An update on genetic risk scores for coronary artery disease: are they useful for predicting disease risk and guiding clinical decisions?

Figures & data

Table 1. Open questions in the context of genetic risk scores (GRS) for coronary artery disease (CAD) and how these issues might be addressed.

Potential unanswered questions	Possible scientific approaches to address the issues
What is the appropriate number of SNPs to incorporate into a GRS and what metrics should we use to decide?	Prediction models with millions of SNPs should be compared to more parsimonious SNP sets with respect to their predictive performance.
How is a high GRS defined?	The relation of a GRS distribution to absolute disease risk for CAD should be addressed in cohort studies of incident disease, stratified by relevant subgroups.
Are there specific subgroups of the population where genetic screening for CAD should be performed?
What drives the CAD risk in individuals with a high GRS?	Prospective studies should explore the trajectories of risk factors according to GRS strata in order to better understand at what age the person develops clinically meaningful CAD risk threshold for the initiation of preventive measures; and to define other mechanisms that link GRS to risk of premature disease.
Are there significant statistical interactions of a GRS for CAD with lifestyle factors?	Potential interactions of GRS with age and with various lifestyle factors and in different ethnicities should be explored in longitudinal settings.
Is there an age-related penetrance of genetic variants associated with CAD?
Does the performance of GRS or do risk alleles differ by race/ethnicity, age or diabetes or HTN status?
What are the clinical consequences of having a high GRS? How to avoid ‘deterministic’ perspectives on the part of health care providers?	It should be addressed in randomized trials (RCT; including post-hoc analyses of RCTs with available genetic data) whether individuals with a high GRS benefit from preventive measures.
Does knowledge of a GRS change the clinical management of the patient (and his/her outcome)?	RCTs should address whether the disclosure of genetic information increases patient compliance or motivation. Psychological effects of knowledge of a high GRS must be assessed.
Does knowledge of a GRS change patient motivation for health-related behavior or compliance or provoke excess anxiety?
Is there a potential for genetic discrimination?	The potential for genetic discrimination should systematically be evaluated and legal options for mitigating such risks examined.
Who is responsible for communicating the results of a GRS and what is the impact of such communication on patients’ psychological state and general health?	Research should develop strategies for the regular evaluation of genetic variants and GRS and for communication with the patient when the categorization of SNPs changes. Assessment of impact on disclosure to patients (behavioral, economic, medical outcomes) warrants careful study.
How do we return genetic results to vulnerable populations?	Research into cultural sensitivity, access to educational tools and to health care resources for additional medical follow-up requires consideration.

SNP indicates single nucleotide polymorphism; HTN, hypertension.