Large-scale plasma proteomics in the UK Biobank modestly improves prediction of major cardiovascular events in a population without previous cardiovascular disease

Aims Improved identification of individuals at high risk of developing cardiovascular disease would enable targeted interventions and potentially lead to reductions in mortality and morbidity. Our aim was to determine whether use of large-scale proteomics improves prediction of cardiovascular events beyond traditional risk factors (TRFs). Methods and results Using proximity extension assays, 2919 plasma proteins were measured in 38 380 participants of the UK Biobank. Both data- and literature-based feature selection and trained models using extreme gradient boosting machine learning were used to predict risk of major cardiovascular events (MACEs: fatal and non-fatal myocardial infarction, stroke, and coronary artery revascularization) during a 10-year follow-up. Area under the curve (AUC) and net reclassification index (NRI) were used to evaluate the additive value of selected protein panels to MACE prediction by Systematic COronary Risk Evaluation 2 (SCORE2) or the 10 TRFs used in SCORE2. SCORE2 and SCORE2 refitted to UK Biobank data predicted MACE with AUCs of 0.740 and 0.749, respectively. Data-driven selection identified 114 proteins of greatest relevance for prediction. Prediction of MACE was not improved by using these proteins alone (AUC of 0.758) but was significantly improved by combining these proteins with SCORE2 or the 10 TRFs (AUC = 0.771, P < 001, NRI = 0.140, and AUC = 0.767, P = 0.03, NRI 0.053, respectively). Literature-based protein selection (113 proteins from five previous studies) also improved risk prediction beyond TRFs while a random selection of 114 proteins did not. Conclusion Large-scale plasma proteomics with data-driven and literature-based protein selection modestly improves prediction of future MACE beyond TRFs. Lay summary The risk of having a myocardial infarction or stroke is usually assessed by clinical scores including traditional risk factors for cardiovascular disease. The development of new technologies enables the rapid measurement of an increasing number of blood proteins. In this study, we applied machine learning techniques in a UK-based cohort of 38 380 participants with 2919 blood proteins measured. We obtained a set of 114 proteins that improved the prediction of the 10-year risk of major cardiovascular event when added to traditional risk factors. Improvements were also achieved using a set of 113 proteins found in previous studies. However, the magnitude of these improvements was relatively low and the clinical utility of combining these proteins with traditional risk factors in primary prevention will have to be further investigated. [GRAPHICS] .