A polygenic risk score based on analysis of 29 discrete genetic variants linked with abdominal aortic aneurysms appeared better than the current criteria that clinicians use to identify people to screen for this disorder, potentially paving the way for more efficient use of screening resources.
Ryuji Suzuki
Dr. Derek Klarin
Future screening guidelines for abdominal aortic aneurysms (AAA) “should consider including individuals with high polygenic risk for screening ultrasonography,” Derek Klarin, MD, said at the virtual Vascular Discovery Scientific Sessions 2020, organized by the American Heart Association.
The data he reported showed that when researchers applied the polygenic risk score (PRS) to men aged older than 50 years in three independent validation cohorts of people with primarily European ancestry, those with scores in the top 5 percentile within each cohort had a collective AAA prevalence rate of 8.6% (95% CI 7.3%-9.8%).
This 8.6% pick-up rate using the PRS to help identify screening candidates for this male demographic subgroup compared favorably with previously reported prevalence rates of AAA detected by ultrasonography (defined as aneurysms of at least 3.0 cm in diameter) in men aged 65 years or older with a history of ever smoking. Last year, the U.S. Preventive Services Task Force issued an updated recommendation to perform a one-time ultrasound screening of 65- to 75-year-old men who ever smoked and cited five reported screening studies that found prevalence rates in these people of 3.3%-7.6% (JAMA. 2019 Dec 10;322[22]:2211-8). An earlier review of the topic by the task force cited an average estimated prevalence of 6%-7% in men at least 65 years old and with a smoking history (Ann Intern Med. 2014 Aug 19;161[4]: 281-90).
“You can use [the PRS] with other risk factors to increase the yield of identifying those at high risk,” Dr. Klarin said during a discussion of his report. He noted the possibility of using it to identify people at-risk early on, at birth, “prior to other risk factors being present,” as well as using the PRS as an add-on to known risk factors when assessing adults. He stressed that validations he has run so far still leave the PRS a step away from routine use, although it is “quite close,” said Dr. Klarin, a vascular surgeon at the University of Florida in Gainesville.
For use in routine practice, the PRS needs “further validation,” including further assessment of its performance in other age groups, in a wider range of ethnic groups, and in women, said Chris Semsarian, MBBS, professor of medicine at the University of Sydney and head of the Molecular Cardiology Program at its Centenary Institute. However, Dr. Semsarian also said that he saw great promise for the potential of the PRS, and its development so far had been solid.
“The study was meticulously undertaken, with a large number of AAA cases and controls. Both the derivation and validation are robust. There is great potential to use such a genetic risk score in the clinical setting, along with other risk factors such as smoking, high blood pressure, and lipid levels. The PRS adds another piece in the puzzle of risk of AAA by adding in genetic or inherited risk. An additional 1%-2% in pick-up rate would still lead to many thousands more AAA detected and lives saved. The PRS doesn’t replace environmental factors that contribute to AAA risk but adds a genetic component” when estimating a person’s overall risk and the appropriateness of screening ultrasound, Dr. Semsarian said in an interview.
The derivation analysis that Dr. Klarin and associates ran used data from the Million Veteran Program that included 7,642 people with AAA and matched them with more than 172,000 controls from the same database. This generated three alternative PRSs that involved testing for 29, 301, or 3,699 different mutations or polymorphisms that discriminated cases from controls. They compared these three scoring formulas in 1,000 AAA cases and 7,700 matched controls from the Mayo Clinic’s patient database, which showed that the 29-item PRS performed best, boosting identification of cases with a statistically significant odds ratio of 1.26.
They then ran the 29-item PRS in four additional large data banks, three that included mostly people of European ancestry and one that included mostly people with an African heritage. In the three data banks with people of mostly European background, the 29-item PRS performed even better than it did using the Mayo Clinic data, but this PRS was less informative in people with African ancestry. The analyses also suggested that the PRS identified elevated AAA risk independently of information on a family history of AAA, Dr. Klarin said.
The study had no commercial funding. Dr. Klarin has been a consultant to Regeneron.
SOURCE: Klarin D et al. Vascular Discovery 2020, abstract 170.