Abstract Current approaches to estimate cardiovascular risk for primary prevention of cardiovascular disease (CVD) are suboptimal and novel strategies that optimise and personalise risk predictions are needed. In this review, we provide an overview of the potential use of circulating cardiovascular biomarkers - C-reactive protein, growth differentiation factor 15, N-terminal pro B-type natriuretic peptide, and cardiac troponins – to improve CVD risk prediction in the general population. We present novel statistical approaches that dynamically capture risk to further optimise and personalise cardiovascular risk estimation systems. While all studied cardiovascular biomarkers are independently associated with CVD risk when using measurements from a single time-point, the added value they provide to traditional risk factors in terms of accuracy is modest. A dynamic risk prediction approach - such as joint modelling - that can incorporate changes in biomarkers and other risk factors over time may better capture the complexities of CVD development, enhancing CVD risk estimates. With the increasing adoption of electronic health records and commercially available assays for these circulating biomarkers, deploying such systems in clinical practice has become feasible. Future research should focus on the development and validation of multi-marker biomarker-driven dynamic risk prediction models, along with development of a standardized methodological approach for comparing performance of dynamic risk estimation systems against current static models. To facilitate its implementation in clinical practice, public health impact and cost-effectiveness needs to be assessed. Finally, optimizing screening intervals through a biomarker-driven dynamic risk approach may offer further improvement in individualising primary CVD prevention strategies.
Bakker et al. (Thu,) studied this question.