Percutaneous hepatic perfusion (PHP) with melphalan for patients with ocular melanoma liver metastases: Preliminary results of FOCUS (PHP-OCM-301/301A) phase III trial.

Abstract

9510 Background: Ocular melanoma, the most common intraocular malignancy, frequently metastasizes to the liver but to date there is no established standard of care for hepatic-dominant ocular melanoma patients. The FOCUS trial began as a randomized, phase III trial (301) comparing PHP with best alternative care (BAC). The trial was subsequently amended (301A) to remove the BAC arm due to enrollment concerns. Methods: Eligible patients with hepatic-dominant ocular melanoma were randomized 1:1 to receive PHP or BAC (investigator’s choice of TACE, pembrolizumab, ipilimumab, or dacarbazine) on the 301 trial. All eligible patients received PHP on the 301A trial. PHP patients could receive up to 6 PHP treatments, repeated every 6-8 weeks with melphalan dosed at 3.0mg/kg ideal body weight (IBW). Patients with progressive disease (PD) were discontinued from study treatment and all patients are followed until death. Patientswere imaged every 12 (±2) weeks until PD. The primary endpoint, ORR (per RECIST 1.1) as assessed by Independent Review Committee, will be characterized by the point estimate and 95% CI for each group (PHP and BAC). Categorical efficacy variables will be presented as frequency counts and percentages and 95% CI. Time-to-event variables will be summarized using Kaplan-Meier methods (median and 95% CI). Results: 144 patients were enrolled overall; 102 were assigned to PHP (301: n=43; 301A: n=59) and 42 were assigned to BAC. 91 PHP patients received treatment (301: n=40; 301A: n=51) and 32 BAC patients received treatment. At the time of this analysis, 4 PHP patients were still ongoing on study treatment with a minimum follow-up of 24 weeks. 79 PHP-treated patients and 29 BAC-treated patients were evaluable for response. ORR among PHP patients was 32.9% (26/79; 95% CI: 22.75-40.40%). ORR among BAC patients was 13.8% (4/29; 95% CI: 3.89-31.66%). The median PFS was 9.03 months (95% CI: 6.24-11.83) among PHP patients and was 3.06 months (95% CI: 2.69-5.65) among BAC patients; this difference was statistically significant ( p=0.0004). Among the 94 patients assessed for safety after treatment with PHP, 40.4% of patients experienced a serious treatment-emergent adverse event, the majority of which were hematological and resolved without sequelae. There were no treatment related deaths in the trial. Conclusions: In this analysis of preliminary data from the FOCUS trial, PHP demonstrates a statistically superior ORR and significantly prolonged PFS in comparison with BAC in the treatment of hepatic metastases from ocular melanoma. The data is encouraging as efficacious treatments for hepatic metastases from ocular melanoma are desperately needed. These early data show an improvement over the previous phase III study in terms of both efficacy (ORR and PFS) as well as toxicity using second generation filters. Clinical trial information: NCT02678572.