Diverse Population Data Sheds Light on Genetics Underlying Coronary Artery Disease

by Adrienne Mueller, PhD
August 3, 2022

Coronary artery disease (CAD) is the leading cause of death worldwide. Large-scale population genetic studies help us understand the inherited basis of complex traits such as CAD. Previous studies suggest fatal CAD has a genetic heritability of 40-60%. As with many diseases, historically the populations used to study CAD were largely of European, South Asian, or East Asian descent. Even 15 years after CAD was known to be partially heritable, there have been no large-scale genetic studies for Black or Hispanic populations – populations that are disproportionately affected by CAD.

To bridge this gap of knowledge, Stanford University’s Themistocles L. Assimes, MD, PhD co-led a study with Philip Tsao, PhD (Stanford University), Yan Sun, PhD (Emory University), and Christopher O'Donnell, MD (Harvard Medical School). They used data from the Million Veteran Program to perform the largest multi-ancestry genome-wide association study of coronary artery disease to date. The Million Veteran Program has data from over 400,000 participants, including approximately 118,00 with CAD. The results of this study were recently reported in a Nature Medicine article, co-first authored by Catherine Tcheandjieu, PhD, Xiang Zhu, PhD, Austin Hilliard, PhD, and Shoa Clarke, MD, PhD. Since the study began, Dr. Tcheandjieu has been appointed faculty at the Gladstone Institute and Dr. Zhu at Penn State University. Dr. Hilliard is a senior computational biologist at PAVIR and is part of the Palo Alto Veterans Association research group working on the Million Veteran Program. Dr. Clarke is currently an Instructor at Stanford.

Figure: Circos plot indicating the association of population-specific and multi-population genome-wide association study meta-analyses with coronary artery disease. Association results are shown for Black (red), Hispanic (green), and White (black) populations, as well as multi-population (blue). The red line indicates genome-wide significance. The outer track lists the nearest mapped gene to the lead SNPs reaching GWS in each of these four meta-analyses including five loci in Blacks (red), three loci in Hispanics (green), 33 novel loci among Whites (black), and 62 additional novel loci after the multi-population meta-analysis (blue).

To identify the genetics underlying CAD in diverse populations, the investigators first assessed the heritability of CAD across four ancestrally very different populations including individuals with a high very high proportion of their DNA originating from either Europe (White), Africa (Black), Japan (East Asian), or the Indigenous (Native) people of America. CAD heritability was similar across all populations. Furthermore, the first eight CAD susceptibility loci uncovered in non-Hispanic Black and Hispanic participants in this study were found to be the same loci that had been uncovered in European and/or East Asian cohorts several years ago. However, meta-analysis of this extremely large dataset allowed the discovery of 95 new loci reaching genome-wide significance (see Figure.)

These new loci contribute to CAD both within individual ancestral populations and across all populations. Lastly, the investigators also tested the performance of polygenic risk scores generated from this large, diverse dataset. Polygenic risk scores are a means of measuring the risk of developing a specific condition, in this case CAD, based on the combined influence of many genetic variants. The Million Veteran Program data from this study yielded improved performance of polygenic risk scores, but these scores still performed worse for Black populations than other populations.

In summary, the Million Veteran Program genome-wide association study has shed light on the genetic underpinnings on the most common cause of death worldwide: coronary artery disease. By including more diverse datasets in this study, the researchers were able to have a more reliable identification of causal variants, identify new loci that contribute to CAD, and enhance the performance of polygenic risk scores. This study strongly benefited from the collaboration of a multidisciplinary team of bioinformaticians, geneticists, epidemiologists, and clinicians, at numerous academic institutions throughout the US and at the US department of Veteran Affairs.

Additional currently or previously CVI-affiliated investigators with this study include Valerio Napolioni,PhD; Shining Ma, PhD; Huaying Fang, PhD; Derek Klarin, MD; Nasa Sinnott-Armstrong, PhD; Jennifer S Lee, MD, PhD; and Hua Tang, PhD.