Exploring the causal relationship between vitiligo and psoriasis: a bidirectional Mendelian randomization analysis

Abstract

Observational studies have demonstrated an association between vitiligo and psoriasis. However, to date, the causal nature of this association remains uncertain. The objective of this study was to investigate the potential bidirectional causal relationship between vitiligo and psoriasis by employing a bidirectional two-sample Mendelian randomization (MR) approach. We utilized summary statistics obtained from the genome-wide association study (GWAS) conducted in European ancestry for vitiligo (N = 44,266) and psoriasis (N = 373,338). We first performed univariate MR analysis to detect potential bidirectional causality between vitiligo and psoriasis. Then, for directions in which univariate MR confirmed a causal relationship, we further conducted multivariate MR analysis to investigate independent causal effects on the outcome considering exposure to confounders. The bidirectional two-sample MR analysis showed genetic liability to vitiligo was significantly associated with an increased risk of psoriasis (OR = 1.094, 95% CI: 1.052, 1.138), but there was no significant association between genetic liability to psoriasis and risk of vitiligo (OR = 1.176, 95% CI: 0.915, 1.511). For the vitiligo to psoriasis direction, multivariate MR adjusting for smoking, drinking, body mass index, and rheumatoid arthritis showed the presumed causality was despite attenuated (OR = 1.060, 95% CI:1.035, 1.085), and remained statistically significant. Our study suggests that vitiligo is a causal risk factor for psoriasis, but the reverse may not be true. It is emphasized by the evidence from this study that enhanced early screening for psoriasis among patients with vitiligo may help to reduce the incidence of psoriasis.