Introduction and methods

Integrating The Effect Of Polygenic Scores, Low Density Lipoproteins And Systolic Blood Pressure On The Lifetime Risk Of Cardiovascular Disease

Presented at ACC.20 by Brian A Ference (Cambridge, UK)

There is a great interest in polygenic scores (PGS) to predict CVD. However, recent studies suggest that adding PGS to short term risk estimation equation may not provide meaningful information. PGS have may potential clinical value in that they predict lifetime risk early in life, before the onset of disease and identify persons with high lifetime CV risk, who will benefit from early intervention. Recent evidence suggest that effects of high LDL-c levels and blood pressure (BP) accumulate over time. Use of PGS may identify those with high lifetime CV risk and these high-risk patients may benefit from early intervention by minimizing lifetime exposure to high LDL-c and BP. This personalized strategy of screening and treatment decisions could have an important clinical impact.

This study examined lifetime CV risk at all levels of a PGS for coronary artery disease (CAD) depending on difference in lifetime exposure to LDL-c and systolic BP (SBP) using Mendelian randomization in order to make inferences about how a PGS for CAD can be combined with information on LDL-c and SBP. 445,566 Participants in the UK Biobank were enrolled in this study (54% female, mean age: 57.2 years). Primary outcome was major coronary events, a composite of first occurrence of fatal or non-fatal myocardial infarction or coronary vascularization. Using all participants in the UK Biobank, a polygenic score with 4,051,820 variants was deconstructed into an LDL score and an SBP score from Mendelian randomization, and a residual PGS representing all other pathways (variants significantly associated with LDL-c or SBP were excluded from constructing PGS).

Main results

• Quintiles for PGS showed a stepwise increase in risk of major coronary events (quintile 5: HR 2.77, 95%CI:2.65-2.89, with quintile 1 as reference).
• Risk of major coronary events was higher in those with highest PGS across all ages.
• Trajectories of CV risk in quintiles of PGS varied substantially by lifetime exposure to LDL-c and SBP. For example, those in the middle PGS quintile with a 10 mg/dL increase in LDL-c and 2 mmHg increase in BP had a similar CV risk as those in the highest PGS quintile. In opposite direction, those in the middle PGS quintile with a 10 mg/dL decrease in LDL-c and 2 mmHg decrease in BP had a similar CV risk as those in the lowest PGS quintile.
• CV risk appeared to be determined by the combination of PGS and lifetime exposure to LDL-c and BP. Across all quintiles of PGS, CV risk remained low with low exposure to LDL-c and BP. Lifetime CV risk was 2-3 fold higher in those in the lowest PGS and high exposure to LDL-c and BP compared to those in the highest PGS and low exposure to LDL-c and BP (P<0.001).
• Those in the highest PGS did not per se benefit the most from treatment of LDL-c and BP lowering. Participants with highest LDL-c and BP levels had highest expected benefit with lowering LDL-c and SBP across the range of PGS quintiles.

Conclusion

Combining PGS for CAD with information on lifetime exposure to LDL-c and SBP more accurately estimates lifetime risk of CVD, more accurately identifies subjects who benefit from early intervention, and better estimates potential benefit from early interventions. This strategy may help in personalizing CV risk and prevention.

- Our coverage of ACC.20 is based on the information provided during the congress –