Polygenic prediction of coronary heart disease among 130,000 Mexican adults

Importance

Coronary heart disease (CHD) is a leading cause of premature mortality globally. Most polygenic risk scores (PRSs) for CHD have been derived in populations of European ancestry. Their utility for CHD risk prediction in other populations is uncertain.

Objective

To evaluate the performance of eight established CHD PRSs in an admixed cohort of Mexican adults.

Design, Setting, Participants

133,207 genotyped participants aged 35–79 years from the Mexico City Prospective Study (MCPS), a cohort recruited between 1998–2004, with follow-up for mortality until September 30, 2022.

Exposures

Eight PRSs for CHD, comprising between 44 and 6,472,620 single nucleotide polymorphism (SNP) variants, were selected and recreated for MCPS participants.

Main outcomes and measure

Premature CHD comprised prior doctor-diagnosed CHD at recruitment or CHD-related death before age 80. Logistic regression adjusted for age, sex, and the first seven genetic principal components (PCs) assessed PRS associations with CHD. Additional analyses evaluated performance by key participant characteristics, and after adjustment for vascular risk factors. Risk discrimination was assessed using C-statistics.

Results

Of the participants, 67% were women, the mean (±SD) age was 51±12 years, and Indigenous American ancestry averaged 67%. Premature CHD occurred in 5,163 participants (3.9%), including 1,901 prevalent and 3,479 fatal cases. All eight PRSs were positively and log-linearly associated with CHD, with odds ratios (ORs) per 1 SD increase ranging from 1.05 (95% CI, 1.03–1.08) to 1.29 (95% CI, 1.25–1.33). Associations were consistent across strata of age, ancestry, and relatedness. For six PRSs, however, associations were stronger in men than women (e.g., for the PRS with the strongest overall association: OR 1.37 [1.32–1.43] in men vs. 1.23 [1.18–1.28] in women). Adjustment for vascular risk factors did not substantially alter associations. Models including age, sex, genetic PCs and a PRS achieved an AUC of 0.72.

Conclusion and Relevance

In this Mexican population, existing PRSs derived from predominantly European ancestry populations predicted premature CHD independently of established vascular risk factors, particularly in men. Polygenic risk scores better capturing genetic variation in Latin American men and women may further enhance CHD risk prediction among Mexican and other Hispanic populations.

Key points