Biomarkers Added to a Polygenic Risk Score Equals Better CAD Prediction

Given that polygenic risk doesn’t change and only needs to be measured once, researchers advocate greater use.

Combining a polygenic risk score (PRS) with traditional biomarkers can help clinicians better estimate a patient’s risk of developing coronary artery disease (CAD), according to new UK Biobank data.

When researchers added the CAD PRS to three more-conventional metrics—LDL cholesterol, high-sensitivity C-reactive protein (hs-CRP), and lipoprotein(a)—for more than 353,000 patients from the database, the model’s discrimination improved compared with using biomarkers alone (C-index 0.739 vs 0.754; P < 1 × 10-300). This relationship remained even when patients on lipid-lowering therapy were excluded from the analysis.

Senior author Arman Qamar, MD, MPH (NorthShore University HealthSystem/Endeavor Health, Evanston, IL), told TCTMD that the study, published online yesterday as a research letter in JACC, was sparked by another paper, published last year in the New England Journal of Medicine¸ showing that LDL cholesterol, hs-CRP, and Lp(a) levels are strongly linked to MI risk in middle-aged women.

Importantly, polygenic risk is consistent over a person’s lifetime. “Our polygenic risk has been there since the time when we were born,” he said. Biomarker numbers might “fluctuate, but the DNA remains completely stable.”

The PRS includes approximately 1 million genetic variants from genome-wide association data that reflect inherited risk for CAD. In the present study, the researchers, including first author, Sumeet A. Khetarpal, MD, PhD (Massachusetts General Hospital, Boston, MA), wanted to see if the addition of the PRS to biomarkers commonly used in risk-prediction calculators could better estimate risk of CAD than the risk factors alone.

They included data from 353,755 participants enrolled in the UK Biobank who were without CAD at baseline. Over a mean follow-up period of 11 years, 7.0% developed CAD—about two-thirds of these individuals were men (64.1%) and 93.7% were of European ancestry. Mean age was also higher among those with CAD compared to those without (60.9 vs 56.5 years).

When comparing patients in the highest quintile of each biomarker to those with lower levels (first through fourth quintiles combined), the risk of developing CAD was always greater: LDL cholesterol (adjusted HR 1.25; 95% CI 1.21-1.29), hs-CRP (adjusted HR 1.43; 95% CI 1.39-1.47), Lp(a) (adjusted HR 1.20; 95% CI 1.16-1.23), and CAD PRS (adjusted HR 1.78; 95% CI 1.73-1.83).

The risk of CAD was doubled among the 3,550 participants in the highest quintile for all three circulating biomarkers (adjusted HR 2.15; 95% CI 1.97-2.35). Similarly, the risk was almost quadrupled among the 1,168 patients with elevated biomarkers who were also in the highest quintile for CAD PRS (adjusted HR 3.71; 95% CI 3.24-4.25). The researchers observed no interaction by patient sex.

In another analysis substituting apolipoprotein B (apoB) for LDL cholesterol, the risk of CAD increased with each quintile and in the composite model for three biomarkers (adjusted HR 2.84; 95% CI 2.68-3.00) as well as when CAD PRS was added (adjusted HR 4.16; 95% CI 3.62-4.78). Adding CAD PRS to this model improved overall discrimination for predicting CAD (C-index 0.740 vs 0.754; P < 1 × 10-300).

“When we think about our personalized medicine approach, [we know] we all don’t look the same,” Qamar said. “That is how we should target our preventive therapies too.” Adding the CAD PRS to the established three-biomarker model “convert[s] it into a robust personalized medicine approach in which we have clinical risk factors and biomarkers and now genomics. And what we found is that that’s the best model.”

He stressed: “The bottom line is that people who have high polygenic risk score, they’re not doomed. . . . But knowing your DNA will change destiny.”

An advantage to this type of screening is that the PRS only needs to be measured once in a lifetime, he added. Qamar said it’s a misconception to think that management tailored to polygenic risk is “medicine of the future,” as his institution is one of the first in the United States to use it “very frequently” in risk assessment, although it’s not yet routinely covered by health insurers.

The test, which costs about $200-$250 according to Qamar, pays off especially in situations where younger patients are treated for STEMI and “people are wondering what happened and why did he have a heart attack,” he said. “I’m checking polygenic risk score in them and I’m seeing, oh my God, they had [high risk]. And I’m telling myself that I wish that someone had done that for them before then. This could have been prevented.”

Also, “our goal is to educate people about the role of genomics,” especially when it comes to genetic ties., Qamar concluded. “If someone had an MI and if they have a lot of family history, they could actually discuss with their family members about the value of PRS testing.”