Department of Molecular and Clinical Medicine, Sahlgrenska Academy, Institute of Medicine, Gothenburg University, PO Box 100,405 30 Gothenburg, Sweden.
Department of Clinical Physiology, Region Västra Götaland, Sahlgrenska University Hospital, 413 45 Gothenburg, Sweden.
Eur J Prev Cardiol. 2024 Oct 10;31(14):1681-1689. doi: 10.1093/eurjpc/zwae124.
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).
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.
Large-scale plasma proteomics with data-driven and literature-based protein selection modestly improves prediction of future MACE beyond TRFs.
提高对心血管疾病高危人群的识别能力，将使针对性干预成为可能，并可能降低死亡率和发病率。我们的目的是确定大规模蛋白质组学的使用是否能提高对心血管事件的预测能力，超越传统风险因素（TRFs）。
在英国生物库的 38380 名参与者中，使用接近延伸测定法测量了 2919 种血浆蛋白。使用极端梯度提升机器学习进行数据和文献的特征选择和训练模型，用于预测 10 年随访期间主要心血管事件（MACE：致命和非致命性心肌梗死、中风和冠状动脉血运重建）的风险。曲线下面积（AUC）和净重新分类指数（NRI）用于评估选定蛋白质组学对 MACE 预测的附加价值，通过系统冠状动脉风险评估 2（SCORE2）或 SCORE2 中使用的 10 个 TRFs 进行评估。SCORE2 和 SCORE2 重新拟合到英国生物库数据预测 MACE 的 AUC 分别为 0.740 和 0.749。数据驱动的选择确定了与预测最相关的 114 种蛋白质。单独使用这些蛋白质并不能提高 MACE 的预测（AUC 为 0.758），但将这些蛋白质与 SCORE2 或 10 个 TRFs 结合使用可以显著提高（AUC = 0.771，P < 001，NRI = 0.140，AUC = 0.767，P = 0.03，NRI 0.053，分别）。基于文献的蛋白质选择（来自五个先前研究的 113 种蛋白质）也可以提高 TRFs 以外的风险预测，而随机选择的 114 种蛋白质则没有。
数据驱动和基于文献的蛋白质选择的大规模血浆蛋白质组学适度提高了对 TRFs 以外未来 MACE 的预测能力。
