Abstract
PURPOSE: Different combination modalities between an anti-PD-1/PD-L1 agent and a platinum-based chemotherapy or another checkpoint inhibitor (with or without a short course or full course of a platinum doublet) proved superior to chemotherapy alone in multiple clinical trials, but these strategies were not directly compared. The aim of this study is to report the real-world data results with different immunotherapy combinations in a series of patients treated in consecutive cohorts at the Ion Chiricuță Oncology Institute. METHODS: A total of 122 patients were successively enrolled in three cohorts: (1A) nivolumab + ipilimumab (18 patients), (1B) nivolumab + ipilimumab + short-course chemotherapy (33 patients), and (2) pembrolizumab plus full-course chemotherapy (71 patients). Endpoints included overall survival (OS), progression-free survival (PFS), objective response (ORR), and univariate and multivariate exploratory analysis of prognostic factors. RESULTS: Median follow-up in the consecutive cohorts 1A, 1B, and 2 was 83 versus 59 versus 14.2 months. Median OS and PFS for all patients were 22.2 and 11.5 months, respectively, and 2-year actuarial OS and PFS were 49% and 35%, respectively. For the nivolumab + ipilimumab (cohorts 1A and 1B) versus pembrolizumab combinations (cohort 2), median OS was 14 vs. 24.8 months (p = 0.18) and 2-year actuarial survival 42% vs. 53%; median PFS was 8.6 vs. 12.7 months (p = 0.41) and 2-year actuarial PFS 34% vs. 35%; response rates were 33.3% vs. 47.9% (p = 0.22). Older age, impaired PS (2 versus 0-1), corticotherapy in the first month of immunotherapy, and >3.81 neutrophils to lymphocytes ratio were independent unfavorable prognostic factors in the multivariate analysis of survival (limited to 2 years follow-up). The 5-year long-term survival was 30.5% and 18.8% for cohorts 1A and 1B, respectively (not enough follow-up for cohort 2). CONCLUSIONS: Efficacy results using different immunotherapy combination strategies were promising and not significantly different between protocols at 2 years. Real-world efficacy and long-term results in our series were in line with those reported in the corresponding registration trials.