Abstract
BACKGROUND: Observational research suggests that vitamin D levels affect psoriasis. However, observational studies are prone to potential confounding or reverse causation, which complicates interpreting the data and drawing causal conclusions. AIM: To apply Mendelian randomization (MR) methods to comprehensively assess a potential association between vitamin D and psoriasis. METHODS: Genetic variants strongly associated with 25-hydroxyvitamin D (25OHD) in genome-wide association study (GWAS) data from 417 580 and 79 366 individuals from two independent studies served as instrumental variables (used as the discovery and replication datasets, respectively). As the outcome variable, we used GWAS data of psoriasis (13 229 people in the case group, 21 543 in the control group). We used (i) biologically validated genetic instruments, and (ii) polygenic genetic instruments to assess the relationship between genetically proxied vitamin D and psoriasis. We carried out inverse-variance weighted (IVW) MR analyses for the primary analysis. In sensitivity analyses, we used robust MR approaches. RESULTS: MR analyses of both the discovery and replication datasets did not show an effect of 25OHD on psoriasis. Neither the IVW MR analysis of the biologically validated instruments [discovery dataset: odds ratio (OR) 0.99; 95% confidence interval (CI) 0.88-1.12, P = 0.873; replication dataset: OR 0.98, 95% CI 0.66-1.46, P = 0.930] nor that of the polygenic genetic instruments (discovery dataset: OR 1.00, 95% CI 0.81-1.22, P = 0.973; replication dataset: OR 0.94, 95% CI 0.64-1.38, P = 0.737) revealed an impact of 25OHD on psoriasis. CONCLUSION: The present MR study did not support the hypothesis that vitamin D levels, measured by 25OHD, affect psoriasis. This study was conducted on Europeans, so the conclusions may not be applicable to all ethnicities.