A phase II study of lenvatinib plus pembrolizumab in patients with recurrent/metastatic salivary gland cancers.

6103 Background: Recurrent/metastatic (R/M) salivary gland cancers (SGCs) are rare diseases without standard therapies. Based on the hypothesis that VEGFR inhibition can enhance immune checkpoint-induced responses against SGCs, we conducted a phase II trial of the multikinase inhibitor lenvatinib (len) plus the programmed death-1 (PD-1) inhibitor pembrolizumab (pem) in two R/M SGC cohorts: adenoid cystic carcinoma (ACC) and non-ACC histologies. Here we report results from the completed non-ACC cohort. Methods: Patients (Pts) with R/M SGC (except ACC) were enrolled. RECIST v1.1 measurable disease was required; prior therapies were allowed. Pts with acinic cell carcinoma (AcCC) were required to have progression of disease (PD) or worsening disease-related symptoms. Len 20 mg oral daily and pem 200 mg intravenously every 3 weeks was given. The primary endpoint was best overall response (BOR) rate using a minimax Simon two-stage design. In the first stage, >1 confirmed complete and/or partial responses (CRs, PRs) was required among 18 pts to enroll 14 more pts. >4 responses among 32 pts would be considered positive (BOR 5% vs 20%, 1-sided alpha 0.1, power 0.9). Secondary objectives were progression-free survival (PFS) and safety/tolerability per CTCAE v5.0. Results: 27 pts with R/M SGC pts were enrolled; 26 evaluable for the study endpoints. Among evaluable pts, the median age was 62 and 15 were men. SGC histologies included 9 AcCC, 8 salivary duct carcinoma (SDC), 4 myoepithelial carcinoma, 2 mucoepidermoid carcinoma, and 1 each of polymorphous adenocarcinoma, epithelial-myoepithelial carcinoma, and mucinous adenocarcinoma. 5/26 (19.2%) had confirmed PR. 18 pts had stable disease (SD), 1 PD as best response; 2 evaluable pts did not reach first scan assessment. As of 1/20/25, median PFS was 47 weeks. Among the 9 AcCC pts, 4 (44.4%) had PR and 5 SD; 8/9 pts had tumor regression in target lesions. For the 8 SDC pts, 1/8 (12.5%) had PR. Per protocol, 2 pts (AcCC and SDC) stopped treatment after 2 years; both resumed treatment when PD occurred, achieving SD and PR, respectively. Six deaths were observed, 4 possibly related to treatment: 3 SDC (respiratory failure due to pneumonitis vs. cancer progression [1]; cardiac arrest after polymyositis/myocarditis/aspiration pneumonia [1]; stroke [1]) and 1 myoepithelial carcinoma (respiratory failure due to pneumonitis vs. infection [1]). Conclusions: Trial enrollment was completed after 26 evaluable pts given the study was positive for the primary BOR endpoint and the grade 5 events observed. Len+pembro may be active and safe among pts with AcCC, though further study is needed. This combination may be less promising for SDC given the grade 5 events and low response rate observed. (Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA provided lenvatinib and pembrolizumab for the study.) Clinical trial information: NCT04209660 .