SCORE2 recalibrated to better estimate CVD risk in Asians

The updated cardiovascular disease (CVD) prediction algorithm SCORE2 can be used to estimate the 10-year CVD risk in individuals without diabetes or CVD in the Asia Pacific/Middle East region, according to a study presented at ESC 2023.

An algorithm tailored to European populations, SCORE2 underwent a systematic recalibration using nationally representative incidence and risk factor data and was subsequently given the name SCORE2-ASIA, which was specifically adjusted to the target population of individuals between 40 and 70 years of age in the Asia Pacific/Middle East region who were free of CVD or DM.

“Rather than direct estimates of total CVD incidence, CVD mortality data were converted with ‘multiplication factors’, [given that] CVD morbidity data are very heterogeneous across countries [and that] data quality in most countries are not sufficient for recalibration,” said one of the SCORE2-ASIA Writing group authors Dr Sofian Johar of Ripas Hospital, Brunei Darussalam.

“These estimates of CVD incidence were [then] used in combination with mean risk factor levels to recalibrate to every region,” Johar added.

The component countries of Asia Pacific/Middle East region were grouped based on WHO’s Global Health Estimates standardized mortality rates: low/moderate risk (≤150 CVD deaths per 100,000), high risk (150-300 CVD deaths per 100,000), and very high risk (>300 CVD deaths per 100,000).

For example, low/moderate-risk countries would include Australia, Japan, New Zealand, Singapore, South Korea, and Thailand, Johar noted. On the other hand, very-high-risk-countries would include Indonesia and the Philippines.

Multiplier data

The cohort used for the multiplier data in the creation of SCORE2-ASIA comprised more than 9 million participants recruited between 2004 and 2017 from low/moderate risk countries (Singapore and South Korea) and high-risk countries (Brunei, China, and Malaysia). These included 426,267 fatal and nonfatal CVD events that occurred over a median of between 5.6 and 15.1 years.

Applying to SCORE2-ASIA model, the data showed 10-year incidence rates of CVD fatal and nonfatal events for men and women across the region. Of note, the 65-to-70-year age group had more than a 60 percent 10-year risk of CVD events, Johar pointed out.

He noted that the SCORE2-ASIA coefficients, in general, were broadly comparable to the original SCORE2 coefficients derived from European data.

When validated in an external cohort of more than 4 million individuals with 150,233 CVD events, SCORE2-ASIA had C-statistics of 0.716 (95 percent confidence interval [CI], 0.676–0.757) overall, 0.726 (95 percent CI, 0.685–0.766) for low-risk populations, and 0.697 (95 percent CI, 0.586–0.808) for high-risk populations. Johar acknowledged that there was not enough data for the very-high-risk group. [Johar S, et al, ESC 2023]

SCORE2-ASIA performed well in the well-defined Singapore cohort. Plotting the observed 10-year risk against the predicted 10-year risk yielded a C-statistic of 0.725 (95 percent CI, 0.723–0.728), with some slight underestimation of the observed 10-year risk in the higher risk cohorts.

Johar stated that their data can be used to generate a risk chart for routine clinical practice.

Referring to the chart on his slide, he said: “For example, if you have a 65-year-old male who is a smoker, with a total cholesterol of 6.0 mmol/L and systolic blood pressure of 165 mm Hg, you will get an expected 22 percent 10-year CVD risk score.”

He added that they will be able to complete their work and share the finalized risk charts by May 2024.