Plasma proteomics for prediction of subclinical coronary artery calcifications in primary prevention
Authors' conclusion
A plasma protein subset that could predict the presence of subclinical CAC was identified but it did not outperform nor improve a model based on traditional risk factors.
Study context
- Population
- Individuals aged 50–64 years without history of ischaemic heart disease, with coronary artery calcifications (CAC) assessed by CT (Impaired Glucose Tolerance and Microbiota (IGTM) training cohort, SCAPIS validation cohort)IGTM 1,827
- Comparison
- Plasma protein subset vs traditional risk factors for predicting CAC
- Outcome
- Prediction of presence of CAC (AUC)
Methodological detailsmachine learning (extreme gradient boosting), external validation cohort, 1,342 plasma proteins measured by proximity extension assays, Models trained in IGTM, validated in SCAPIS
- Machine learning (extreme gradient boosting)
- External validation cohort
- 1,342 plasma proteins measured by proximity extension assays
- Models trained in IGTM, validated in SCAPIS
Performance
As reported in the abstract
Presence of CAC, validation cohort
| Test or model | AUC |
|---|---|
| Proteins | 0.691 |
| Traditional risk factors | 0.710 |
| Proteins + traditional risk factors | 0.705 |
AUC
Proteins0.691
Traditional risk factors0.710
Proteins + traditional risk factors0.705
- Proteins vs Traditional risk factors: P = 0.17
- Proteins + traditional risk factors vs Traditional risk factors: P = 0.6
Further findings
Correlation of the 44 proteins with traditional risk factors
- Most of the 44 proteins highly correlated with traditional risk factors
Original publication
Plasma proteomics for prediction of subclinical coronary artery calcifications in primary prevention. Am Heart J. Published 5 February 2024. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →