Percutaneous hepatic perfusion combined with ipilimumab and nivolumab for metastatic uveal melanoma (CHOPIN): a single-centre, open-label, randomised, phase 2 trial.

In this investigator-initiated, single-centre, open-label, randomised, phase 2 trial, adults aged 18-80 years with unresectable liver-only or liver-dominant metastatic uveal melanoma, WHO performance status 0-1, and with no previous systemic therapy were randomly assigned (1:1) to receive percutaneous hepatic perfusion alone (perfusion group) or percutaneous hepatic perfusion combined with ipilimumab plus nivolumab (combination group). Two perfusions with melphalan (3 mg/kg; maximum 220 mg) were scheduled in weeks 1 and 7. The combination group also received intravenous ipilimumab (1 mg/kg) and nivolumab (3 mg/kg) once every 3 weeks in weeks 0, 3, 6, and 9 without maintenance therapy. The primary endpoint, 1-year progression-free survival, was analysed in the intention-to-treat population; the safety analysis included all treated patients and was prospectively collected and graded using Common Terminology Criteria for Adverse Events version 5.0. The trial is registered with ClinicalTrials.gov (NCT04283890) and EudraCT (2018-004248-49) and is ongoing but no longer enrolling.

Percutaneous hepatic perfusion can lead to meaningful hepatic tumour control in metastatic uveal melanoma, but its benefit is confined to liver metastases and does not address extrahepatic disease. Immune checkpoint inhibitors ipilimumab and nivolumab show limited activity in uveal melanoma, although retrospective studies suggest improved outcomes when combined with liver-directed therapies. This study aimed to evaluate the efficacy and safety of combining percutaneous hepatic perfusion with ipilimumab and nivolumab.

Between Dec 10, 2020, and Nov 15, 2024, 80 patients were screened, of whom 76 were eligible and assigned to the combination group (n=38) or perfusion group (n=38), including 49 (64%) men and 27 (36%) women. Median follow-up was 24·9 months (IQR 15·4-36·0). 1-year progression-free survival was 54·7% (95% CI 36·8-69·5) with combination therapy versus 15·8% (5·8-30·1) with perfusion alone (adjusted hazard ratio 0·34 [95% CI 0·19-0·60]; p=0·0002). Grade 3-4 treatment-related adverse events occurred in 31 (82%) of 38 patients with combination therapy compared to 15 (41%) of 37 patients with perfusion alone, due to both a higher incidence of severe perfusion-related events and immunotherapy-related adverse events. The most common grade 3-4 adverse events were thrombocytopenia (13 [34%] in the combination group vs five [14%] in the perfusion group), leukopenia (ten [26%] vs five [14%]), γ-glutamyl transferase increase (seven [18%] vs three [8%]), and anaemia (five [13%] vs one [3%]). One treatment-related death (due to triple M syndrome) occurred in the combination group.

Adding ipilimumab and nivolumab to percutaneous hepatic perfusion significantly improved progression-free survival, but with a higher rate of adverse events. The combination therapy offers a promising new treatment paradigm for patients with metastatic uveal melanoma. These results would ideally be validated in larger, multicentre randomised trials; however, conducting such studies is challenging due to the low incidence of uveal melanoma.

Leiden University Medical Centre, Delcath Systems, and Bristol Myers Squibb.