Common Medications Linked to Worse Survival in Lung Cancer Patients Receiving Keytruda

Use of certain antibiotics, steroids and proton pump inhibitors (PPIs) at the start of immunotherapy was associated with shorter survival in patients with advanced non-small cell lung cancer (NSCLC), according to a nationwide cohort study from France published Sept. 10, 2025, in JAMA Network Open.

Immunotherapy drugs such as Keytruda (pembrolizumab) have transformed the outlook for people with advanced NSCLC, both as stand-alone treatment and in combination with chemotherapy. However, questions remain about how other medications may influence their effectiveness. Earlier research has raised concerns that certain common medications might reduce the benefits of immunotherapy — for example, corticosteroids (e.g., prednisone) because they suppress the immune system, and antibiotics or proton pump inhibitors (PPIs, such as omeprazole) because they can alter gut bacteria. Past meta-analyses have linked antibiotic or PPI use with worse survival, although findings have not always been consistent, especially for steroids.

The new study was led by Adrien Rousseau, M.D., of Gustave Roussy, Paris-Saclay University, with colleagues in Villejuif, France. The investigators analyzed national health care data to assess whether commonly prescribed medications influence outcomes with Keytruda for advanced NSCLC.

The cohort included 41,529 patients who initiated Keytruda treatment between January 2015 and December 2022. Of these, 27,826 (67%) were men, and the median age was 65 years (range 19–97). More than one-third (35.7%) received Keytruda alone, while 64.3% received the drug in combination with chemotherapy.

Concomitant medication use was common at treatment initiation: 41.9% of patients had received antibiotics, 59.1% received steroids, and 53.7% were prescribed PPIs. Investigators evaluated outcomes using inverse probability of treatment weighting to adjust for confounding factors.

The analysis showed that antibiotic use was generally linked to shorter survival, with patients exposed to these drugs having about an 8% higher risk of death. The effect depended on the type of antibiotic, with macrolides and penicillins not showing the same association.

Steroids had a more nuanced effect. Overall use was not tied to worse outcomes, but higher doses were. Specifically, more than 20 milligrams per day of prednisone equivalent for patients receiving Keytruda alone or more than 30 mg per day for those receiving Keytruda plus chemotherapy were linked to shorter survival.

Proton pump inhibitors were also linked to worse outcomes. Patients taking PPIs at Keytruda initiation had a 13% higher risk of death compared with those who were not.

The researchers concluded that survival in patients with advanced NSCLC was negatively affected by exposure to antibiotics, PPIs, and high-dose steroids at the start of Keytruda. “These findings suggest that comedications should be monitored carefully in patients with immunotherapy,” they wrote.

When antibiotics are necessary for serious infections, macrolides or penicillins may be preferable because they were not linked to worse outcomes, they suggested. Low-dose steroids should not delay or prevent immunotherapy, they wrote, since the negative association was seen only with higher doses. As for PPIs, the authors cautioned that these drugs should not be used for patients receiving immunotherapy unless clearly indicated for conditions such as gastric ulcers.