Leveraging human genetics to inform the effect of blood pressure lowering drugs on adverse pregnancy outcomes

Background: There is very limited evidence on the comparative benefits and adverse effects of different antihypertensive drug classes in pregnancy. Mendelian randomization (MR) can provide insights into drug efficacy and safety in pregnancy by using genetic variants to mimic pharmacological modulation of proteins targeted by drugs.

Objectives: We used MR to explore the effect of modulating targets of antihypertensive drugs on pregnancy-related outcomes.

Methods:

Design: MR (observational)
Setting: Up to 468,950 women from five contributing studies in the UK, Norway and Finland.
Exposures: Genetically predicted perturbation of targets of several antihypertensive drugs (i.e. angiotensin-converting enzyme inhibitors (ACEi), β-blockers (BB), calcium channel blockers (CCB), and thiazide diuretic agents (TZD)).
Main outcome measures: miscarriage, stillbirth, hypertensive disorders of pregnancy (HDP), preeclampsia, preterm birth, and small-for-gestational age (SGA).
Statistical analysis: We selected genetic variants in the drug target gene region that were strongly (P < 5*10-8) and independently (R2 < 0.2) associated with systolic (SBP) or diastolic (DBP) blood pressure. We used these genetic variants to mimic pharmacological modulation of targets of antihypertensive drugs and tested their effect on pregnancy-related outcomes using the generalized inverse-variance weighted MR method. Results for each drug target are presented as odds ratio (OR) per 1 standard deviation (SD) lower SBP or DBP. We performed additional analyses to test the validity of our findings, including genetic colocalization, to explore confounding by linkage disequilibrium, and adjustments for fetal genotype, to explore bias by fetal genetic effects.

Results: Modulating targets of CCB was related to lower risk of HDP (CACNA1C, CACNA1D and CACNB2), preeclampsia (CACNB2), and preterm birth (CACNA1C) but mixed effects on SGA. Modulating a target of BB (ADRB1) was related to lower risk of HDP, preeclampsia, and miscarriages. Inhibiting ACE, the target of ACEi, was related to lower risk of HDP and SGA. For example, the OR for HDP was 0.87 for CACNA1C (95% confidence interval (CI): 0.76, 0.99), 0.84 for ADRB1 (95%CI: 0.79, 0.90), and 0.91 (95%CI: 0.84, 0.98) for ACE. Results for TZD were imprecise. In some instances, findings for SGA were attenuated after accounting for fetal genetic effects.

Conclusions: Our study supports the efficacy of inhibiting targets of antihypertensive drugs in reducing the risk of HDP (across drug classes), preeclampsia (for CCB and BB), and preterm birth (for CCB).