rs116427366 (STK32B): VLDL and Macular Degeneration

Key takeaways

• A UK Biobank study of 72,376 participants found 84 blood metabolites significantly associated with age-related macular degeneration (AMD), with lipoprotein subclasses making up 39% of those associations
• Very small VLDL (very low-density lipoprotein) subfractions were identified as likely causally involved in AMD via Mendelian randomization, a method that uses genetic variants to test cause-and-effect relationships
• Of 19 metabolites with likely causal roles in AMD, 6 were very small VLDL fractions, and one was the phospholipid-to-total-lipid ratio in medium VLDL
• Age is the dominant AMD risk factor; metabolite levels including lipoproteins contribute a smaller but measurable additional role in risk prediction

What the research says A UK Biobank analysis examining 325 blood metabolites in 72,376 participants (1,353 AMD cases and 71,023 controls) identified 84 metabolic markers significantly associated with AMD after false discovery rate (FDR) correction, with 39% of those markers being lipoprotein subclasses. Two-sample Mendelian randomization, a method that uses inherited genetic variants as instruments to estimate whether a metabolite causes a disease rather than merely being associated with it, was applied across 98,316 European UK Biobank participants and identified 19 metabolites with likely causal roles in AMD, including 6 very small VLDL fractions and the phospholipid-to-total-lipid ratio in medium VLDL. A machine learning risk prediction model built from metabolites, age, and sex confirmed age as the primary predictive factor for AMD, with metabolites contributing a much smaller additional role.

Reported associations

• Age-related macular degeneration (AMD): 84 blood metabolic markers significantly associated with AMD (FDR-adjusted p < 0.05) in 72,376 UK Biobank participants (1,353 AMD cases, 71,023 controls); lipoprotein subclasses were the most common category at 39% of associations
• Very small VLDL depletion: Depletion of circulating very small VLDL was identified as likely causal for AMD, with 6 very small VLDL fractions and the phospholipid-to-total-lipid ratio in medium VLDL implicated via two-sample Mendelian randomization in 98,316 European participants
• Lipid-related gene expression in retinal tissue: Lipid homeostasis genes identified in prior AMD GWAS show tissue-specific retinal expression: ApoE in Muller glia, LIPC in the retinal pigment epithelium, and CETP in photoreceptor outer segments and the choriocapillaris, suggesting both systemic and local retinal mechanisms contribute to AMD pathogenesis

Evidence quality The evidence base is a large UK Biobank study of 72,376 participants for AMD-metabolite associations and 98,316 European participants for metabolite GWAS. Significance was assessed at FDR-adjusted p < 0.05. Causal inference via two-sample Mendelian randomization is methodologically stronger than simple observational association, but it depends on the validity of the genetic instruments used. The provided study excerpt covers the AMD-metabolite relationship broadly; variant-specific data (odds ratios, beta coefficients, or p-values) for rs116427366 at the STK32B locus are not detailed in the available text. Replication of these causal metabolite findings in independent cohorts is not described in the provided excerpt. The evidence for very small VLDL as a likely causal factor in AMD should be considered preliminary pending further replication.

Lifestyle considerations No lifestyle considerations on file for this variant.