In the final analysis of the phase III LEAP-017 study, the combination of lenvatinib plus pembrolizumab failed to improve outcomes over the standard of care in previously treated patients with metastatic colorectal cancer that lacked high microsatellite instability or mismatch repair deficiency (non–MSI-H/dMMR).1
“While there was a trend toward longer overall survival with lenvatinib plus pembrolizumab in the final analysis, the combination did not meet the prespecified threshold for statistical significance for the overall survival primary endpoint,” said Akihito Kawazoe, MD, PhD, of the National Cancer Center Hospital East in Kashiwa, Japan, who reported the late-breaking findings at the European Society for Medical Oncology (ESMO) World Congress on Gastrointestinal Cancer 2023.
“Some separation of the overall survival curve was seen after 12 months, suggesting a possible delayed treatment effect with lenvatinib/pembrolizumab….”— Akihito Kawazoe, MD, PhD
Tweet this quote
The multiple-receptor tyrosine kinase inhibitor lenvatinib and the checkpoint inhibitor pembrolizumab were combined and compared against two standards of care for previously treated metastatic non–MSI-H/dMMR tumors: regorafenib and trifluridine/tipiracil. Both are standard treatment options for patients who relapse after initial treatment with fluorouracil-based regimens with or without targeted agents, he said.
LEAP-017 randomly assigned 480 patients to the combination arm or standard treatment arm. Patients in the combination arm received lenvatinib at 20 mg daily plus pembrolizumab at 400 mg every 6 weeks; pembrolizumab could be continued for up to 2 years alone or with lenvatinib for more than 25 cycles. The control arm received the investigator’s choice of treatment, either regorafenib at 160 mg daily on a 4-week cycle or trifluridine/tiparicil at 35 mg/m2 on a 4-week cycle. Selection between the two regimens was roughly equal.
At baseline, 70% of patients in both arms had liver metastases; 35% of the combination arm and 40% of the standard treatment arm had a combined positive score (CPS) of at least 1; 41% and 46%, respectively, had BRAF/RAS wild-type tumors; 58% and 54% had RAS mutations; and 2% and 5% had BRAF mutations.
For statistical significance in the final analysis, the overall survival efficacy threshold was .0124; for progression-free survival and objective response rate, significance at the interim analysis was set at .0125. The latter two secondary endpoints were to be tested only if the overall survival null hypothesis was rejected.
Improvements Over Standard of
Care Were ‘Trends’
At a median follow-up of 18.6 months, median overall survival was 9.8 months with the combination and 9.3 months with standard therapy, and the 12-month survival rates were 42.7% and 40.4%, respectively (hazard ratio [HR] = 0.83; P = .0379; 95% confidence interval [CI] = 0.68–1.02). The threshold for showing superiority was .0214; therefore, the primary endpoint was not met. A favorable trend was shown for virtually all key subgroups, but the confidence intervals crossed 1, Dr. Kawazoe reported.
“Some separation of the overall survival curve was seen after 12 months, suggesting a possible delayed treatment effect with lenvatinib/pembrolizumab, which is consistent with previous trials of immune checkpoint inhibitors in other malignancies. However, frequent censoring beyond this time point limits the interpretation and limits drawing definitive conclusions,” he said.
In the final analysis, median progression-free survival was 3.8 months vs 3.3 months, respectively (HR = 0.69; 95% CI = 0.56–0.85). Again, hazard ratios favored the combination in all key subgroups, with somewhat stronger hazard ratios than were seen for overall survival.
Control Arm Performed Better Than Expected
Dr. Kawazoe pointed out that the control arm performed better than expected for reasons not yet understood. “There is no clear imbalance in the baseline characteristics that could explain the overperformance of the control arm in terms of overall survival,” he said.
Although median overall survival in the control arm was 9.3 months in his study, it was 6.4 months in the CORRECT trial of regorafenib2 and 7.1 months in the RECOURSE trial of trifluridine/tipiracil.3
“Longer overall survival in LEAP-017, for standard therapy, could also be impacted by the relatively higher rate of subsequent anticancer therapy, particularly subsequent chemotherapy, compared with previous trials in salvage lines for metastatic colorectal cancer,” he suggested.
Numerical Improvements in Response
The confirmed response rate was 10.4% vs 1.7%—an 8.7% increase with lenvatinib/pembrolizumab—and the median duration of response was 11.1 months vs 7.6 months, respectively. The disease control rate was 63.1% with the combination vs 52.7% with standard therapy.
The greatest differences in response, favoring the combination, were seen in patients with a CPS of at least 1 (+10.9%), patients older than age 65 (+10.1%), patients with an Eastern Cooperative Oncology Group performance status of 0 (+10.9%), patients without liver metastases (+17.7%), and patients with RAS mutations (+13.5%), he reported.
“Novel therapeutic options for patients with previously treated non–MSI-H/dMMR metastatic colorectal cancer remain an area of unmet need.”— Akihito Kawazoe, MD, PhD
Tweet this quote
Grade ≥ 3 drug-related adverse events were observed in 58.4% receiving the combination and 42.1% treated with the standard of care. Two patients on the combination and none receiving standard therapy experienced a fatal drug-related adverse event. Approximately 46% vs 59% of patients, respectively, received poststudy anticancer therapy.
Based on the disappointing findings of LEAP-017, Dr. Kawazoe commented: “Novel therapeutic options for patients with previously treated non–MSI-H/dMMR metastatic colorectal cancer remain an area of unmet need.”
DISCLOSURE: Dr. Kawazoe has received honoraria from Taiho Pharmaceuticals, Eli Lilly, and Daiichi Sankyo; has served as an advisor or consultant to Zymeworks and Merck & Co; and has received institutional research grants or funding from Merck & Co.
1. Kawazoe A, Xu R, Passhak M, et al: Lenvatinib plus pembrolizumab vs standard of care for previously treated metastatic colorectal cancer: The phase 3 LEAP-017 study. ESMO World Congress on Gastrointestinal Cancer 2023. Abstract LBA-5. Presented July 1, 2023.
2. Grothey A, Van Cutsem E, Sobrero A, et al: Regorafenib monotherapy for previously treated metastatic colorectal cancer (CORRECT): An international, multicentre, randomised, placebo-controlled, phase 3 trial. Lancet 381:303-312, 2013.
3. Van Cutsem E, Mayer RJ, Laurent S, et al: The subgroups of the phase III RECOURSE trial of trifluridine/tipiracil (TAS-102) versus placebo with best supportive care in patients with metastatic colorectal cancer. Eur J Cancer 90:63-72, 2018.
Cathy Eng, MD, FACP, FASCO
Cathy Eng, MD, FACP, FASCO, the David H. Johnson Chair in Surgical and Medical Oncology, Professor of Medicine, Hematology and Oncology, Co-Director of GI Oncology and Co-Leader of the Gastrointestinal Cancer Research Program at Vanderbilt-Ingram Cancer Center,...