WASHINGTON – A genome-wide association study of the Million Veteran Program confirmed three specific genes associated with nonalcoholic fatty liver disease, underscoring the robustness of those loci as well as the clinical phenotyping in the program.
Marina Serper, MD, of the Cpl. Michael J. Crescenz Veterans Affairs Medical Center, and University of Pennsylvania, both in Philadelphia, and her colleagues looked at patients with NAFLD in the Million Veterans Program (MVP), a project of the federal Precision Medicine Initiative designed to leverage the data and experience associated with the Veterans Health Care Administration, Dr. Serper said at the annual meeting of the American Association for the Study of Liver Diseases. Currently, more than 600,000 veterans have been enrolled at over 50 sites across the United States, with a goal of 1 million participants by 2020.
About one-third (108,458) of 352,953 MVP enrollees whose DNA has been analyzed met the study definition of NAFLD. In their study, Dr. Serper and her associates defined the clinical phenotype of NAFLD as patients having abnormal alanine aminotransferase levels (greater than 30 U/L for men and greater than 20 U/L for women) detected twice in a 2-year period, plus at least 1 metabolic risk factor, such as body mass index of 30 kg/m2 or greater, type 2 diabetes or prediabetes, hypertension, or dyslipidemia. Further, included patients did not have alcohol misuse disorders or viral hepatitis.
Most patients were male (90%) and white (72%), with a median age of 64 years. More than half (56%) had a BMI of 30 or greater, 30% were diagnosed with type 2 diabetes, and 71% with dyslipidemia – aligning the cohort closely with rest of the MVP population, Dr. Serper said.
Logistic regression analysis adjusted for age, sex, and principal components stratified by ancestry (European, African American, and Hispanic). On initial analysis, 21 genetic loci met the criteria for genome-wide significant association; specifically, investigators successfully replicated three key variants that have been previously seen associated with NAFLD – PNPLA3, ERLIN1, and TRIB1.
“We were able to use clinical VA data to come up with a robust and clinically relevant definition and validate that definition because the genes we found associated with our definition of NAFLD have previously been shown by others who used biopsy data and imaging data for steatosis,” Dr. Serper said in a video interview. “This is important because the diagnosis of fatty liver disease is really a clinical diagnosis.”
Panel moderator Elizabeth K. Speliotes, MD, of the University of Michigan, Ann Arbor, said, “Really what makes us unique is our genetics and our exposures and the environment, and if we can capture that better, then we can use that to more precisely tailor diagnoses and treatments for patients. That’s really the hope of the next generation.”
The study was supported by the VA Office of Research and Development award 1I01BX003362. Dr. Serper disclosed no relevant conflicts of interest.
Watch this video interview with Dr. Serper and Dr. Chang for more information on the Million Veteran Program.
On Twitter @denisefulton