Pembrolizumab for vulvar squamous cell carcinoma: results from the phase 2 KEYNOTE-158 study

Abstract

<h3>Objectives:</h3> Advanced vulvar carcinoma typically affects the elderly and treatment options are limited. Moderate antitumor activity of pembrolizumab (ORR, 6%) was noted in a small population of patients (pts) with PD-L1-positive vulvar cancer in the phase 1b KEYNOTE-028 study. KEYNOTE-158 (ClinicalTrials.gov, NCT02628067) is a larger, nonrandomized, multicohort, open-label, phase 2 study that evaluates pembrolizumab in pts with previously treated advanced cancers irrespective of tumor PD-L1 expression. We report results for pts in the vulvar carcinoma cohort receiving pembrolizumab monotherapy. <h3>Methods:</h3> Enrolled pts were aged ≥18 y with histologically/cytologically documented advanced vulvar squamous cell carcinoma with prior treatment failure; measurable disease per RECIST v1.1; ECOG performance status ≤1; and tumor samples available for biomarker analysis, including PD-L1. PD-L1 positivity was defined as combined positive score ≥1 with expression evaluated using the PD-L1 IHC 22C3 pharmDx assay (Agilent Technologies, Carpinteria, CA, USA). Pts received pembrolizumab 200 mg Q3W until disease progression, unacceptable toxicity, or completion of 35 treatment cycles. Tumor imaging was performed every 9 weeks for 1 y and every 12 weeks thereafter. Response was assessed per RECIST v1.1 by independent central radiologic review. The primary endpoint was ORR. Secondary endpoints included duration of response (DOR), PFS, OS, and safety. <h3>Results:</h3> A total of 101 pts were enrolled. As of October 5, 2020, median time from start of treatment to data cutoff was 36.0 (range, 15.4–55.2) months. Median age was 64.0 (range, 31–87) y; 9 (8.9%) pts had no prior therapy, 57 (56.4%) had 1, 23 (22.8%) had 2, and 11 (10.9%) had 3 or more lines of prior therapy. Tumors were PD-L1–positive in 84 pts (83.2%), PD-L1–negative in 7 pts (6.9%), and non-evaluable/not assessed in 10 pts (9.9%). Overall, ORR was 10.9% (95% CI, 5.6–18.7); 1 pt had a CR and 10 pts had PR. Among pts with PD-L1–positive tumors, 8 of 84 (9.5%; 95% CI, 4.2–17.9) had CR or PR; among pts with PD-L1–negative tumors, 2 of 7 (28.6%; 95% CI, 3.7–71.0) had CR or PR. Among pts with a response, median DOR was 20.4 months (range, 2.1+ to 28.0). Among all enrolled pts (n=101), median PFS was 2.1 months (95% CI, 2.0–2.1) and median OS was 6.2 months (95% CI, 4.9–9.4 months). Treatment-related AEs occurred in 51 pts (50.5%), including 12 pts (11.9%) who had grade 3–5 events (no event occurred in >1 pt). There were 2 deaths (2.0%) considered related to treatment (hepatitis, n=1; fulminant hepatitis, n=1). Neither pt had comorbidities or pre-existing liver disease to contribute to these events; no other contributing factors were identified. A total of 5 pts (5.0%) discontinued because of treatment-related AEs. Immune-mediated AEs occurred in 18 pts (17.8%). <h3>Conclusions:</h3> Pembrolizumab monotherapy was associated with durable responses in a subset of pts with vulvar carcinoma (ORR, 10.9%; median DOR, 20.4 months). Responses occurred in both PD-L1-positive and PD-L1-negative pts.