Genomic Determinants of Cardiometabolic Traits in South Australian Aboriginal Communities

Abstract

Background: Cardiovascular disease is the leading contributor to the mortality gap between Aboriginal and Torres Strait Islander peoples and non-Indigenous Australians, with events occurring 10-20 years earlier and mortality rates approximately 1.5 times higher. While social determinants substantially drive these disparities, the contribution of genetic factors has remained unexplored due to the underrepresentation of Aboriginal people in genomic research. Underrepresentation creates direct barriers to precision medicine, as variant interpretation algorithms, polygenic risk scores (PRS), and clinical genomics tools developed predominantly in European populations perform poorly when applied to populations with diverse genetic ancestries.

Aim: This thesis investigates whether genomics can improve the understanding and prediction of coronary heart disease for Aboriginal people, addressing a critical gap in Indigenous health research.

Methods: The Predicting Renal, Ophthalmic and Heart Events in the Aboriginal Community (PROPHECY) study enrolled 1,376 Aboriginal people living in South Australia with comprehensive phenotyping for type 2 diabetes, coronary heart disease, and cardiometabolic risk factors. Whole-genome sequencing for 1,231 participants generated the largest Aboriginal genomic dataset to date. Analyses included characterisation of population genetic diversity and comparison with global reference databases; the first systematic screening of 72 monogenic cardiovascular disease genes for loss-of-function variants in Aboriginal people; the first genome-wide association study of 14 cardiometabolic traits under additive and recessive inheritance models with local ancestry inference for Aboriginal peoples; and evaluation of clinical risk prediction tools and polygenic risk scores for coronary heart disease.

Results: Approximately 24% of the genetic variants identified in South Australian Aboriginal communities were absent from gnomAD, the world's largest genomic reference database, demonstrating substantial underrepresentation with direct clinical implications for variant interpretation. Screening of monogenic cardiovascular disease genes identified 35 high-confidence loss-of-function variants across 17 genes, with 37% of these variants absent from global databases. Ten variants with pathogenic classifications span conditions including hypertrophic cardiomyopathy, dilated cardiomyopathy, long QT syndrome, and familial hypercholesterolaemia. Genome-wide association analyses identified 13 genome-wide significant and 5 suggestive associations with cardiometabolic traits, including strong signals at LPA and APOE loci, with 10 associations detected exclusively under recessive inheritance models. Current clinical risk calculators showed only moderate discrimination for prevalent coronary heart disease (AUC 0.77 to 0.81), with substantial underperformance in adults under 40 years. PRS demonstrated independent predictive value, with combined clinical-genetic models achieving superior performance and moderate-risk reclassification strategies yielding net reclassification improvement of 35% to 55% for males and females, respectively.

Conclusions: Aboriginal communities in Australia possess a unique genomic architecture shaped by over 50,000 years of evolutionary history, with substantial variation invisible to current clinical databases. This thesis establishes foundational resources for precision medicine equity, demonstrating that monogenic cardiovascular disease variants are present in Aboriginal communities, that both shared and population-specific genetic architecture underlie cardiometabolic traits, and that genetic information can improve cardiovascular risk prediction where clinical tools underperform. Conducted under Aboriginal governance and aligned with Indigenous data sovereignty principles, this work provides a model for ethical genomic research that advances both scientific knowledge and health equity. Show more