Background: Up to 30% of UM patients will develop metastases, with a median survival of 12 months in the metastating setting. Prognostic factors combine clinical features of the primary tumor (diameter, thickness, retinal detachment, extra-scleral extension) and genetic factors (monosomy 3, 8 q gain and class 1 /2 gene expression profiling).The genomic analysis is feasible by fine needle aspiration biopsies before radiotherapy for small UM or on enucleated eyes.

Methods: Multicenter randomized phase 3 trial with adjuvant fotemustine, 6 cycles, 100 mg/m2 versus surveillance for 3 years (liver tests/3 months, liver MRI or CT/6 months, whole body CT/12 months) in patients with high risk of recurrence, defined by clinical criteria (diameter > 15 mm with extra scleral extension and/or retinal detachment or diameter > 18 mm) or genomic high risk signature by array-CGH (monosomy 3 or deletion of 3p associated with gain of chromosome 8). The primary objective was 5-year Metastasis Free Survival (MFS). With an expected increase of 5-year MFS from 50 to 70%, 302 patients and 99 events were required to achieve an 95%-power with a 5% type I error rate. Secondary objectives were overall survival (OS), safety (NCI-CTC v3), quality of life (QLQ-C30). Interim analyses were planned for safety and after 50 events, disclosed to an independent safety monitoring board.

Results: The trial was stopped for futility after 244 patients had been recruited between June 2009 and January 2016. No unexpected toxicity was found in the chemotherapy group. The study was amended to go on with intensive surveillance in new high risk patients. Ninety-one metastases and 43 deaths were reported, with no treatment-related death. With a median follow-up of 3 years, the 3-year MFS is 60.3% in the chemo group and 60.7% in the surveillance group (HR 0.97 [0.64-1.47]). The 3-year OS is 79.4% [73.2-85.7], with no difference between the 2 groups of patients.

Conclusions: FOTEADJ is the first adjuvant randomized phase 3 trial based on genomic analysis in high risk UM patients. Despite negative results, it shows the feasibility of multicenter adjuvant studies in this rare cancer and provides genomic data in small tumors for future trials.

Clinical trial information: NCT02843386.