NEW YORK (GenomeWeb) – Despite a limited budget, the US Department of Veterans Affairs last year met its goal to enroll 500,000 participants into a study that aims to recruit 1 million and use genetics and other data to study the diseases that impact former military service members.
While all participants have had samples genotyped for research use, the VA has been steadily increasing its capacity to whole-genome sequence enrollees, and now has the funds to sequence as many as 30,000 enrollees over the next two years.
The Million Veteran Program, launched in 2011, aims to collect a variety of data — genetic, medical, lifestyle, and military exposure information — from 1 million vets and glean from that a better understanding of the factors that cause conditions such as post-traumatic stress disorder, schizophrenia, and bipolar disorder. The goals of the MVP are similar to the Precision Medicine Initiative, and MVP has agreed to help recruit veterans interested in joining the national effort to advance personalized medicine research. After President Barack Obama launched PMI two years ago, the White House gave the department a goal to recruit half the MVP cohort by the end of 2016.
When the project was launched six years ago, the VA had aimed to enroll the entire cohort over five to seven years. The VA is just more than halfway there and may still meet that target timeframe though it has been operating on a tight budget — a challenge that the PMI may also contend with as it embarks on its goal to recruit a million participants over four years.
According to MVP Director Sumitra Muralidhar, the VA spent less than $200 million on the program over the last five to six years. Comparatively, in FY2016, the PMI received $200 million, of which $130 million went to the National Institutes of Health to begin building the cohort. The $300 million Obama requested for PMI in FY2017 isn't secure, particularly in light of President Donald Trump's budget blueprint proposing to cut NIH's budget by $1.2 billion.
The prospect of such deep budgetary cuts to the NIH sent shockwaves through the life sciences field. Although many policy experts have opined that Republican legislators will not vote to defund the country's research enterprise to this degree, some are already sounding the death knell for the PMI and fear that amid reshuffling policy priorities, personalized medicine will not rank high for the Trump administration.
There are some important differences between the PMI and MVP, but the steady progress in the latter project despite a lean budget should be an encouraging sign. "It was really about how much we could do with the money and where we could expand our bandwidth," Muralidhar said.
While the NIH is taking a high-tech route with the PMI in some respects, the VA took a low-tech approach with the VA. For example, the NIH is building a web portal through which individuals can enroll directly into the initiative. The experts involved in planning the PMI have envisioned maintaining contact with participants to glean longitudinal data through emails and mobile apps. The NIH will be doing genetic testing in a CLIA-certified lab setting, and has promised to return aggregate research data and some individual-level genetic testing results back to participants.
Within MVP, genetic testing is not done in a CLIA-certified lab, so results aren't returned, which in turn helps cut back on costs. The VA's population, by comparison, also requires a "high touch," Muralidhar said, and wouldn't readily use an online portal to enroll into MVP. There are more than 9 million veterans enrolled in the VA, 90 percent of whom are male, and the median age of those using VA benefits is 64 years old.
"So, we didn't go for high tech. We went for a more basic approach," she said. "We used strategies that we knew our population would respond to better."
The MVP team mails out invitations about joining the study to veterans and includes information about the program in their pharmacy packets. Recruiters engage with veterans when they go in for regular hospital appointments. Physicians and researchers at local VA hospitals also help spread the word about the program.
"We started slow," Muralidhar said, recalling that when the MVP first launched, it was recruiting participants at only two VA hospitals. Now, the project is enrolling at 50 VA hospitals around the country and at 60 VA community-based outpatient clinics.
The project's enrollment has also been reflective of the larger veteran population in the US in terms of race and ethnicity, she added. VA users are 81 percent white, 16 percent African American, 1 percent Native American, and 6 percent Hispanic. Within MVP, 77 percent of participants are white, 20 percent African American, 3 percent Native American, and 8 percent Hispanic.
Two subpopulations that are underrepresented within MVP are women and younger veterans. Women comprise approximately 10 percent of veterans who receive health benefits from the VA, Muralidhar said, but make up between 8 percent and 9 percent of the MVP cohort.
