ATLANTA – Researchers at the American College of Cardiology's annual scientific meeting presented data from numerous studies, in which they tested different ways of developing and using polygenic risk scores in patients with cardiovascular diseases, signaling this as an active area of interest for the field.
The excitement over polygenic risk scores has been building in cardiology for years, as researchers have suggested that such scores could support early intervention for at-risk patients and guide prescribing decisions for patients who could benefit from certain drugs.
"There's been intense interest within the cardiovascular community as to whether polygenic risk scores can enhance the way we clinically manage patients," Lathan Liou, a medical student at Icahn School of Medicine at Mount Sinai in New York City, said during a poster session on Monday.
Liou presented on different approaches of creating risk scores that predict types of coronary artery disease (CAD), including findings that a particular approach, pathway-based polygenic risk scores, can increase predictive power in certain conditions.
CAD is a complex and heterogeneous disease, with multiple potential contributors, researchers wrote in their abstract. As such, they defined five subtypes of CAD, characterized by whether a patient had ST-elevated myocardial infarction, high low-density lipoprotein (LDL), high lipoprotein(a), occlusive disease, or unstable disease.
Researchers then tested multiple risk scores, including one that only contained clinical risk factors, as well as others that combined clinical factors and polygenic risk scores, and clinical factors and pathway polygenic risk scores. Specifically, pathway polygenic risk scores incorporate single-nucleotide polymorphisms that have been linked with biochemical and metabolic pathways associated with CAD.
They applied the risk scores to data from nearly 21,000 patients with CAD from the UK Biobank and found that inclusion of pathway polygenic risk scores improved prediction of certain disease subtypes, such as CAD characterized by high LDL and high Lp(a). In particular, polygenic risk scores related to pathways for fibronectin binding, apolipoprotein binding, and aerine-type endopeptidase activity proved strongly predictive of high Lp(a).
However, in unstable CAD, for example, the pathway polygenic risk score held less predictive value. "This suggests to us that for particular subtypes it could be that the risk is mediated less by your genome … and more so by a lifetime's worth of accumulated risk factors," Liou said.
These risk scores may illuminate the biological underpinnings of CAD, which could identify targets for precision therapeutics, researchers wrote in the abstract.
In a poster session on Saturday, researchers reported that combining a polygenic risk score with the coronary artery calcium (CAC) score was more effective than either score individually at identifying patients likely to develop high systolic blood pressure, and thus, these patients may benefit from antihypertensive therapy. The polygenic risk score, which comprises about 1.1 million genetic variants, was previously developed and validated as an effective predictor of cardiovascular events by the same research team, which included Naman Shetty, an instructor fellow at the University of Alabama at Birmingham and a coauthor of the poster. However, in this study, Shetty and colleagues wanted to build evidence on the score's ability to support treatment decisions.
In the retrospective analysis, researchers drew on data from more than 5,200 patients within multiple datasets. They excluded patients with prevalent atherosclerotic cardiovascular disease (ASCVD) and those already on antihypertensives. Researchers used hypertension treatment guidelines from the ACC and the American Heart Association to identify patients with elevated blood pressure or low-risk stage I hypertension for whom antihypertensive treatment typically wouldn't be recommended. They then stratified these patients using the polygenic risk score and the CAC score. They compared predictions from the risk scores against estimates from the US National Institutes of Health's Systolic Blood Pressure Intervention Trial study on 10-year ASCVD incidence rate and the number of patients that would need to be treated to prevent ASCVD events.
The polygenic risk score demonstrated "modest value" as a tool for identifying patients at high risk for ASCVD events unlikely to get treatment based on current guidelines. While the CAC score outperformed the polygenic risk score slightly, ultimately, "using both of these tools would improve risk stratification and help guide antihypertensive treatment," Shetty said.
Risk assessments that include social determinants of health and lifestyle contributors to disease alongside polygenic risk and clinical factors could also improve predictions, according to research presented Sunday. Researchers using data from the UK Biobank determined that both a polygenic risk score and social determinants of health and lifestyle factors risk score correlated with clinical risk of patients developing incident coronary heart disease, but a combined measure outperformed both. And by being able to better predict which patients are likely to develop disease, physicians hope to intervene earlier to prevent it.
Researchers on Saturday also presented data showing that a polygenic risk score, initially developed to predict coronary artery disease, was effective at identifying measures of subclinical CAD, such as the presence of plaque and stenosis, specifically in HIV patients. Such a score could inform prevention efforts in this population, for which CAD is a leading cause of death, said Roger Zou, an internal medicine resident at Massachusetts General Hospital and a poster coauthor. The retrospective analysis included data from about 660 patients whose HIV was well-controlled, who were on antiretroviral therapy, and who otherwise would be considered to have low-to-moderate ASCVD risk based on clinical and environmental factors.
In yet another study researchers discussed on Saturday, a polygenic risk score developed to predict atrial fibrillation also showed promise in predicting incidence of certain cerebrovascular events, such as ischemic stroke and transient ischemic attacks. That analysis was based on data from about 2,500 patients enrolled in the Sanford Heart Screening Program. Such a polygenic risk score for identifying patients' predisposition to these cerebrovascular events could support earlier treatment initiation, said Andrii Maryniak, a cardiology fellow at the University of South Dakota’s Sanford School of Medicine. "There definitely is a role for polygenic risk scoring," he said.