Combination of vemurafenib with cobimetinib in BRAF V600E/K mutated melanoma patients to normalize LDH and optimize immunotherapy with nivolumab and ipilimumab (COWBOY): A randomized, open-label phase 2 clinical trial.

9513 Background: Elevated levels of serum lactate dehydrogenase (LDH) in patients (pts) with metastatic melanoma are associated with reduced response to immune checkpoint inhibitors (ICIs). In a phase 2 trial, we investigated whether upfront reduction of tumor burden and normalization of LDH through six weeks of BRAF/MEK inhibitor (BRAF/MEKi) induction, prior to the initiation of ICIs, could improve response rates compared to standard first-line dual ICI. Methods: In this two-arm, multicenter, phase 2 trial, treatment naïve pts with unresectable stage III or stage IV BRAFV600E/K mutation-positive melanoma and elevated baseline LDH levels (< 5x upper limit of normal) were randomized 1:1 (stratified by LDH) to one of two treatment arms. Arm A received 6 weeks of induction with vemurafenib 960 mg bid + cobimetinib 60 mg QD on a 21/7-day schedule, directly followed by four courses of ipilimumab 3 mg/kg q3wk (ipi3) plus nivolumab 1 mg/kg q3wk (nivo1). Arm B directly started with four courses of ipi3 + nivo1. Both arms received response-adapted nivolumab maintenance. The primary endpoint was to compare the best overall response rate (BORR) within 18 weeks from the start of treatment between two treatment arms, as assessed by RECIST 1.1. Secondary endpoints included progression-free survival (PFS), overall survival (OS), and the rate of grade 3/4 toxicity. Exploratory endpoints focused on circulating tumor DNA dynamics (ctDNA), blood transcriptome, and tumor immunoprofiling. ClinicalTrials.gov identifier: NCT02968303. Results: Between 2017-2021, 71 pts were randomly assigned, 36 to arm A and 35 to arm B. The study was prematurely terminated due to slow patient accrual (planned accrual: 200 pts). The median follow-up time was 27.9 months (IQR 16.8-57.8). In arm A, the BOR was partial response (PR) in 75% of the pts (36.1% had an ongoing response after treatment switch), stable disease (SD) in 22.2%, and progressive disease (PD) in 2.8%. In arm B, 54.3% had PR, 22.9% SD, and 20% PD (1 pt was not evaluable). BRAF/MEKi induction prior to ICI was associated with a significantly shorter median PFS as compared with upfront dual ICI (4.4 months vs 25.0 months, HR 0.52, 95% CI 0.30-0.91; p=0.02). Median OS was also shorter in arm A (13.5 months), whereas the median OS was not reached in arm B (HR 0.38, 95% CI 0.19-0.68; p=0.002). Treatment-related grade ≥3 adverse events occurred in 25 of 36 pts (69%) in arm A and 15 of 35 pts (43%) in arm B. When combining both arms, pts with undetected ctDNA at 6-12 weeks after ICI initiation had significantly longer PFS than those with detected ctDNA (HR 2.05, 95% CI 1.05-4.0; p=0.03). Conclusions: BRAF/MEKi induction prior to dual ICI did not improve response rates to ICI and was associated with worse PFS and OS in pts with BRAFV600E/K mutated melanoma and elevated LDH compared to upfront dual ICI. Clinical trial information: NCT02968303 .