Background: Data on long-term survivors with advanced non-small-cell lung cancer (NSCLC) treated with nivolumab are available from randomized trials. Characteristics, management, and healthcare resources of those patients need to be confirmed with real-world data.
Methods: The UNIVOC retrospective observational study included all patients with advanced NSCLC recorded in the French national hospital database starting nivolumab in 2015 and followed them until December 2020. The Kaplan-Meier method estimated the overall survival (OS). A machine learning approach identified patients with similar treatment sequences.
Results: Within the 3,050 patients who had nivolumab initiation,5-year OS rate was 14.6 % (95 %CI 13.3 %-16.2 %). In total, data covering at least 5 years of follow-up were retrieved for 231 surviving patients. Survivors were younger, often female and had fewer comorbidities than non-survivors. Three clusters of patients with different nivolumab treatment durations were identified: 1/ Continuous nivolumab treatment; 2/ Long period of nivolumab treatment followed by chemotherapy or no treatment; 3/ Short period of nivolumab treatment then chemotherapy or no treatment. At 5 years, 61.0 % of survivors were no longer receiving systemic therapy, 26.4 % were treated with nivolumab, 8.7 % chemotherapy, and 3.9 % other immunotherapies. Among 5-y survivor patients, the average number of hospitalisations per patient decreased from 23.4 to 12.8 between the 1st and the 5th year. In the 5th year, 46 % of patients had no more hospitalization for lung cancer.
Conclusions: This large nationwide study confirms the long-term benefit of nivolumab treatment for advanced NSCLC patients in the real-world setting, with a 5-year survival rate similar to that reported in clinical trials.
Keywords: Immunotherapy; Long-term survivors; Machine learning; Non-small cell lung cancer; PMSI.
Copyright © 2023 The Author(s). Published by Elsevier Masson SAS.. All rights reserved.