Abstract
Background: Treatment-free survival (TFS) characterizes disease control after discontinuation of immune checkpoint inhibitors (ICIs) until subsequent therapy or death. We previously evaluated TFS in a pooled analysis of the CheckMate 067 and CheckMate 069 trials of the ICIs nivolumab and ipilimumab, alone or in combination, in patients with advanced melanoma after minimum follow-up of 36 months. This analysis investigated TFS differences between treatments in CheckMate 067 after a minimum follow-up of 60 months, and their relation to overall survival (OS) differences. Methods: Data were from 937 patients who initiated treatment (nivolumab plus ipilimumab, nivolumab, or ipilimumab) in CheckMate 067 (NCT01844505). TFS was defined as the area between the Kaplan-Meier curves for time to protocol therapy cessation and time to subsequent systemic therapy initiation or death, each measured from randomization. TFS was partitioned as time with and without toxicity. Toxicity included persistent and late-onset grade ≥2 select treatment-related adverse events (ie, those of potential immunologic etiology). The area between Kaplan-Meier curves was estimated by the difference in 60-month restricted-mean times of the endpoints. Between-group differences were estimated with bootstrapped 95% CIs. Results: At 60 months from randomization, 39%, 24%, and 11% of patients assigned to treatment with nivolumab plus ipilimumab, nivolumab, and ipilimumab, respectively, had survived and were treatment-free. The 60-month mean TFS was approximately twice as long with the combination (19.7 months) than with nivolumab (9.9 months; absolute difference, 9.8 (95% CI 6.7 to 12.8)) or ipilimumab (11.9 months; absolute difference, 7.8 (95% CI 4.6 to 11.0)). In the respective groups, mean TFS represented 33% (8% with and 25% without toxicity), 17% (2% and 14%), and 20% (3% and 17%) of the 60-month period. Compared with 36-month estimates, mean TFS over the 60-month period represented slightly greater percentages of time in the nivolumab-containing regimen groups and a lesser percentage in the ipilimumab group. TFS differences between the combination and either monotherapy increased with longer follow-up. Conclusions: Along with improved long-term OS with the nivolumab-containing regimens versus ipilimumab, TFS without toxicity was sustained with nivolumab plus ipilimumab versus either monotherapy, demonstrating larger between-group differences with extended follow-up.