Marston NA. Circulation. 2021 Feb 2;143(5):470-478.

Genetic risk scores illustrate a procedure of summating a person’s genetic propensity for a given phenotype, and have accumulated interest for their prospective to enhance risk prediction in many common disorders. An early attempts at with a GRS for ischemic stroke exhibited limited predictive ability, regardless of promise in other conditions, possibly because of the little number of stroke susceptibility loci recognised and the biologic heterogeneity of ischemic stroke. Thus, Marston NA, et al., conducted a study to analyse whether a GRS could recognise patients at greater risk for ischemic stroke following accounting for traditional clinical risk factors in five trials over the spectrum of cardiometabolic disorder. The level of risk was estimated and conferred by each genetic risk tertile, high genetic risk compared the magnitude of risk to well-established clinical risk factors, and analysed the GRS performance over key subgroups. Lastly, whether genetic risk classification could refine stroke risk prediction in patients with atrial fibrillation was explored, with particular attention to those with lower CHA2DS2-VASc scores in whom high genetic risk might notify the decision about beginning anticoagulation.

A genetic cohort evaluation was executed with pooling of individual patient-level data from five cardiovascular clinical trials: ENGAGE AF-TIMI 48, SOLID-TIMI 52, SAVOR-TIMI 53, PEGASUS-TIMI 54, and FOURIER. All patients were given written informed consent. A set of 32 SNPs correlated with ischemic stroke was used to estimate a GRS in each patient and recognise tertiles of genetic risk. Hazard ratios for ischemic stroke over genetic risk groups were estimated using a cox model, adjusted for clinical risk factors. 51,288 patients were eligible to be incorporated in this evaluation over the five trials studied. The average age of the study population was 65.9 years, and 28.3% were female. Most patients (81.7%) showed coronary or peripheral artery disease. Many patients showed traditional clinical risk factors for stroke, including hypertension (82.3%), hyperlipidemia (60.2%), and diabetes (41.9%). A smaller percentage of the group showed prior transient ischemic attack or stroke (14.0%), were current smokers (18.2%), showed atrial fibrillation (28.4%), or showed a history of congestive heart failure (29.1%). A total of 960 patients exhibited an ischemic stroke across a median follow up of 2.5 years.

The low genetic risk score group showed significantly lower Kaplan-Meier event rates for ischemic stroke at 3 years as compared to intermediate and high genetic risk groups (1.95% (n=272) vs. 2.24% (n=322) vs. 2.56% (n=366)) (Figure 1).

Figure 1: Kaplan-Meier event rates for ischemic stroke by tertile of genetic risk at 3 years

Between subgroups, 44,095 patients without prior stroke (adjusted HR of top tertile vs lowest tertile 1.27, 95% CI 1.04-1.53) with no clear predictive utility showed stronger performance of GRS as compared to the 7,193 patients with prior stroke (adjusted HR 1.06, 95% CI 0.81-1.41) (Figure 2).

Figure 2: Hazard ratios for top tertile of genetic risk score in patients with and without prior stroke. Hazard ratios are adjusted for age, sex, hypertension, hyperlipidemia, smoking, diabetes mellitus, atrial fibrillation, coronary artery disease, and congestive heart failure. In all groups, 95% confidence intervals are shown. GRS=genetic risk score, HR=hazard ratio.

Between atrial fibrillation group, patients with lower CHA2DS2-VASc of 2 scores showed substantially higher predictive ability of the GRS (p-trend=0.04), including a 4-fold increased risk of stroke and absolute risk equivalent to those with CHA2DS2-VASc of 3 (HR 3.97 (95% CI 1.04-15.2), Figure 3).

Figure 3: Absolute risk of ischemic stroke in anticoagulated patients with atrial fibrillation stratified by CHA2DS2-VASc score and genetic risk in ENGAGE AF-TIMI 48. Median follow up of 2.8 years

Thus, it was concluded that over five large clinical trials of patients with cardiometabolic disease, a 32-SNP GRS was a distinct, individualistic predictor of ischemic stroke. Patients without prior stroke and patients with little clinical risk factors showed highest predictive value of the GRS. Additionally, in patients with atrial fibrillation, the GRS identified patients with risk however lower CHA2DS2-VASc scores were comparable to the higher CHA2DS2-VASc scores.

GRS: Genetic risk score; SNPs: Single nucleotide polymorphisms; CHA2DS2-VASc: Congestive heart failure, hypertension, age ≥ 75 years, diabetes mellitus, stroke or transient ischemic attack, vascular disease, age 65 to 74 years, sex category; ENGAGE AF-TIMI 48: Global Study to Assess the Safety and Effectiveness of Edoxaban vs Standard Practice of Dosing With Warfarin in Patients With Atrial Fibrillation; SOLID-TIMI 52:  SAVOR-TIMI 53: Saxagliptin Assessment of Vascular Outcomes Recorded in Patients with Diabetes Mellitus–Thrombolysis in Myocardial Infarction trial; PEGASUS-TIMI 54: Prevention of Cardiovascular Events in Patients With Prior Heart Attack Using Ticagrelor Compared to Placebo on a Background of Aspirin-Thrombolysis in Myocardial Infarction 54; FOURIER: Further Cardiovascular Outcomes Research With PCSK9 Inhibition in Subjects With Elevated Risk