Association between opioid use and treatment discontinuation in patients receiving nivolumab: a real-world database study.

PURPOSE: We evaluated the association between baseline opioid use and time to treatment discontinuation (TTD) in patients receiving nivolumab monotherapy using a nationwide real-world database. METHODS: In this retrospective cohort study using the JMDC database (enrollment: 2015-2020; follow-up through 2022), patients with non-small cell lung, gastric, colorectal, or renal cell carcinoma initiating nivolumab were classified by baseline opioid prescriptions during the 90-day pre-index period. TTD associations were evaluated using multivariable Cox proportional hazards models. Propensity score matching (PSM) and 30- and 90-day landmark analyses were performed as sensitivity analyses. RESULTS: Among 1,879 patients (763 opioid users, 1,116 non-users), baseline opioid use was associated with shorter median TTD (112 vs. 284 days; adjusted hazard ratio [aHR], 1.93; 95% CI, 1.68-2.22; P < 0.001). This association remained consistent after PSM (aHR, 1.86; 95% CI, 1.59-2.17) and in landmark analyses. No significant interaction was observed across cancer types (P for interaction = 0.452). CONCLUSION: Baseline opioid use was associated with shorter treatment persistence among patients receiving nivolumab. Baseline opioid use may serve as a pragmatic real-world marker to identify patients at increased risk of early treatment discontinuation who may benefit from closer clinical assessment and supportive care.