High-dose vitamin D supplementation does not improve outcome in a cutaneous melanoma population: results of a randomized double-blind placebo-controlled study (ViDMe trial).

Abstract

Background: Observational studies in cutaneous melanoma (CM) have indicated an inverse relationship between levels of 25-hydroxyvitamin D and Breslow thickness, in addition to a protective effect of high 25-hydroxyvitamin D levels on clinical outcome.

Objectives: To evaluate whether high-dose vitamin D supplementation in curatively resected CM reduces melanoma relapse.

Methods: In a prospective randomized double-blind placebo-controlled trial, 436 patients with resected CM stage IA to III (8th American Joint Committee on Cancer staging) were randomized. Among them, 218 received a placebo while 218 received monthly 100 000 IU cholecalciferol for a minimum of 6 months and a maximum of 42 months (treatment arm). Following randomization, patients were followed for a median of 52 months, with a maximum follow-up of 116 months. The primary endpoint was relapse-free survival. Secondary endpoints were melanoma-related mortality, overall survival, and the evolution of 25-hydroxyvitamin D serum levels over time.

Results: In our population (mean age 55 years, 54% female sex) vitamin D supplementation increased 25-hydroxyvitamin D serum levels after 6 months of supplementation in the treatment arm by a median 17 ng mL-1 [95% confidence interval (CI) 9-26] compared with 0 ng mL-1 (95% CI 6-8) in the placebo arm (P < 0.001, Wilcoxon test) and remained at a steady state during the whole treatment period. The estimated event rate for relapse-free survival at 72 months after inclusion was 26.51% in the vitamin D supplemented arm (95% CI 19.37-35.64) vs. 20.70% (95% CI 14.26-29.52) in the placebo arm (hazard ratio 1.27, 95% CI 0.79-2.03; P = 0.32). After adjusting for confounding factors (including baseline stage, body mass index, age, sex and baseline season), the hazard ratio was 1.20 (95% CI 0.74-1.94, P = 0.46). The number of deaths from progression of CM and nonmelanoma-related deaths was similar in both the vitamin D supplemented and placebo groups (deaths from progression of CM, n = 10 and n = 11, respectively; nonmelanoma-related deaths, n = 3 and n = 2, respectively). No major adverse events were observed during the study.

Conclusions: In patients with CM, monthly high-dose vitamin D supplementation was safe, resulted in a sustained increase in 25-hydroxyvitamin D levels during the treatment period, but did not improve relapse-free survival, melanoma-related death or overall survival.