Scientific studies aimed at unlocking the genetic code of all humans have had problems with their data: it was mostly collected from people with “European” heritage, leaving wide gaps in the study of DNA from populations around the world.
To close the gap, the authors of one recent study turned to military veterans, a group whose health and genetics are well studied during and after military service, and whose genetics come from a wide swath of the world.
Researchers working with the Department of Veterans Affairs were granted access to the VA’s database of DNA known as the Million Veteran Program. Using the DNA database from the VA, researchers found genetic markers for prostate cancer, anemia, Alzheimer’s dementia and cirrhosis, according to a study published in Science.
“Most of the genetic data available to researchers are still derived from individuals of European descent,” researchers wrote. “This shortcoming limits both the biological insights that can be gleaned from these data and their clinical applications to non-European patients, who may not match up well with the traditional study participants.”
The Million Veteran Program was launched in 2011 as a research effort to improve veteran health care. It’s also one of the largest biobanks in the world, collecting DNA and other health information on veterans for medical research. Military veterans have long been a more racially and genetically diverse group than the U.S. as a whole.
The recent VA-funded study was done in collaboration with the Department of Energy in order to use their supercomputers to run thousands of genetic-disease analyses using MVP data.
According to Anurag Verma, a researcher at the Corporal Michael J Crescenz VA Medical Center, most genomic studies rely heavily on European ancestry DNA which limits the accuracy of research results.
“If we include more diversity in these studies, then we are able to overall improve the risk prediction,” said Verma, a lead author of the study and an assistant professor of medicine at the University of Pennsylvania.
In the study with Veteran DNA, researchers were able to find 101 traits – including hemolytic anemias, sarcoidosis, keloid scarring and susceptibility to gout – among veterans with African ancestry that were twice as prevalent than in veterans with European ancestry. The research also validated previous studies on African ancestral populations which found a higher prevalence of traits linked to prostate cancer, reduced white blood count levels and kidney-related conditions such as end-stage renal disease.
Among veterans with East Asian and Admixed American ancestry (a term that typically encompass those who self-identify as Hispanic or Latino), researchers found 18 traits “with at least twice the prevalence” of veterans with European ancestry including Alopecia areata in Admixed American and viral hepatitis B in East Asian ancestry.
“This is an example where the donation that the million-plus veterans made to this program, it’s really a gift to the world,” said Sumitra Muralidhar, director of the Million Veteran Program.
In most genetic association studies, research teams study one disease and determine which genetics are associated with it or a researcher identifies one genetic marker and they try to link conditions or health conditions that are associated with that specific marker, Muralidhar said.
“By taking 42 million-or-so-plus genetic markers and about 2,000 health traits all at once and looking at this, we’ve already completed the first step so-to-speak for a number of health traits,” she said. “Now other researchers can really take this as a jumping point and expedite discovery and move it towards translation much faster.”
Closing the gap
The lack of diverse DNA in genomic research has been well-documented in published studies and news reports. A systematic review of existing genome-wide association studies from January 2024, found that 82% of 123 studies looking at neurodegenerative disease connections to DNA predominantly featured participants with European ancestry.
Access to veteran DNA, however, is helping to close that gap.
With the MVP study, which began in 2018, researchers were able to use veteran data from the VA’s biobank, which at the time was just over 638,000 individuals and about 29% had non-European ancestry.
“Not only the percentage is high, but absolute number of the individuals in this study is also massive in comparison to whatever has been published in the past,” Verma said.
The VA has since reached more than one million participants and as of July 24, MVP had 1,037,886 participants. In the current breakdown, around 25% of the cohort are racially diverse (non-European ancestry) with 18% African ancestry and 7% other racial minorities; 8% ethnically diverse; namely, Hispanic.
By conducting a diverse, cross-population analysis, researchers were able to identify 834 previously unreported variant-trait associations and 15 signals from coding variants that are either rare or not observed in non-European populations.
With a substantial amount of African ancestry data, researchers also found numerous pleiotropic genes, which are genes that control more than one trait. A common example of pleiotropic genes is phenylketonuria, a disorder caused by an enzyme deficiency that can result in multiple characteristics like mental retardation, eczema, and lighter skin pigments.
“This highlights the substantial contribution conferred by including diverse populations in genetic research,” the study states. “At the same time, cross-population heritability analyses, fine mapping, and heterogeneity analyses demonstrated substantial similarities in the genetic architecture between population groups driven by variants common across populations.”
Why does it matter?
Genome-wide association studies have long been the foundation of research into complex biological traits and drug development.
“People carry different genes and some genes where we call variants are more common in one population compared to another population and so how this plays out is in drugs,” said Katherine Liao, a lead researcher of the study and staff physician at the VA Boston Healthcare System. “There are certain drugs that if you carry a certain genetic defect, you’re gonna have a really bad side effect.”
Liao gave an example of 1.5% of one population carrying a specific gene and another population where 10% carry the gene.
“If you were to give everybody the same drug, the population where they don’t have the gene that gives you a side effect, nothing happens. So if you only test on that population, you think ‘Oh, this drug is really safe,’ but in another population, 10% are having some kind of massive side effect,” Liao, also a professor at Harvard Medical School said. “That’s where it really matters.”
Subscribe to Task & Purpose today. Get the latest military news and culture in your inbox daily.
Genome-wide association studies are also helping in the field of precision medicine where doctors look at genetics, environment, and lifestyle to select the best treatment for a patient.
“The promise of precision medicine is finding the best drug for the patient and also how do we manage it the best. If there’s only that one drug, it’s not like we want to avoid it, but it’s like how do we deal with that?,” Liao said. “How do we tell the patient ‘watch out for rash’ or ‘these are the issues you need to watch out for.’”
Expanding MVP
A number of other studies are underway using MVP data, with researchers looking at links between DNA and prevalence of certain types of cancers, diabetes, and cardiovascular disease, as well as substance abuse and mental health disorders.
But MVP has already led to “some of the largest studies ever done,” said M, MVP director.
One study that looked at the genetics of PTSD, had 165,000 veterans “which had never been done before” and another, which was “the largest study on genetics of anxiety” used data from 200,000 veterans.
During the pandemic, a research team observed African Americans with COVID-19 were dying of acute kidney disease at much higher rates than the rest of the population. By diving into MVP data, researchers found a gene called APOL1 that increased African Americans’ risk of death. With their findings, Muralidhar said, pharmaceutical companies can develop drugs that target the gene and reduce mortality risk.
While the study does note that MVP data is ancestry-diverse, its veteran population is predominantly male and older, making the research “less well-powered to study conditions more prevalent in females or younger populations.” But even if only 10% of MVP is made up of female veterans, the absolute number equals 100,000 female participants which MVP officials and the study’s researchers said is larger than the majority of existing biobanks.
Muralidhar said MVP has launched a couple of campaigns aimed at enrolling more women veterans. During one marketing campaign, MVP doubled the number of women participants and are developing focused campaigns for different races, ethnicities, genders, ages and even geographies for groups like rural veterans who are harder to reach. As part of the MVP sign-up process, veterans have to give a blood sample at a VA facility, but in order to expand the enrollment, MVP has started to mail a kit home for blood specimens.
Participation in MVP is voluntary and requires consent from each veteran. Enrollees have to complete online or mail surveys on their health, lifestyle habits, military experience, personal and family history, give a blood sample for genetic analysis, and agree to future contact from MVP.
“The altruism of veterans has made this possible – really without which we would never have been here,” Muralidhar said. “They really look at this as another opportunity to serve.”