Background
Observational studies and previous Mendelian randomization (MR) studies have shown that genetically low 25-hydroxyvitamin D (25OHD) levels are associated with a high susceptibility to multiple sclerosis (MS). The present MR study aims to update the causal estimates for the effects of 25OHD levels on MS risk.
Methods
To date, the largest genome-wide association study (GWAS) for serum 25OHD (n = 401,460) and MS (14,498 MS cases and 24,091 controls) was used to assess the effect of serum 25OHD levels on MS. All participants were of European ancestry. The MR-egger_intercept test and Cochran’s Q statistic were used to determine the pleiotropy and the heterogeneity, respectively. MR-egger, weighted median, inverse variance weighted (multiplicative random effects), simple mode, and weighted mode methods were used to evaluate the causal association of serum 25OHD levels with MS. Finally, the effect of a single 25OHD SNP (single nucleotide polymorphism) on MS was used to test the SNP bias.
Results
One hundred and fifteen newly identified serum 25OHD genetic variants were extracted from a large-scale serum 25OHD GWAS dataset. The 20 most effective and independent 25OHD genetic instrumental variables were extracted from the MS GWAS summary statistics. Pleiotropy analysis suggested no significant pleiotropic variant among the 20 selected 25OHD genetic instrument variants in MS GWAS datasets. As serum levels of 25OHD based on genetic changes increased, the risk of MS decreased using MR-egger (Beta = − 0.940, p = 0.001; OR = 0.391), weighted median (Beta = − 0.835, p = 0.000; OR = 0.434), IVW (Beta = − 0.781, p = 0.000; OR = 0.458), simple mode (Beta = − 1.484, p = 0.016; OR = 0.227), and weighted mode (Beta = − 0.913, p = 0.000; OR = 0.401). Our results were robust, with no obvious bias based on investigating the single 25OHD SNP on MS.
Conclusions
Our analysis suggested a causal association between genetically increased serum 25OHD levels and reduced MS in the European population.
