inhibitor efficacy



– Antibiotic exposure in the month before cancer immunotherapy starts may hamper the efficacy of immune checkpoint inhibitors, investigators caution.

Dr. David J. Pinato of Imperial College, London Neil Osterweil/MDedge News

Dr. David James Pinato

A prospective study of 196 patients treated with immune checkpoint inhibitors for various cancers showed that the 29 patients who received antibiotics within 30 days of starting immunotherapy had significantly worse overall survival than patients without antibiotic exposure; this effect was seen across cancer types, reported David James Pinato, MD, PhD, from Imperial College London.

In contrast, concurrent antibiotic and checkpoint inhibitor use was not significantly associated with overall survival differences, he said at the American Society of Clinical Oncology (ASCO) – Society for Immunotherapy of Cancer (SITC): Clinical Immuno-Oncology Symposium.

“I think these data are quite interesting in showing an independent detrimental effect, both on response and survival, in unselected patients treated with immune checkpoint inhibitors in routine clinical practice,” Dr. Pinato said.

The data also suggest “the timing of antibiotic exposure is crucial,” he added. Antibiotic treatment concurrent with immunotherapy did not appear to affect prognosis. Alternatively, prior antibiotic therapy appeared to have “a sort of a priming effect towards the immune system.”

Broad-spectrum antibiotics can affect the diversity of the gut microbiome, which influences mucosal immunity, dendritic cell function, and antigen presentation. Alternatively, enrichment of the microbiome with several bacterial species can enhance the potency of checkpoint inhibitors by facilitating the process of tumor rejection, Dr. Pinato explained.

To see whether antibiotic disruption, or “dysbiosis” of the gut microbiome, could hinder responsiveness to checkpoint inhibitors regardless of the tumor site and whether there were time-dependent effects of antibiotic exposure on response to checkpoint inhibitors, the investigators conducted a prospective, observational study in 196 patients treated with checkpoint inhibitors for non–small cell lung cancer (NSCLC), melanoma, renal cell carcinoma, head and neck cancer, transitional cell carcinoma of the bladder, and other cancers.

The researchers defined prior antibiotic exposure as more than 30 days before the start of checkpoint inhibitor therapy and concurrent exposure as antibiotics begun on the first day of the first cycle of checkpoint inhibitor dosing.

Of the 196 patients, 29 had previously received antibiotics, and 68 received them concurrently. The most frequently prescribed antibiotics were beta-lactam agents given in a single, short course. Other classes of drugs, used in eight or fewer patients each, included quinolones, macrolides, sulfonamides, tetracyclines, aminoglycosides, and nitroimidazole.

Median overall survival for the entire cohort, one of two primary outcomes, was 2 months for patients who had received prior antibiotics and 26 months for patients with no prior exposure. This difference was similar for patients with NSCLC (2.5 vs. 26 months), melanoma (3.9 vs. 14 months), and other cancers combined (1.1 vs. 11.0 months; log-rank P less than .01 for all comparisons).

In multivariate analysis, only response to checkpoints inhibitors (complete vs. partial response, stable disease, or progression) and prior antibiotic exposure were significantly associated with survival. The hazard ratio for survival for patients who had not previously received antibiotics was 3.5 (P less than .001).

In contrast, concurrent antibiotic and checkpoint inhibitor use did not have a significant effect on survival.

An analysis of radiologic responses also showed that patients with prior antibiotic exposure had a significantly higher probability of primary disease progression than those without (81% vs. 44%; P less than .001). There were no associations, however, between specific classes of antibiotics or corticosteroid use.

The findings indicate that “certainly, mechanistic studies are required here, not just to investigate the prognostic role of antibiotic-mediated dysbiosis, but perhaps transform this into an actual driver of antitumor immunity,” Dr. Pinato concluded.

The study was internally supported. Dr. Pinato reported receiving grant funding from Merck and Bristol-Myers Squibb unrelated to the study, as well as honoraria from ViiV Healthcare.

SOURCE: Pinato DJ et al. ASCO-SITC, Abstract 147.

Next Article: