One year on from the approval of lenvatinib plus pembrolizumab for previously treated non-microsatellite instability-high (MSI-H), mismatch repair (MMR)-proficient endometrial cancer, the phase II LEAP-005 study has reported that a number of other previously treated advanced solid tumour types also appear likely to respond to this combination (LBA41).
The first results of the open-label, multicohort study were presented as a Late-Breaking Abstract in a today’s Proffered Paper session at ESMO Virtual Congress 2020. A total of 187 patients ≥18 years old with disease meeting one of the six designated tumour categories – triple-negative breast (TNBC), ovarian, gastric cancer (GC), colorectal cancer (CRC) (non-MSI-H/mismatch repair proficient), glioblastoma multiforme (GBM), or biliary tract cancer (BTC, excluding ampulla of Vater) – received lenvatinib 20 mg/day plus pembrolizumab 200 mg every 3 weeks. Treatment was continued for 35 weeks or until confirmed disease progression or unacceptable toxicity. The tumour cohorts comprised second- or third-line TNBC, fourth-line ovarian cancer, third-line GC, third-line CRC (non-MSI-H/MMR proficient), second-line GBM and second-line BTC. Each cohort included 31 patients, except for the CRC cohort (n=32), with the intention to expand the number of patients treated up to 100 for cohorts in which sufficient efficacy was observed.
The primary endpoints were objective response rate (ORR) by blinded central review per RECIST v1.1 – or RANO, in the case of GBM – and safety.
At the data cut-off of 10 April 2020, the median study follow-up was 8.6 months (range 1.9–13.1). Efficacy was encouraging across all tumour types, with ORRs increasing from 10% in GC (95% confidence interval [CI] 2–26) and BTC (95% CI 2–26), to 16% (95% CI 6–34) in GBM, 22% (95% CI 9–40) in CRC, 29% (95% CI 14–48) in TNBC and 32% (95% CI 17–51) in ovarian cancer. Corresponding disease control rates were 48% in GC, 68% in BTC, 58% in GBM, 47% in CRC, 58% in TNBC and 74% in ovarian cancer. The median duration of response (DOR) among responders was 3.2 months in GBM, 5.3 months in BTC and, at the time of cut-off, had not been reached in the other cohorts. The longest DOR observed was 10.4+ months, recorded in a patient with CRC.
The toxicity of the lenvatinib–pembrolizumab combination was manageable. Grade ≥3 treatment-related adverse events (TRAEs) were most frequent in the ovarian cancer cohort (68%) and least common in the GBM cohort (35%). Rates in the other tumour types were 42% in GC, 48% in BTC, 50% in CRC and 55% in TNBC. The proportion of patients discontinuing treatment due to a TRAE was also highest in the ovarian cancer group (13%), followed by TNBC (10%) and CRC (9%); the remaining three cohorts (GC, GBM and BTC) all had discontinuation rates of 6%.
The study is ongoing and, given the promising activity observed across tumour types, all cohorts are being expanded.
Abstract and session details
- Lwin Z et al. LEAP-005: Phase 2 Study of Lenvatinib (Len) Plus Pembrolizumab (Pembro) in Patients (Pts) With Previously Treated Advanced Solid Tumors. ESMO Virtual Congress 2020, LBA41
Proffered Paper - Investigational immunotherapy, 20.09.2020, h. 14:25 – 16:05, Channel 2