Polygenic Risk Scores for Cardiovascular Disease

Publication Date: July 18, 2022
Last Updated: July 25, 2022

Key Conclusions

  • What Are PRSs? PRSs are single scores reflecting the cumulative weighted risk of individual genetic variation for a set of traits. These individual genetic variants confer an incrementally small disease risk, but summated, they have been shown to be predictive of many cardiovascular diseases.
  • Polygenic Versus Monogenic Risk Variants: Monogenic risk variants are typically single, protein-truncating variants conferring a relatively large risk of disease. Examples of monogenic risk variants for cardiovascular disease include LDLR for FH, GCK for diabetes, KCNQ1 for AF, and F5 for venous thromboembolic disease. PRSs independently associate with disease risk and, at particularly high scores, may yield similar estimated disease risk as monogenic risk variants.
  • Atrial Fibrillation: PRSs for AF have consistently shown incremental predictive capabilities in addition to clinical risk factors. Proposed utility has been to refine the identification of individuals meriting close surveillance for AF development.
  • Coronary Artery Disease: CAD is perhaps the most studied cardiovascular phenotype for PRSs. Among middle-aged adults, a CAD PRS performs similarly to conventional risk factors and provides additional prognostic information for CAD, but the clinical significance of this improvement is contentious. Observational evidence suggests that CAD PRSs may have utility in guiding pharmacological management (particularly for LDL-C lowering) attributable to increased estimated disease risk and greater estimated treatment benefit.
  • Hypercholesterolemia: LDL-C PRSs have been shown to be predictive of LDL-C concentrations,including severe hypercholesterolemia. Because an LDL-C PRS is predictive of ASCVD events independently of LDL-C, whether it should be used to help allocate novel LDL-C–lowering medicines (as FH variants currently allow) requires further study.
  • Type 2 Diabetes: Early research suggested that PRSs for T2D had a predictive ability similar to that of clinical risk factors. More recent evidence suggests that PRSs may be additive to clinical risk factors. However, the identification of those at high risk of T2D currently has unclear value; lifestyle modification and metformin treatment for T2D prevention did not appear to have different effects across genetic risk strata. Nevertheless, T2D PRSs may help guide T2D management through both sulfonylurea responsiveness and intensity of glucose management.
  • Venous Thromboembolic Disease: A VTE PRS is associated with incident venous thromboembolic disease (VTE) risk. Because the clinical utility of identifying inherited thrombophilias is unknown, the clinical utility of a VTE PRS also remains unknown. The benefits and risk of prolonged anticoagulation in those at high risk (determined using both clinical and genetic factors) require further study.
  • PRSs for Pharmacogenomics: Pharmacogenetic PRSs have addressed the following: In regard to PRSs for drug efficacy, most research has focused on statins with CAD PRSs, sulfonylureas with T2D PRSs, and PCSK9 inhibitors with CAD PRSs, whereas drug toxicity PRSs have focused on QTc prolongation with QTc PRSs. Ongoing disease-associated PRS analyses in completed clinical trials will continue to inform pharmacogenetics. Discovery genetic association analyses within clinical trials may enable novel pharmacogenomic PRSs.
  • Criteria for Implementing PRSs in Cardiovascular Clinical Practice: Estimates of incremental efficacy and harm and logistical challenges are key aspects for health care systems to consider when evaluating PRSs. Clinician and patient education on their interpretation and limitations may require additional infrastructure and personnel.
  • Calibrating PRSs to the Population of a Health Care System: Recalibration of PRSs to target populations may improve performance but may lead to challenges in transferability.
  • Considerations for Commercial Genetics Orga- nizations: Commercial genetics organizations should be aware of the current, but likely changing, regulatory approval process for LDTs. Currently, LDT regulation falls largely under CLIA/CAP regulation, but a series of decisions by the FDA over the past couple years seem to indicate that this may change in the future.
  • How to Consider the Financial Integration of PRSs Into Clinical Practice: There are a paucity of cost-effectiveness studies addressing cardiovascular PRSs. Decreasing genotyping costs and the negligible incremental direct costs of virtually limitless PRSs lead to modeling challenges.
  • Interpretation and Relevance of PRSs in the Context of Insurance Policies: The Genetic Information Nondiscrimination Act protects people in the United States from being discriminated against on the basis of their genetic information, specifically the provision of health insurance or employment hiring, firing, pay, or promotion. However, this protection does not extend to life and disability insurance.
  • Challenges and Future Directions: The addition of PRSs to clinical risk tools consistently enhances the predictive ability. However, significant but surmountable challenges exist. The lack of diversity of participant inclusion in biobanks, GWAS, and consequently PRSs is a major current limitation. The lack of inclusion of non-Europeans in PRS construction makes the use of PRS in non-European populations suboptimal. Recognizing this major limitation, large samples of non-Europeans are being enrolled in newer biobanks. For some populations, study of large numbers of non-European participants has already begun (ie, Japan Biobank, Million Veteran Program), but much work is still required. Further future work should address the inclusion of rare variants in scores, improvement in phenotyping in biobanks, and inclusion of PRSs within randomized controlled trials; future work on other cardiometabolic diseases also is needed.



Polygenic Risk Scores for Cardiovascular Disease

Authoring Organization