Concomitant medications on survival in patients treated with immune checkpoint inhibitors
The effectiveness of immune checkpoint inhibitors (ICIs) depends on systemic immune homeostasis, which may be modified by concomitant medications. Although certain drugs have been associated with altered outcomes, a comprehensive evaluation of how a broad range of commonly used medications affects survival outcome remains lacking. Identifying medications associated with impaired survival is an important unmet need for optimizing ICI therapy. We systematically evaluated associations between 25 concomitant medication classes and short-term mortality among patients receiving ICIs. Using nationwide healthcare data in South Korea, we conducted a nested case–control study within a cohort of patients aged ≥18 years with cancer who newly initiated ICIs between August 2017 and December 2023. Cases were deaths occurring within 1 year of ICI initiation and were matched to one event-free control on age, sex, and date of ICI initiation (±90 days). Concomitant medication exposure across 25 drug classes was independently assessed during the 90 days before and 30 days after ICI initiation. Adjusted odds ratios (ORs) and 95% confidence intervals (CIs) were estimated using conditional logistic regression with adjustment for covariate imbalances after matching.Among 30,201 matched case–control pairs (mean age 68.1 years; 77.5% male; 44.9% receiving atezolizumab; 64.5% with lung cancer), the most frequently used medications (>25%) were corticosteroids, antibiotics, acid suppressants, nonsteroidal anti-inflammatory drugs, anxiolytics/hypnotics/sedatives, and diuretics. Use of these medications was associated with higher 1-year mortality, with the largest ORs for diuretics (OR 1.29, 95% CI 1.24–1.34 before ICI initiation; OR 2.34, 95% CI 2.25–2.45 after ICI initiation). Some less commonly used medications (< 10%) were also associated with increased mortality, including immunosuppressants (OR 1.51, 95% CI 1.32–1.71; OR 1.77, 95% CI 1.51–2.08), antipsychotics (OR 1.33, 95% CI 1.24–1.43; OR 1.53, 95% CI 1.47–1.59), direct oral anticoagulants (OR 1.18, 95% CI 1.10–1.27; OR 1.44, 95% CI 1.34-1.54). Angiotensin receptor blockers were associated with lower mortality (OR 0.93, 95% CI 0.90–0.97; OR 0.94, 95% CI 0.91–0.98). Similar associations were observed for 6-month and 1-year mortality. Subgroup analyses across major cancer types, including lung, liver, and urinary tract cancers, showed consistent results. Concomitant medication use around ICI initiation may influence treatment outcomes. As the use of ICIs expands, commonly used medications showing potential mortality signals should be prioritized for follow-up studies, and careful medication review in clinical practice may be warranted.
2026 Spring Convention