Skip to main content
Premium Trial:

Request an Annual Quote

Addition of Genetic Risk Score Improves Chances of IDing Atrial Fibrillation Patients, Study Finds

Premium

NEW YORK (GenomeWeb) – Researchers from Quest Diagnostics and the Scripps Translational Science Institute published a study earlier this month demonstrating that adding a genetic risk score to the standard clinical considerations (weight, age, high blood pressure and other factors) can better identify patients with atrial fibrillation (AF).

According to Quest, the predictive power demonstrated in the study — which was published March 13 in PLoS One — is persuasive enough to support efforts to translate the approach to a clinical test, although there are still important questions to be answered before its clear that that would be feasible and/or profitable.

While genetic factors have been associated with AF in previous research, studies have mostly been retrospective and focused on long term risk.

AF is a commonly encountered arrhythmia associated with a higher risk of stroke. But because individuals may not know that they have the condition before experiencing a stroke, there is a significant clinical need for better ways to identify patients with the highest risk for AF so that they can be watched more carefully, diagnosed, and treated if necessary.

In the Scripps study this month, investigators looked at risk on a short time frame — evaluating whether their genetic score could pick out the individuals most likely to be diagnosed with AF during a two-week monitoring period after they presented to a doctor with suspicious symptoms.

Investigators recruited patients who were are least 40 years old, took a blood sample for genetic analysis using a weighted contribution of 12 risk loci, and followed them using an adhesive patch monitor or a longer-term device called a Holter monitor and recorded whether they went on to be diagnosed with AF based on an ECG or either of the monitoring devices.

According to the authors, 85 out of the 904 participants manifested AF. Those in the highest quintile of the genetic risk score were more likely (an odds ratio of 3.11) to have had an AF event than participants in the lowest quintile even after adjusting for age, sex, smoking status, BMI, hypertension, diabetes mellitus, heart failure, and prior myocardial infarction.

"This study was really about trying to look at this in a specific clinical setting, and what we chose was patients who show up with a complaint … putting them on monitoring, and then asking the question, can we use this risk score to identify those at greater risk," said Dov Shiffman, a Cardiology Research and Development Fellow at Quest Diagnostics and an author of the new study.

"Based on the results, it looks like we can," he added. "We saw that top quintile had a threefold greater risk than those in the bottom, which is even better than what we saw in earlier studies, which was about a twofold [increase]."

AF is not the only cardiovascular condition where genetic risk testing is being investigated. Polygenic and gene expression scores have also garnered much attention in predicting coronary artery disease (CAD).

So far, few genetic scores have been taken as far as the Quest and Scripps team's approach, though some have. CardioDx, for example, commercialized a gene expression panel for assessing CAD risk several years ago. And more recently, researchers from the Broad Institute showed in prospective analyses that patients at a higher risk for CAD, based on a polygenic score they developed, appeared to benefit significantly from healthy lifestyle factors like not smoking, being slim, and exercising regularly.

Polygenic risk scores are also gaining ground in other medical settings. Boutique health coaching company Arivale said last year, for example, that it was incorporating polygenic analyses to gauge customers' risk for high lipids, obesity, and exercise tolerance or ability.

Myriad genetics has also launched what it calls riskScore — an 86-SNP breast cancer risk algorithm that it believes can help inform the care of the large proportion of women who lack mutations in well-known cancer-linked genes like BRCA.

Based on the strength of the results in its AF study, Quest is now evaluating options for a test service. He did not provide details on how Quest may be arranging next steps toward bringing a test to the clinic, but said that the prospective study results merit significant attention being paid to the AF signature.

"I'm more on the research side, but I'm pushing this very hard and [the company] is looking into this seriously," he said, adding that the clinical utility proposition for the genetic risk score is fairly simple to demonstrate.

"We know that anticoagulants can dramatically reduce the risk of stroke when you have AF," he explained. "Its up to clinicians [how to treat patients], but many would agree that if they know a patient has a higher AF risk, they would send them to monitoring, and then there are established clinical guidelines for when to prescribe anticoagulants," he added.

In the Scripps study, Thermo Fisher Scientific's SNaPshot multiplex genotyping instrument was used to simultaneously genotype the 12 SNPs comprising the AF risk score, but a commercial test could potentially use a variety of platforms.

The challenge, according to Shiffman, is establishing whether the risk score can be adapted into a practical and cost-effective test, as well as cementing the best clinical niche to target it at.

For example, he said, the group's current study focused on predicting which patients would be diagnosed with AF during monitoring with a wearable sensor. But another area that this might be useful to clinicians is helping them determine which patients who have had a stroke but haven't been diagnosed with AF might most need to be assessed for it.

Important limitations of the current study also include the fact that the recruited cohort was very ethnically homogenous, while evidence has shown that the same variants can confer different risks of AF in white and black individuals, for example.

The 12-gene signature used in the study also doesn't include some more recently identified risk loci, which could potentially improve its predictive ability, the authors wrote.