Survival Outcomes Associated With Prolonged Pembrolizumab After Definitive Chemoradiation for Non-small Cell Lung...

The optimal duration of pembrolizumab after definitive chemoradiation for non-small cell lung cancer remains uncertain in routine practice. Although prospective trials support immunotherapy-based strategies in locally advanced disease, real-world treatment duration varies because of toxicity, progression, patient preference, access, and clinician decision-making. We evaluated survival and healthcare utilization outcomes associated with prolonged versus shorter pembrolizumab exposure after chemoradiation.

We conducted a retrospective cohort study using the TriNetX U.S. Collaborative Network, a de-identified electronic medical record database including 70 healthcare organizations. Adult patients with non-small cell lung cancer who received chemoradiation followed by pembrolizumab were identified. Prolonged pembrolizumab exposure was defined by documented pembrolizumab use during the 9- to 14-month window after initial pembrolizumab exposure, with evidence of healthcare follow-up at least 1 year after initiation. The shorter-duration cohort had pembrolizumab exposure within the first 6 months after initiation, no documented pembrolizumab use during the 9- to 14-month window, and evidence of healthcare follow-up at least 1 year after initiation. Propensity score matching was performed for demographic and clinical characteristics. The primary outcome was all-cause mortality. Secondary outcomes included hospital admission, critical care utilization, palliative care encounters, emergency care, and prolonged services.

The initial query identified 914 patients in the prolonged-exposure cohort and 471 patients in the shorter-duration cohort. In the propensity score matching population, 899 and 458 patients were available before matching, and 450 patients were matched in each cohort. For mortality analysis, 10 patients in each cohort were excluded because mortality was recorded before the analysis window, leaving 448 and 446 evaluable patients, respectively. All-cause mortality occurred in 166 of 448 patients receiving prolonged pembrolizumab and 207 of 446 patients receiving shorter-duration therapy, corresponding to risks of 37.1% and 46.4%, respectively. Prolonged pembrolizumab exposure was associated with lower mortality risk: risk difference -9.4% (95% CI, -15.8% to -2.9%; p = 0.005), risk ratio 0.798 (95% CI, 0.683-0.934), and odds ratio 0.680 (95% CI, 0.520-0.888). Median survival was 1505 days versus 791 days, respectively; hazard ratio 0.674 (95% CI, 0.550-0.827; log-rank p < 0.001). Hospital admission, critical care utilization, palliative care encounters, and emergency care did not differ significantly between cohorts.

In this real-world cohort, prolonged pembrolizumab exposure after chemoradiation was associated with improved overall survival without increased measured healthcare utilization. These findings should be interpreted as hypothesis-generating because treatment duration may reflect survivorship, treatment tolerance, disease biology, performance status, PD-L1 expression, response to therapy, and other unmeasured confounders. Further studies using landmark analyses, time-dependent exposure modeling, and prospective treatment-duration assessment are needed.