Causal effects of antihypertensive drugs on thyroid diseases

Thyroid disorders frequently coexist with cardiovascular disease, including hypertension, but whether long-term modulation of antihypertensive drug targets influences thyroid disease risk remains uncertain. We aimed to evaluate the associations of 7 antihypertensive drug-target classes with thyroid diseases in European and East Asian populations using a trans-ancestry drug-target Mendelian randomization (MR) framework. We performed a two-sample, trans-ancestry drug-target MR analysis of 7 antihypertensive target classes. Target-proximal single-nucleotide polymorphisms associated with systolic blood pressure (SBP) were used as genetic proxies. Estimates were scaled per 1-mm Hg genetically predicted increase in SBP; the putative effect of target inhibition that lowers SBP was therefore interpreted in the opposite direction. Summary statistics were obtained for 4 primary thyroid diseases, benign thyroid tumors where available, and 3 thyroid-related proteins in European and East Asian datasets. Hypertension was used as a positive-control outcome. Random-effects inverse-variance weighting was the primary method, with Wald ratio estimation for single-nucleotide polymorphism instruments and weighted-median, Cochran's Q, MR-Egger, and Mendelian Randomization Pleiotropy RESidual Sum and Outlier analyses as sensitivity assessments. Within the European populace, genetic surrogates for β-blockers exhibit (odds ratio = 0.93; 95% confidence interval = 0.87-0.99) a close association with heightened susceptibility to thyroid carcinoma, whereas KCNH2 (1.15; 1.01-1.32) demonstrates an inverse correlation with thyroid cancer risk. Moreover, calcium channel blockers exhibit a conspicuous linkage with an elevated incidence of hyperthyroidism (0.97; 0.96-0.99) alongside a diminished likelihood of thyroid cancer (1.07; 1.03-1.10). CYP17A1 exerts a protective influence against hyperthyroidism (1.04; 10.1-1.07) and Graves disease (1.10; 1.05-1.16) but manifests a pathogenic effect concerning hypothyroidism (0.99; 0.97-1.00) and thyroid cancer (0.92; 0.86-0.98). Conversely, KCNJ11 demonstrates a proximate correlation between diminished hypothyroidism risk and augmented thyroid cancer susceptibility (0.83; 0.73-0.95). Analogous estimations were noted for CYP17A1 within the East Asian cohort. Genetically proxied perturbation of several antihypertensive drug targets was associated with thyroid disease risk, with CYP17A1 showing the broadest cross-ancestry pattern. These estimates reflect lifelong target-proximal genetic effects rather than the clinical effects of prescribed drugs and should be considered hypothesis-generating pending replication, pharmacoepidemiological studies, and randomized clinical trials.