Lenvatinib plus pembrolizumab versus standard of care for previously treated metastatic colorectal cancer (mCRC): the phase 3 LEAP-017 study

Effective treatment options remain an unmet need for patients (pts) with previously treated non-microsatellite instability-high/mismatch repair deficient (non-MSI-H/dMMR) mCRC. We present results from the global, randomized, open-label phase 3 LEAP-017 study (NCT04776148) evaluating the efficacy and safety of lenvatinib + pembrolizumab versus standard-of-care (SOC) in previously treated non-MSI-H/dMMR mCRC. Eligible pts aged ≥18 years with unresectable, non-MSI-H/dMMR mCRC that had progressed on or was intolerant to prior treatment including fluoropyrimidine, oxaliplatin, irinotecan and targeted therapies such as anti-VEGF/EGFR mAbs, or BRAF-inhibitors, were randomized 1:1 to lenvatinib 20 mg orally QD + pembrolizumab 400 mg IV Q6W or investigator’s choice (selected before randomization) of oral regorafenib 160 mg QD (d 1-21, no dose on d 22-28 [cycle 1 dose escalation from 80 mg QD permitted as per local guidelines]) Q4W or trifluridine/tiparicil 35 mg/m2 Q4W (BID on d 1-5 and d 8-12, no doses on d 6-7 or d 13-28 [SOC]). Randomization was stratified by presence/absence of liver metastases. Treatment with pembrolizumab could continue for up to 2 years, and with lenvatinib until progression or discontinuation. The primary endpoint was OS. Key secondary endpoints included PFS, and ORR per RECIST v1.1 by BICR. The protocol specified one interim analysis (IA) and one final analysis (FA) for OS. Prespecified efficacy boundaries were one-sided p=0.0214 for OS (FA), and one sided p=0.0125 for PFS and ORR (IA). At FA (data cut-off Feb 20, 2023), 480 pts were randomized (241 lenvatinib + pembrolizumab; 239 SOC). Median age of pts was 58 years (range, 21-87), and 70% had liver metastases at randomization. After median follow-up of 18.6 months (mo), 366 OS events had occurred and 22 pts receiving lenvatinib + pembrolizumab and none receiving SOC remained on treatment. Median OS was 9.8 vs 9.3 mo (HR 0.83; 95% CI, 0.68-1.02; p=0.0379), and median PFS was 3.8 vs 3.3 mo (HR 0.69; 95% CI 0.56-0.85). Confirmed ORR was 10.4% with lenvatinib + pembrolizumab vs 1.7% with SOC. The median (range) DOR was 11.1 (3.7 to 18.5+) vs 7.6 mo (5.6+ to 9.0). At IA (data cut-off Jul 21, 2022) with median follow-up of 11.6 mo, median PFS was 3.8 vs 3.3 mo (HR 0.71; 95% CI 0.57-0.87) and ORR was 10.4% vs 1.7%. Grade ≥3 drug-related AE rates were 58.4% with lenvatinib + pembrolizumab vs 42.1% with SOC. Two pts receiving lenvatinib + pembrolizumab and none receiving SOC had a grade 5 drug-related AE. Approximately 46% vs 59% of pts, respectively, received post-study anti-cancer therapy. Treatment with lenvatinib + pembrolizumab versus SOC trended toward longer OS, PFS, and ORR, but did not meet prespecified threshold for statistical significance, in pts with previously treated non-MSI-H/dMMR mCRC. No new safety signals were observed.