Integrating a Polygenic Risk Score for Coronary Artery Disease as a Risk‐Enhancing Factor in the Pooled Cohort Equation: A Cost‐Effectiveness Analysis Study

Figures

• 

  Figure 1. Model structure. The Markov model structure used in this study is shown with a total of 18 health states. The initial cohort was distributed in 2 groups: high‐risk cohort and nonhigh‐risk cohort. We defined high risk as individuals in the top quintile of the polygenic risk score (PRS) for coronary artery disease (CAD‐PRS) distribution or having other risk‐enhancing factor (eg, family history), while the nonhigh‐risk group included individuals in the bottom 80% of the CAD‐PRS distribution without any risk‐enhancing factor. In the pooled cohort equation (PCE)+CAD‐PRS strategy, all of the high‐risk cohort was initiated on statin preventive therapy to reduce the risk of coronary artery disease (CAD) and stroke, while for the PCE‐alone strategy only a proportion of patients with other risk‐enhancing factors initiated statins. We accounted for statin side effects such as diabetes, myopathy, and hemorrhagic stroke and subsequent risk of ischemic stroke and CAD. In the PCE‐alone strategy, CAD‐PRS was not considered as a risk‐enhancing factor, so only those with other risk‐enhancing factors initiated statins. Health outcomes were not examined for the nonhigh‐risk cohort.

• 

  Figure 2. Life expectancy in the US general population compared with the event‐free cohort in the model. Projected life expectancy of the event‐free cohort in the model with the mean ages of 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, and 90 years. The projected years of survival were comparable to the life expectancy of Americans with the same age. 52

• 

  Figure 3. Life expectancy postmyocardial infarction. Life expectancy of the cohort after acute myocardial infarction (MI) estimated from the model compared with data from the literature. 53 , 54 We found the life expectancy generated by the model to be within 95% CIs of the life‐expectancy values from the literature.

• 

  Figure 4. The 5‐year incremental net monetary benefit of pooled cohort equation (PCE) alone vs PCE+polygenic risk score for coronary artery disease (CAD‐PRS), with PCE+CAD‐PRS preferred across all parameter value variations except when the annual risk of developing CAD was <0.006. CAD indicates coronary artery disease; EV, expected value; and PRS, polygenic risk score.

• 

  Figure 5. The 10‐year incremental net monetary benefit of pooled cohort equation (PCE) alone vs PCE+polygenic risk score for coronary artery disease (CAD‐PRS), with PCE+CAD‐PRS preferred across all parameter value variations. Figures 4 and 5 show the incremental net monetary benefit of pooled cohort equation (PCE) alone vs PCE+polygenic risk score for coronary artery disease (CAD‐PRS), with PCE+CAD‐PRS preferred across all parameter value variations except when the annual risk of developing CAD was <0.006 in the 5‐year time horizon.

• 

  Figure 6. Cost‐effectiveness acceptability curves. Results from the probabilistic sensitivity analysis indicating the probability of the pooled cohort equation (PCE)+polygenic risk score for coronary artery disease (CAD‐PRS) being cost‐effective at different willingness‐to‐pay (WTP) thresholds. Compared with PCE alone, PCE+CAD‐PRS is likely to be cost‐effective with a probability of 94% and 99% at $50 000 WTP threshold and 98% and 99% at $100 000 WTP threshold in 5‐ and 10‐year time horizons, respectively.

• 

  Figure 7. Expected value of perfect information (EVPI) at different willingness‐to‐pay (WTP) thresholds. Population EVPI of ≈$64.3 M and $3.53 M at a $50 000 WTP threshold in the 5‐ and 10‐year time horizons. EVPI decreased with increase in WTP thresholds suggesting high certainty in the cost‐effectiveness analysis results.