NEW YORK (GenomeWeb) – A team led by investigators at the Fred Hutchinson Cancer Research Center, Icahn School of Medicine at Mount Sinai, and the University of North Carolina at Chapel Hill demonstrated that the common genetic risk variants implicated in human traits and diseases — and the extent of heritability explained by them — can differ significantly across populations with different ancestral backgrounds.
Those involved in the study argued that such realizations need to be taken into account when developing genomics-based risk assessment or treatment strategies to meet the healthcare needs of diverse human populations, while narrowing existing healthcare gaps.
"Because the availability of non-European genomic data is limited, existing clinical therapies may disproportionately benefit those of European descent, further widening the health disparities gap," co-senior author Eimear Kenny, a genomic health, personalized medicine, and psychiatric genomics researcher at Icahn School of Medicine, said in a statement.
In their report, published online today in Nature, she and her colleagues added that "[i]n the United States — where minority populations have disproportionately higher burden of chronic conditions — the lack of representation of diverse populations in genetic research will result in inequitable access to precision medicine for those with the highest burden of disease."
As part of the National Institute on Minority Health and Heath Disparities and National Human Genome Research Institute's "Population Architecture using Genomics and Epidemiology" (PAGE) study, the researchers included more than 49,800 non-European individuals in GWAS spanning more than two-dozen clinical conditions or behavioral phenotypes, ranging from height, body mass index, or cholesterol levels to blood pressure, coffee consumption, or kidney disease.
Along with known common genetic contributors, the team detected associations at 27 loci not linked to the conditions or physical features in the past. Even when SNP associations did replicate in the non-Europeans, though, the effect sizes associated with those variants often differed significantly from one ancestral group to the next, highlighting the importance of fine-mapping and population-specific genetic analyses.
"Previous articles have alluded to the need for multiethnic diversity in genome-wide studies, but our study is among the first to clearly delineate the scope of the problem, using detailed analyses of minority genetic samples," co-senior author Christopher Carlson, a public health sciences researcher at the Fred Hutchinson Cancer Research Center, said in a statement.
For the new set of GWAS, collaborators at the Johns Hopkins University Center for Inherited Disease Research used Illumina's Multi-Ethnic Genotyping Array to genotype 49,839 participants: a group that included 22,216 individuals who self-identified as Hispanic/Latino; 17,299 African Americans; 4,680 Asians; 3,940 Native Hawaiian participants; and 652 individuals who identified as Native American. The remaining 1,052 individuals came from still other self-reported ancestral groups.
Reasoning that the participants "reside on a continuum of genetic ancestry, rather than discrete population groups," the authors developed a joint analysis, multi-ethnic-tailored computational approach for modeling population structure, evaluating heterogeneous genetic variation along this spectrum, and picking up genetic associations with the 26 clinical, physical, or behavioral phenotypes considered.
"Critical variants may be missed if they have a low frequency or are completely absent in European populations, especially as the field shifts its attention towards rare variants, which are more likely to be population-specific," the authors explained, noting that "effect sizes and their derived risk prediction scores derived in one population may not accurately extrapolate to other populations."
By further taking into account minor allele-frequencies, the team narrowed in on 16 significant trait-locus associations not described in the past, along with 11 more suggestive associations involving loci with low-frequency variants. Another 32 significant and six suggestive secondary associations turned up at known loci.
Along with analyses focused on associations that tracked with specific ancestral components in participant genomes, the researchers compared the trait-locus associations with those documented in the GWAS Catalog for the 26 phenotypes considered. They replicated 1,444 associations, though more than half of the associations that reached genome-wide significance fell at loci with ancestry-related genetic heterogeneity.
Likewise, when the team set the anthropometric trait associations alongside those described for more than 250,000 European GIANT study participants or 50,000 individuals from the UK Biobank, it found that loci picked up by GIANT explained far more phenotypic variance in the cohorts with European ancestry than in the PAGE participants.
Based on these and other findings, the PAGE study authors touted the value of ancestry-specific fine-mapping at risk loci, along with efforts to engage and study diverse populations for future association and precision medicine efforts.
"We strongly advocate for continued, large genome-wide efforts in diverse populations to maximize genetic discovery and reduce health disparities," the authors wrote, pointing to efforts such as the All of Us Research program.