The PROG-IMT project: Individual progression of carotid intima media thickness as a surrogate for vascular risk.

颈动脉内膜中层厚度(IMT)是早期动脉粥样硬化的超声生物标志物。单次IMT可预测未来的血管事件，如心肌梗死或中风。IMT进展来自至少两次超声扫描，通常被认为是血管风险的替代指标；它经常被用于随机对照试验和病理生理学研究。目前IMT进展对临床血管终点风险的替代作用的证据稀少且不一致。提交的项目旨在使用当前最先进的个人参与者数据(IPD)荟萃分析方法评估IMT进展与血管风险之间的关联，以及替代的标准。这一分析是在申请者领导的大型、有效和快速增长的国际合作中进行的。根据知情估计，全球约有200个符合我们荟萃分析条件的随机对照试验。其中53家已经加入了我们的合作，并分享了他们的数据集。我们的预测-基于六年的类似数据经验-显示在未来两年内将有大约100个额外的RCT合作。通过对PubMed、EMBASE和其他临床试验数据库的广泛搜索、对所有已识别论文的参考列表的手动搜索、PubMed对IMT综述文章的研究及其参考列表的手动搜索以及所有研究小组成员的个人联系，确定了优秀的试验。纳入标准是(I)使用介入臂和安慰剂或标准治疗臂的随机对照试验，(Ii)前瞻性纵向研究设计，(Iii)研究调脂剂、降压药或任何其他类型干预的研究，当研究这种干预类型的至少五个试验可用时，(Iv)明确和公开的纳入标准和招募策略，(V)至少两次超声检查，其中确定颈动脉IMT，以及(Vi)记录心肌梗塞、中风或死亡或其中几项的临床随访。获得的数据集保存在协调中心(法兰克福)，在那里进行协调性和可信性检查。统一的数据集被转移到统计中心(剑桥)。对于荟萃分析，将使用Daniels和Hughes方法，其中对替代标记物的干预效应(平均IMT进展的差异)和对临床终点的干预效应(Cox回归模型的对数风险比)之间的关联用双变量正态分布建模，并用贝叶斯模型进行拟合。IPD荟萃分析使我们能够考虑IMT进展的不精确性，以及IMT进展与心血管疾病原发病风险比的试验内相关性。

Carotid Intima Media Thickness (IMT) is an ultrasound biomarker of early atherosclerosis. Single-time IMT is predictive of future vascular events, such as myocardial infarction or stroke. IMT progression, derived from at least two ultrasound scans, is frequently perceived to be a surrogate of vascular risk; and it is often used in randomized controlled trials and in pathophysiological studies. Current evidence for the surrogacy of IMT progression for the risk of clinical vascular endpoints is sparse and inconsistent.The submitted project aims at assessing the association between IMT progression and vascular risk, and the criteria of surrogacy, with current state-of-the-art methods of individual participant data (IPD) meta-analysis. This analysis is organized within a large, effective and fast growing international collaboration, led by the applicant. Based on informed estimations, approximately 200 RCTs that are eligible for our meta-analysis exist worldwide. 53 of these already joined our collaboration, and shared their datasets. Our projections - based on six years' experience with similar data - show that approximately 100 additional RCTs will cooperate within the next two years.Eligible trials are identified with an extensive search of PubMed, EMBASE and other databases of clinical trials, handsearch of the reference lists of all identified papers, PubMed research of review articles on IMT and handsearch of their reference lists, and personal contacts of all study group members. Inclusion criteria are (i) RCT with interventional arm and placebo or standard treatment arm, (ii) prospective longitudinal study design, (iii) investigation of lipid-lowering agents, antihypertensive drugs, or any other type of intervention when at least five trials studying this intervention type are available with the necessary data, (iv) well-defined and disclosed inclusion criteria and recruitment strategy, (v) at least two ultrasound visits where carotid IMT was determined, and (vi) a clinical follow-up recording MI, stroke, or death or several of these.Acquired datasets are kept in the coordination centre (Frankfurt), where harmonization and plausibility checks are done. Uniform datasets are transferred to the statistics centre (Cambridge). The analyses are managed in close cooperation between Frankfurt and Cambridge, and interpreted together.For the meta-analysis, the Daniels & Hughes approach will be used, where the association between the intervention effect on the surrogate marker (difference in mean IMT progression) and the intervention effect on the clinical endpoint (log hazard ratio of a Cox regression model) is modeled with a bivariate normal distribution, and fitted with a Bayesian model. The IPD meta-analysis enables us to allow for the imprecision of IMT progression, and for the within-trial correlation of IMT progression and log hazard ratio of CVD.