The utilization of lenvatinib with pembrolizumab for the treatment of recurrent endometrial cancer.

Abstract

e17627 Background: The landscape for the treatment of microsatellite-stable-recurrent endometrial cancer has changed with the published results of the KEYNOTE-775, demonstrating an overall response rate of 37.2% with Pembrolizumab and Lenvatinib 20 mg. However, recent retrospective studies have suggested that a lower dose of Lenvatinib may be safer and equally efficacious. We aimed to retrospectively review our institutional data to compare the efficacy and toxicity of Lenvatinib at a low (≤10 mg) vs. high (>10 mg) dose. Methods: Retrospective review of patients with recurrent endometrial cancer who received ≥ 3 cycles of Pembrolizumab with Lenvatinib. A Univariate analysis was used to estimate the 1-year progression-free survival (PFS) and overall survival (OS), and a Multivariate Cox PH regression model was used to estimate the hazard ratio (HR) with a 95% confidence interval (CI) and p<0.05. Results: 77 patients were included in this analysis, the median age and BMI were 70 years and 28.5 kg/m2, respectively (Table 1). Most tumors were PDL-1<1% (67.2%) and were microsatellite stable (95%). 4 patients had MSI-H. Most patients started at a lower dose (66%, median of 10 mg), and 34% of patients started at a higher dose (n=26, median of 14 mg). The most common dose was 10 mg (41%, n=32) and 13% of patients received 20 mg (n=10). In the lower dose group, 71% had ECOG 0-1 vs. 77% in the higher dose, and were more likely to have > 2 lines of treatment (45% vs. 22%). The median time to interruption was 41 vs. 51 days respectively. The PFS and OS were not statically different (Table 1). Multivariate analysis showed a significantly higher HR in the lower dose group (2.5 [CI 1.12-5.56], p=0.025). Conclusions: TheFDA-recommended dose of Lenvatinib of 20 mg is perceived to be difficult to tolerate and patients are often started on a lower dose. In this study, we observed that the time to interruption was longer in the higher dose than in the lower dose group. Though we failed to detect a significant differences in PFS/OS, when adjusted for age, we observed a significantly higher HR of progression or death in the lower dose group. Possible explanations for our observations are that patients who received a lower dose were older and frailer. Our findings call to question the common practice of starting Lenvatinib at a low dose. Further analysis is warranted to delineate the required dose that is both efficacious and safe. [Table: see text]