Integrating genetic risk into clinical risk assessment tool improves CVD risk prediction

In a population-based study, the relative genetic risk (determined by a polygenic risk score) and absolute clinical risk (determined by SCORE2) provided independent information of an individual’s total CVD risk.

This summary is based on the publication of Li L, Pang S, Starnecker FS, et al. - Integration of a polygenic score into guideline-recommended prediction of cardiovascular disease. Eur Heart J. 2024 Mar 29:ehae048 [Online ahead of print]. doi: 10.1093/eurheartj/ehae048

Introduction and methods

Background

Guideline-recommended tools to predict CVD risk, such as SCORE2, are based on established risk factors and are used to estimate the absolute clinical risk [1-3]. In contrast, polygenic risk scores (PRSs) report the relative genetic risk in relation to the average of a given population, but the ESC and ACC/AHA do no recommend the use of a PRS at this time [1,2]. It is unknown how a PRS is best integrated into guideline-recommended CVD risk assessment tools.

Aim of the study

The authors assessed (1) whether a factor measuring relative genetic risk (PRS-factor) can confer a constant relative risk across the spectrum of clinical risk and (2) whether application of the PRS-factor can increase the precision of risk prediction in a clinically meaningful way for individuals at intermediate CVD risk as estimated by clinical risk assessment tools.

Methods

The PRS for coronary artery disease (CAD) was calculated in 432,981 UK Biobank participants aged 40–69 years based on individual-level genotype data and various phenotype and health-related data. To estimate the PRS-factor, participants were separated equally into deciles (10 groups) based on the PRS distribution. Within each PRS decile, the relative genetic risks (odds ratios [OR]) for CVD (i.e., CAD or stroke)—referred to as the PRS-factor—were computed by comparing them with the reference group (i.e., combination of fifth and sixth deciles). The PRS-factor was compared in the entire cohort and subgroups representing the spectrum of clinical risk.

To calculate the SCORE2-predicted clinical risk, the UK Biobank phenotype data were mapped to the reference table of a low-risk region. To test for consistency in clinical risk assessment tools, the QRESEARCH risk estimator version 3 (QRISK3) was also calculated.

To examine the stability of the PRS-factor, the findings were replicated in the combined population of the Framingham Heart and Atherosclerosis Risk in Communities studies (n=10,757). Additionally, the authors constructed a new risk prediction model to estimate an individual’s total risk, by multiplying the absolute clinical risk estimated by SCORE2 or QRISK3 with the relative genetic risk measured as PRS-factor.

Main results

Stability of the PRS-factor

• The CVD prevalence in the genetic reference group was 6.12%, which was designated as OR=1.0. There was an exponential increase in the ORs for CVD with an increasing PRS group (r²=0.98; P=1.4e-7). The mean CVD ORs was 1.72 (95%CI: 1.67-1.78) in individuals from the highest PRS decile, and 0.63 (95%CI: 0.6-0.65) in individuals from the lowest PRS decile.
• The distribution of the PRS-factor—measured as mean CVD ORs—showed that within each PRS decile, there was little (nonsignificant) variation between the entire cohort and subgroups carrying traditional risk factors, including male sex, age ≥50 years, current smoking, obesity, diabetes mellitus, high cholesterol, and hypertension.
• The PRS-factor within each PRS decile was consistent across the 3 clinical risk categories of SCORE2 (i.e., low, intermediate, and high).
• Similar findings were observed when the QRISK3 was used, which demonstrated stability of the PRS-factor across the risk categories of clinical risk assessment tools.

Application of the PRS-factor

• The PRS and SCORE2 (or QRISK3) had no significant interactive effects on CVD risk, which suggested that the relative polygenic contribution was largely independent of the absolute clinical risk as estimated by the clinical risk assessment tools.
• With the newly developed risk prediction model “SCORE2 × PRS-factor = total risk”, a considerable number of UK Biobank participants was reclassified. For example, by adding the PRS-factor, 13,886 participants (9.55%) with an intermediate risk based on SCORE2 (n=145,337) were reclassified as “high risk,” thereby increasing the number of individuals considered to be at high CVD risk by 56.6%.
• The incidence of CVD events was 8.08% in individuals who were reclassified by addition of the PRS-factor from intermediate to high risk, which was ~2 times higher than the CVD incidence in those who remained at intermediate risk (4.08%).
• Replication in the combined population of the Framingham Heart and Atherosclerosis Risk in Communities studies showed similar results.

Conclusion

In this population-based study, the absolute clinical risk (as determined by a clinical risk score, such as SCORE2) and relative genetic risk (as determined by the PRS-factor) provided independent information of an individual’s total CVD risk. Applying the new multiplicative model “SCORE2 × PRS-factor = total risk” to UK Biobank participants with an intermediate CVD risk according to SCORE2 reclassified 10% of them as “high risk,” thereby increasing the number of individuals considered to be at high CVD risk by 57%. The authors conclude “the PRS-factor appears to have the most clinical relevance for individuals at intermediate clinical risk—but at high genetic risk (PRS-factor >1)—who may be considered for an intensified preventive treatment.”

Find this article online at Eur Heart J.

References

1. Visseren FLJ, Mach F, Smulders YM, Carballo D, Koskinas KC, Bäck M, et al. 2021 ESC guidelines on cardiovascular disease prevention in clinical practice developed by the task force for cardiovascular disease prevention in clinical practice with representatives of the European Society of Cardiology and 12 medical societies with the special contribution of the European Association of Preventive Cardiology (EAPC). Eur Heart J 2021;42: 3227–337. https://doi.org/10.1093/eurheartj/ehab484
2. Arnett DK, Blumenthal RS, Albert MA, Buroker AB, Goldberger ZD, Hahn EJ, et al. 2019 ACC/AHA guideline on the primary prevention of cardiovascular disease: a report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation 2019;140:e596–646. https://doi.org/10.1161/CIR.0000000000000678
3. Hippisley-Cox J, Coupland C, Brindle P. Development and validation of QRISK3 risk prediction algorithms to estimate future risk of cardiovascular disease: prospective cohort study. BMJ (Online) 2017;357:j2099. https://doi.org/10.1136/bmj.j2099