The Racial and Ethnic Inequity in Genomic Medicine -- A Bridge Too Far

This is an exciting time for cardiovascular disease research. The sequencing of the human genome has ushered in a new era of genomic research, which during the last decade has led to new avenues for cardiovascular disease prevention and treatment. Results from large clinical population studies and GWAS are being translated into clinical practice to improve risk stratification, ^1-4 identify responders and non-responders to pharmacologic agents, ^5-7 and develop more personalized approaches to health care. ⁸ Numerous highly penetrant risk alleles have been identified that are currently being used for genetic testing of patients with presumed heritable forms of heart disease and their family members.^{9, 10}

Regrettably, there are huge inequities in those likely to reap the benefits accruing from this new era of genomic medicine. As with other inequities in the health care system, non-European populations are being left out. A driving force in this inequity is that minority (or more specifically, non-European) populations are much less likely to participate in genomic research from which many translatable discoveries are made. In the U.S. ~ 40% of the population is of primarily non-European ancestry, and that proportion is only expected to increase. Globally, non-European ancestry populations make up 84% of the world’s population.¹¹ Despite these numbers, only 22% of individuals included in genome-wide association studies (GWAS) are of non-European descent, according to an analysis of studies published through 2018.¹² In recognition of this gap, a number of efforts have been made to increase genetic diversity and representation in GWAS studies across a range of non-EUR populations. These include the Human Heredity and Health in Africa (H3Africa) initiative, supported by NIH and the Wellcome Trust, and in the U.S., NIH-supported initiatives such as the All of Us Program and the Trans -Omics for Precision Medicine (TOPMed) Consortium.

The underrepresentation of minority populations in current genomic research leads to a significant European bias in terms of the composition of what variants are discovered. Because of differing population histories, many disease-causing and disease-associated variants present in Europeans may not be present in non-European populations or may be present at lower allele frequencies. Conversely, many disease-causing and disease-associated variants important in non-European populations may not be identifiable in European populations. As a result, it is well appreciated that polygenic risk scores (PRS), which are increasingly being used for risk stratification, significantly under-perform when applied to non-European populations because most have been generated based on GWAS results derived from European populations.¹¹ Sound application of PRS requires well-powered GWAS within the ancestry group in which the PRS is to be applied.

Inclusion of ancestrally diverse populations in genomic studies is not only equitable, but also generates new genetic discoveries that may be translatable to members of all populations. Two examples are provided below, but there are many others. The first example is the discovery of a loss of function variant in APOC3 in a genetically isolated Amish community. This variant produces a dysfunctional APOC protein, which is unable to inhibit lipoprotein lipase and hepatic lipase from breaking down triglyceride. As a result, mutation carriers have decreased levels of serum triglycerides, and reduced risk of coronary heart disease.^{13, 14} Even though this mutation was discovered in the Amish, the idea of inhibiting APOC3 to decrease triglycerides and promote cardiovascular health is generalizable to individuals across ancestry groups. The second example is the 2016 discovery in an indigenous Samoan population of a mutation in CREBRF that is associated with a marked increase in body mass index, yet paradoxically, also which confers significant protection against type 2 diabetes.^{15, 16} Ongoing research of this gene may provide novel insights about fat storage and energy use that may help dissociate the co-occurrence of obesity and diabetes and reduce the burden of diabetes. As with the APOC3 example, the biological insights derived from this discovery can potentially benefit all people, including populations in whom the mutation is not present.

So why are minority and indigenous populations being left out of genomic research? The reasons for their underrepresentation are complex and multifactorial, but some of the barriers to participation may be confounded by mistrust by prior experiences with genetic research. Examples of historical transgressions experienced by some communities that have been well publicized include the conduct of unethical trials (e.g., Tuskegee Study of Untreated Syphilis¹⁷), the unauthorized use of tissues and DNA taken from patients without permission (e.g., the establishment of HeLa cell lines from Henrietta Lacks ¹⁸), and a lack of sensitivities to cultural norms of the population.¹⁹

Against this backdrop, Mudd-Martin et al. have prepared a thoughtful discussion outlining considerations for promoting the engagement of racial and ethnic minority groups and indigenous populations in genomic research. The authors provide an overview of ethical guidelines for genetic and genomic research, discuss special considerations regarding genetic and genomic research, and outline a series of practical considerations when conducting cardiovascular genetic and genomic research with minority and Indigenous populations. They discuss issues relating to establishing community collaborations, the informed consent process, transparency around sample use and in data analyses, dissemination of findings, return of genetic results, and protections surrounding data sharing.

Increasing the engagement of minority and underrepresented populations in genomic research is no easy task. Including the representation of more diverse populations in genome-wide association studies is a necessary, but not sufficient, first step. As described by Mudd-Martin et al., active engagement requires the establishment of trusting community partnerships. There is a lot of work to be done but bridging this gap in participation and access to genomic research must be a priority.