Poster Presentation GENEMAPPERS 2026

Genetic dissection of polygenic risk scores for cardiovascular disease following hypertensive disorders of pregnancy (#112)

Gayathry Krishnamurthy 1 , James A Temple 1 , Mustafa A Ahmed 1 , Jennie Hui 2 3 , Hoang T Phan 1 4 , Shaun P Brennecke 5 6 , Eric K Moses 1 , Phillip E Melton 1 3 7
  1. Menzies Institute for Medical Research, College of Health and Medicine, The University of Tasmania, Hobart, Tasmania, Australia
  2. Busselton Population Medical Research Institute, Busselton, Western Australia, Australia
  3. School of Population and Global Health, The University of Western Australia, Crawley, Western Australia, Australia
  4. Menzies School of Health Research, Centre for Child Development, The Charles Darwin University, Darwin, Northern Territory, Australia
  5. Pregnancy Research Centre, Department of Maternal‐Fetal Medicine, The Royal Women’s Hospital, Melbourne, Victoria, Australia
  6. Department of Obstetrics and Gynaecology, The Royal Women’s Hospital, The University of Melbourne, Melbourne, Victoria, Australia
  7. Curtin School of Allied Health, Faculty of Health Sciences, Curtin University, Perth, Western Australia, Australia

Cardiovascular disease (CVD) is a leading cause of mortality and morbidity for women. Women with hypertensive disorders of pregnancy (HDP) have an increased risk of CVD and whether this reflects shared genetic susceptibility across these two traits is unclear. This study conducted a comprehensive polygenic risk score (PRS) screening to determine if women with a history of HDP have an increased polygenic risk of CVD.

We analysed 1,261 CVD related PRS from 1,582 female participants from the Busselton Health Study. CVD cases were determined from ICD CVD outcomes obtained from WA Department of Health-linked data. This included a total of 944 CVD cases, 638 controls, including 373 women with a previous self-reported history of HDP (221 had CVD).  To determine association of PRS for CVD following HDP, we utilised logistic and linear regression on three outcomes including CVD status (yes/no), age at first CVD event, and total number of hospitalisations, adjusting models for age, anthropometry, blood chemistry, blood pressure, smoking and 10 principal components. Multiple testing was adjusted using FDR.

None of the 1261 PRS remained significant following FDR correction for multiple testing. However, a small subset of PRS were nominally significant (unadjusted p-value <0.05). These included those for coronary artery disease (PGS000748, p-value=0.002) for CVD status, total cholesterol (PGS000192, p-value=0.0003) for age at first CVD event, and BMI (PGS004902, p-value=0.001) for number of hospitalisations. We further investigated the impact of HDP on these models and identified two blood pressure PRS, for diastolic blood pressure (PGS004232, p-value=0.04) and systolic blood pressure (PGS005015, p-value=0.04))

The lack of statistically significant PRS suggests that any shared genetic contribution is likely modest and requires larger sample sizes or more refined PRS to detect CVD following HDP. Future work should also explore integrated models combining genetic, with other omic layers to improve PRS prediction.