OBJECTIVE: The MAPK pathway is constitutively activated in uveal melanoma (UM). Selumetinib (AZD6244, ARRY-142886), a MEK inhibitor, has shown limited activity as monotherapy in metastatic UM. Pre-clinical studies support synergistic cytotoxic activity for MEK inhibitors combined with taxanes, and here we sought to assess the clinical efficacy of combining selumetinib and paclitaxel.
METHODS: Seventy-seven patients with metastatic UM who had not received prior chemotherapy were randomised to selumetinib alone, or combined with paclitaxel with or without interruption in selumetinib two days before paclitaxel. The primary endpoint was progression free survival (PFS). After amendment, the combination arms were combined for analysis and the sample size adjusted to detect a hazard ratio (HR): 0.55, 80% power at 1-sided 5% significance level.
RESULTS: The median PFS in the combination arms was 4.8 months (95% CI: 3.8 - 5.6) compared with 3.4 months (2.0 - 3.9) in the selumetinib arm (HR 0.62 [90% CI 0.41 - 0.92], 1-sided p-value = 0.022). ORR was 14% and 4% in the combination and monotherapy arms respectively. Median OS was 9 months for the combination and was not significantly different from selumetinib alone (10 months) with HR of 0.98 [90% CI 0.58 - 1.66], 1-sided p-value = 0.469. Toxicity was in keeping with the known profiles of the agents involved.
CONCLUSIONS: SelPac met its primary endpoint, demonstrating an improvement in PFS for combination selumetinib and paclitaxel. No improvement in OS was observed, and the modest improvement in PFS is not practice changing.

Uveal melanoma (UM) is an orphan disease with a mortality of 80% within one year upon the development of metastatic disease. UM does hardly respond to chemotherapy and kinase inhibitors and is largely resistant to checkpoint inhibition. Hence, further therapy approaches are urgently needed. To improve clinical outcome, we designed a trial employing the 3rd generation personalized IKKβ-matured RNA-transfected dendritic cell (DC) vaccine which primes T cells and in addition activates NK cells. This ongoing phase I trial [NCT04335890 (www.clinicaltrials.gov), Eudract: 2018-004390-28 (www.clinicaltrialsregister.eu)] investigates patients with treatment-naive metastatic UM. Monocytes are isolated by leukapheresis, differentiated to immature DCs, matured with a cytokine cocktail, and activated via the NF-κB pathway by electroporation with RNA encoding a constitutively active mutant of IKKβ. Three types of antigen-RNA are co-electroporated: i) amplified mRNA of the tumor representing the whole transcriptome, ii) RNA encoding driver mutations identified by exome sequencing, and iii) overexpressed non-mutated tumor antigens detected by transcriptome sequencing. This highly personalized DC vaccine is applied by 9 intravenous infusions in a staggered schedule over one year. Parallel to the vaccination, standard therapy, usually an immune checkpoint blockade (ICB) as mono (anti-PD-1) or combined (anti-CTLA4 and anti-PD-1) regimen is initiated. The coordinated vaccine-induced immune response encompassing tumor-specific T cells and innate NK cells should synergize with ICB, perhaps resulting in measurable clinical responses in this resistant tumor entity. Primary outcome measures of this trial are safety, tolerability and toxicity; secondary outcome measures comprise overall survival and induction of antigen-specific T cells.