The MVP is teaming up with the Center for Minority Veterans and the Center for Women Veterans to ramp up recruitment and engagement efforts. Last month, the MVP also held a women's health initiative to further encourage enrollment, and the women's health hospitals within the VA are working in tandem with program coordinators to over sample women.
But 8 percent of 550,000 enrollees still "gives a good number from a research perspective to study women's health issues," Muralidhar said.
Recruiting younger vets has also been a challenge. More than half the enrollees into MVP are between 60 and 69 years old, but only 3 percent of veterans younger than 30 years have joined. This may be due to the fact that younger service members are healthier and don't come into the hospitals or clinics regularly. MVP's low-tech approach to recruitment could be another factor.
"We prioritize [enrolling] veterans who have upcoming appointments," Muralidhar said, noting that invitations to join the study are mailed out to those who have upcoming hospital appointments in the next two or three months.
"It may be for the younger generation, if you get something in the mail, they don't often open something and don't actually respond," she noted. To address this, Muralidhar noted that her group has been piloting surveys on the web for MVP and will award a contract for the development of web portal.
While MVP has been recruiting participants, it has also rolled out a number of early-phase studies that are helping to test out protocols and the computing infrastructure before opening up data access to the larger research community. Two alpha-phase studies are nearing completion — one investigating genetic and other variables contributing to PTSD in combat-exposed veterans, and another on schizophrenia and bipolar disease.
MVP is also doing beta studies on a range of chronic conditions impacting veterans, such as cardiovascular disease, chronic kidney disease, and macular degeneration. Another study on the genetics of Gulf War illness — a set of unexplained symptoms such as fatigue, headaches, insomnia, and memory problems — is just getting off the ground. And VA also put out an RFA earlier this year seeking to fund an additional 10 to 12 projects by the end of this year.
When the data within MVP is available to researchers, they will have access through the GenISIS (Genomic Information System for Integrative Science) computing environment. Within MVP there are teams that are cleaning and curating the data from electronic health records (HER) and genetic testing.
Muralidhar noted that the MVP has partnered with the US Department of Energy last year to leverage its high-performance computing infrastructure and expertise in data science. A copy of the EHR data and genotyping results in MVP has already been moved to one of DOE's national labs. "We're just starting to explore some of the capabilities in this partnership and planning some pilots," she said.
In terms of the genetic information, so far all enrollees' samples are genotyped using a custom Affymetrix array, and the data have been cleaned, curated, and imputed, and are available for the alpha and beta studies. Exome sequencing has been performed on a subset of 28,000 samples using two different platforms, and whole-genome sequencing has been done on 2,000 samples. These data are still being processed for research use.
"We decided for our budget we can't sequence everybody," Muralidhar said.
However, according to its 2017 budget and 2018 advance appropriations request, the VA is "prioritizing its research portfolio towards precision medicine" this year and making a "substantial investment" in genomic sequencing MVP enrollees. Muralidhar said that MVP has $40 million — $10 million from FY2016 and $30 million from advanced FY2017 appropriations — that her team will be able to put toward whole-genome sequencing between 25,000 and 30,000 samples.
"As our budget allows, we're going to be getting more and more in-depth data on these samples, cleaning them to the extent possible, and making them available to researchers," she said.
While MVP is currently a veterans-only project, Muralidhar and her group will be allowing active duty members to join the study. They are working with the US Department of Defense's Millennium Cohort Study, which is investigating the impact of military exposures on 200,000 participants, and involves active service members and veterans. Muralidhar said MVP will begin by enrolling the vets in this study who are already in the VA system, then move on to veterans who are not enrolled, and then expand to active duty members.
Again, sequencing will take place as the budget allows. "Instead of randomly sequencing everyone, we will start with phenotypes that are of relevance to veterans and active duty members," she said. "That way we do something unique to the VA and DoD."
Finally, although study participants are currently not getting any genetic test results, Muralidhar indicated this could also change in the future. The VA anticipates launching a pilot project in May to test out how to return pharmacogenetic testing results for veterans taking drugs for major depressive disorder.
"We're going to look at the results of that pilot and see what the best practices are around returning results," she reflected. "We hope to learn from that … and as we get closer to enrolling our millionth veteran, there might be a subset of the population for whom we can do testing in a CLIA-certified lab and return results."