Background: Immune checkpoint inhibitor (ICPi) rechallenge could represent an attractive option in non-small-cell lung cancer (NSCLC), yet no sufficient data supporting this strategy are available. This retrospective observational multicenter national study explored the efficacy of anti-programmed cell death 1 (PD-1)/programmed death ligand 1 (PD-L1) rechallenge in advanced NSCLC patients, looking for potential clinical features associated with greater outcomes.
Patients and methods: We retrospectively collected data from 144 advanced NSCLC patients whose disease was rechallenged with ICPis after ≥ 12 weeks of discontinuation. The progression-free survival (PFS) and overall survival (OS) were calculated from first or second ICPi initiation to disease progression (PFS1 and PFSR, respectively), death, or last follow-up (OS1, OSR), respectively.
Results: The median (interquartile range) age was 63 (58-70) years. Most patients were male (67%) and smokers (87%). Most had adenocarcinomas (62%) and/or stage IV disease at diagnosis (66%). The best response at rechallenge was not associated with that under the first ICPi (P = 1.10-1). The median (95% confidence interval) PFS1 and PFSR were 13 (10-16.5) and 4.4 (3-6.5) months, respectively. The median (95% confidence interval) OS1 and OSR were 3.3 (2.9-3.9) and 1.5 (1.0-2.1) years, respectively. Longer PFSR and OSR were found in patients discontinuing first ICPi because of toxicity or clinical decision, those not receiving systemic treatment between the two ICPis, and those with good Eastern Cooperative Oncology Group performance status at rechallenge. Only performance status proved to affect outcomes at multivariate analysis.
Conclusion: Patients discontinuing first ICPi because of toxicity or clinical decision, those able to maintain a treatment-free period, and those with good performance status may be potential candidates for rechallenge.
Keywords: Efficacy; Immune checkpoint inhibitors; NSCLC; Re-challenge; Selection criteria.
Copyright © 2020 Elsevier Inc. All rights reserved.