Belzutifan for patients with von Hippel-Lindau disease-associated CNS haemangioblastomas (LITESPARK-004): a multicentre, single-arm, phase 2 study.

backgroundThe first-in-class hypoxia-inducible factor-2α inhibitor, belzutifan, showed clinically meaningful antitumour activity in von Hippel-Lindau (VHL) disease-associated neoplasms in the ongoing, single-arm, phase 2 LITESPARK-004 study. We aimed to investigate antitumour activity with an additional 16 months of follow-up and present updated results for the subgroup of patients with CNS haemangioblastomas.

methodsIn the multicentre, single-arm, phase 2 LITESPARK-004 study, adults (aged ≥18 years) from 11 cancer centres or hospitals in the USA, Denmark, France, and the UK, with germline VHL alterations, at least one measurable renal cell carcinoma tumour, no renal cell carcinoma tumour greater than 3 cm requiring immediate surgical intervention, an Eastern Cooperative Oncology Group performance status 0 or 1, and no previous systemic therapy received oral belzutifan 120 mg once daily until unacceptable toxicity, disease progression, or patient decision to withdraw. The primary endpoint, evaluated in patients with CNS haemangioblastomas, was the proportion of patients with an objective response per RECIST version 1.1 by an independent review committee. We assessed response using two approaches. In approach 1, we evaluated all measurable (≥1 cm maximum diameter) or non-measurable lesions at baseline, including both the solid lesion and the associated cystic component if present. In approach 2, we evaluated only baseline lesions with a measurable (≥1 cm maximum diameter) solid lesion. Antitumour activity was assessed in all patients who received at least one dose of belzutifan. This study is no longer recruiting but is ongoing, and is registered with Clinicaltrials.gov, NCT03401788.

findingsBetween May 31, 2018, and March 29, 2019, of 67 patients screened, 61 (32 [52%] male and 29 [48%] female) were enrolled; 50 (82%) had at least one CNS haemangioblastoma evaluable at baseline (184 total lesions). Median follow-up for the 50 patients with CNS haemangioblastomas was 38·0 months (IQR 36·7-40·1). In approach 1, 22 of 50 patients (44% [95% CI 30-59]) had an objective response. In approach 2, 19 of 25 patients (76% [55-91] had an objective response. 23 (46%) of 50 patients had a grade 3-5 all-cause adverse event. 19 (38%) patients reported grade 3 adverse events, the most common of which was anaemia (in 6 [12%] patients). Two of 50 patients (4%) reported grade 4 events (retinal vein occlusion and embolism). Two patients died owing to adverse events not considered treatment-related (suicide and toxicity to various agents).

interpretationBelzutifan showed meaningful antitumour activity in VHL disease-associated CNS haemangioblastomas that was sustained for more than 3 years of treatment. These results continue to support belzutifan as a systemic treatment option for patients with VHL disease-related CNS haemangioblastomas.

fundingMerck Sharp & Dohme, National Institutes of Health, and National Cancer Institute.