Pembrolizumab with or without bevacizumab in platinum-resistant recurrent or metastatic nasopharyngeal carcinoma: a randomised, open-label, phase 2 trial

Abstract

Background

Vascular endothelial growth factor (VEGF) is overexpressed in nasopharyngeal carcinoma and suppresses the anti-tumour immune response. Previous studies have shown that adding anti-VEGF treatment to PD-1 inhibition treatment strategies improves tumour response. We aimed to compare the efficacy of pembrolizumab, a PD-1 inhibitor, with or without bevacizumab, a VEGF inhibitor, in nasopharyngeal carcinoma.

Methods

In this randomised, open-label, phase 2 trial done at two hospitals (National University Cancer Institute and Tan Tock Seng Hospital) in Singapore, patients with platinum-resistant recurrent or metastatic nasophayngeal carcinoma were eligible if they were aged 21 years or older and had an Eastern Cooperative Oncology Group (ECOG) performance status of 0–1. Patients were assigned (1:1; using random permuted blocks with varying sizes of 4 and 6) to receive either intravenous pembrolizumab (200 mg) every 21 days or a combination of pembrolizumab with intravenous bevacizumab (7·5 mg/kg) administered 1 week prior to each dose, until radiographic disease progression, unacceptable toxicity, completion of 32 cycles, or withdrawal of consent. The study was open label, therefore no masking of treatment assignment was implemented. The primary endpoint was objective response rate, assessed using RECIST (version 1.1) by independent radiologists and analysed in the intention-to-treat population (ie, all randomly assigned patients). This trial is registered with ClinicalTrials.gov, NCT03813394, and enrolment has closed.

Findings

Between May 13, 2019, and Dec 6, 2023, we assessed 60 individuals for eligibility, 12 were excluded, and 48 were randomly allocated to pembrolizumab alone (n=24) or a combination of bevacizumab and pembrolizumab (n=24). The median age was 56 years (IQR 48–65), and 40 (83%) of 48 patients were male and eight (17%) were female. The median follow-up was 28·3 months (IQR 15·1–55·9). The objective response rate was significantly higher in the bevacizumab and pembrolizumab group (58·3% [95% CI 36·6–77·9] than in the pembrolizumab group (12·5% [2·7–32·4]; unadjusted RR 4·67 [95% CI 1·54–14·18]; p=0·0010). Grade 3 treatment-related adverse events occurred in two (8%) of 24 patients in the pembrolizumab group and in seven (29%) of 24 patients in the bevacizumab and pembrolizumab group; the most common severe or grade 3–4 treatment-related adverse events were thrombosis or bleeding (four [17%] of 24 patients in the bevacizumab and pembrolizumab group vs none of 24 patients in the pembrolizumab group), and others were transaminitis (none vs 1 [4%]), colitis (1 [4%] vs none]), cytopenias (none vs 1 [4%]), dermatological toxicities (1 [4%] vs none]), hypertension (1 [4%] vs none]), and proteinuria (1 [4%] vs none]). There were no grade 4 treatment-related adverse events or treatment-related deaths in either group.

Interpretation

Pembrolizumab in combination with bevacizumab was more efficacious than pembrolizumab monotherapy, with manageable toxicities in platinum-resistant nasopharyngeal carcinoma. If validated in a phase 3 trial, the combination therapy could be a new standard of care in this population of patients.

Funding

National Medical Research Council of Singapore, National Research Foundation Singapore, Singapore Ministry of Education under its Research Centres of Excellence initiatives, and Merck Sharp & Dohme.