Lipids, lipid-modifying drug target genes and migraine: a Mendelian randomization study

Migraine, a prevalent headache disorder with unclear mechanisms and limited treatments, may be influenced by dyslipidemia and genetic factors. Statins and emerging lipid-modifying agents show potential but lack evidence for migraine management. Mendelian Randomization analysis offers insights into causal relationships and therapeutic targets. This study aims to explore genetically predicted lipid traits, drug targets, and their association with migraine risk.

We conducted Mendelian randomization (MR) analyses utilizing genetic variants associated with lipid traits and variants in genes encoding the protein targets of various classes of lipid-lowering drugs. The specific drug classes investigated included HMGCR, PCSK9, NPC1L1, ABCG5/ABCG8, LDLR, LPL, ANGPTL3, APOB, CETP, and APOC3. To determine the effects on migraine risk, we meta-analyzed MR estimates for regional variants using data from two large sample sets. The genetic variants were weighted based on their associations with specific lipid traits, such as low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), Apolipoprotein A1, and Apolipoprotein B. To obtain association weights, we utilized data from lipid genetics consortia. For lipid-modifying drug targets that exhibited suggestive significance, we further employed expression quantitative trait locus (eQTL) data. Additionally, we performed colocalization analysis to assess genetic confounding.

The use of genetic proxies for HMGCR inhibition demonstrated a significant association with a decreased risk of migraine in the FinnGen dataset (OR = 0.64, 95% CI: 0.46-0.88, p = 0.0006) and a nearly significant association in the Choquet dataset (OR = 0.78, 95% CI: 0.60-1.01, p = 0.06). When pooling the estimates, the overall effect size showed a reduced risk of migraine (OR = 0.73, 95% CI: 0.60-0.89, p = 0.0016). Similarly, genetic mimicry of LPL enhancement was associated with a lower risk of migraine in the FinnGen dataset (OR = 0.82, 95% CI: 0.69-0.96, p = 0.01) and the Choquet dataset (OR = 0.91, 95% CI: 0.83-0.99, p = 0.03). Pooling the estimates showed a consistent effect size (OR = 0.89, 95% CI: 0.83-0.96, p = 0.002). Sensitivity analyses yielded no statistically significant evidence of bias arising from pleiotropy or genetic confounding.

In the study, it was observed that among the 10 lipid-lowering drug targets investigated, LPL and HMGCR showed significant associations with migraine risk. These findings indicate that LPL and HMGCR have the potential to serve as candidate drug targets for the treatment or prevention of migraines.