219 Long-term clinical outcomes associated with sequential treatment of BRAF mutant advanced melanoma patients

Background Patients with BRAF mutant advanced melanoma can be treated sequentially with immunotherapies (IO) and BRAF+MEK inhibitors. We evaluated the clinical outcomes associated with various treatment sequences for BRAF mutant advanced melanoma based on the 5-year follow-up data from clinical trials. Methods In the absence of head-to-head trial data, a matching-adjusted indirect comparison (MAIC) was conducted for IO vs. BRAF+MEK inhibitors, using the longest follow-up available in the published literature. Multivariate risk equations were developed to predict time-to-event outcomes based on patient-level data from pooled CheckMate-067 &-069 trials. Risk equations were inserted into a discrete event simulation to estimate the average life-years (LYs) and quality-adjusted life-years (QALYs) that can be gained with various treatment sequences over a lifetime horizon. Treatment sequences and corresponding efficacy data sources are presented below (table 1). Utility weights for quality-adjustment of LYs were obtained from published literature. Results Treatment sequences starting with IO followed by BRAF+MEK were associated with 2.9–4.3 years of additional survival and 2.2–3.3 years of quality-adjusted survival versus sequences starting with BRAF+MEK followed by anti-PD-1. After 1L IO, the time spent in the treatment-free interval (TFI) is 3.3–5.0 years. LYs, QALYs, and time spent in TFI were higher with sequences starting with anti-PD-1+anti-CTLA-4 vs. anti-PD-1 alone. Conclusions In this sequencing model with 5-year data from randomized clinical trials, initiating 1L treatment with IO provided prolonged survival compared to initiating 1L treatment with BRAF+MEK. Time spent in TFI represents a significant proportion of survival time for patients on IO initiating sequences. Limitations of the study are the reliance on published information for BRAF+MEK, which could lead to biases due to unmeasured differences in the patient populations or trial conduct and the absence of data on 2L combination IO. Anti-PD-1+anti-CTLA-4 as second line option has not been included because of a lack of clinical evidence. Findings from this analysis will require validation in ongoing prospective randomized clinical trials.