Vitamin D and the risk of treatment-resistant and atypical depression: A Mendelian randomization study

Ryan Arathimos,Fiona Gaughran,Matthew Hotopf,Amy Ronaldson,Laurence J Howe,Alexandru Dregan,Chiara Fabbri,Saskia Hagenaars,Cathryn M Lewis

doi:10.1038/s41398-021-01674-3

Abstract

Observational evidence has implicated vitamin D levels as a risk factor in major depressive disorder (MDD). Confounding or reverse causation may be driving these observed associations, with studies using genetics indicating little evidence of an effect. However, genetic studies have relied on broad definitions of depression. The genetic architecture of different depression subtypes may vary since MDD is a highly heterogenous condition, implying potentially diverging requirements in therapeutic approaches. We explored the associations between vitamin D and two subtypes of MDD, for which evidence of a causal link could have the greatest clinical benefits: treatment-resistant depression (TRD) and atypical depression (AD). We used a dual approach, combining observational data with genetic evidence from polygenic risk scores (PRS) and two-sample Mendelian randomization (MR), in the UK Biobank. There was some evidence of a weak association between vitamin D and both incident TRD (Ncases = 830) and AD (Ncases = 2366) in observational analyses, which largely attenuated when adjusting for confounders. Genetic evidence from PRS and two-sample MR, did not support a causal link between vitamin D and either TRD (Ncases = 1891, OR = 1.01 [95%CI 0.78, 1.31]) or AD (Ncases = 2101, OR = 1.04 [95%CI 0.80, 1.36]). Our comprehensive investigations indicated some evidence of an association between vitamin D and TRD/AD observationally, but little evidence of association when using PRS and MR, mirroring findings of genetic studies of vitamin D on broad depression phenotypes. Results do not support further clinical trials of vitamin D in these MDD subtypes but do not rule out that small effects may exist that require larger samples to detect.

Highlights

Major depressive disorder (MDD) is the leading cause of disability globally affecting 4.7% of the population [1]
Two-sample Mendelian randomization analyses provided little evidence of an effect of vitamin D on prevalent treatment-resistant depression (TRD) when using instrumental variable (IV) set A (27 SNPs, IVW odds ratios (OR) 1.13 [95% CI 0.87, 1.45]) or IV set B (110 SNPs, IVW OR 1.01 [95%CI 0.78, 1.31]) (Fig. 1 and Supplementary Tables 7, 8)
We found no evidence for heterogeneity amongst IVs in either IV set when using the IVW or Mendelian randomization (MR)

Summary

Introduction

Major depressive disorder (MDD) is the leading cause of disability globally affecting 4.7% of the population [1]. Vitamin D deficiency is a major global health problem [2] and in the UK rates of both vitamin D deficiency and insufficiency are considerable [3]. Vitamin D is considered essential for bone health and immune function [4], but is thought to affect the expression of certain neurotransmitters. Several meta-analyses of longitudinal studies have shown that vitamin D status is prospectively associated with depression in older adults [6,7,8]. Using a sample three times larger than the world literature (n = 139,128), a recent study has shown that both vitamin D deficiency and insufficiency are associated with new onset depression in middle-aged adults [9]

Methods

Discussion

Conclusion