Quantitative MRI-based metrics for multi-center stroke treatment assessment

描述（由申请人提供）：神经影像学在识别急性中风后组织恢复能力的标记方面有很大的希望，因此是加速评估新疗法的理想工具。在多中心试验中使用成像的一个限制因素是，这些技术主要是定性的，由于不同的成像协议和平台，这使得跨中心的数据聚合变得复杂。我们项目的目标是开发基于定量成像的工具，这些工具既可以在单个中心内以更高的可靠性使用，也可以在多个中心之间具有更高的可重复性。我们的建议侧重于通过(1)减少非生理因素的混杂影响和(2)改善对自然史患者和接受介入治疗的患者的生理变化的客观监测来提高对可抢救组织的识别。先前，我们已经证明，在人类和实验脑卒中模型中，结合多种急性MRI方法测量组织和灌注状态，比任何单一成像方式都能客观、更准确地预测组织梗死。这些算法具有重要的临床意义，可以减少传统中风试验中需要的样本量，以证明治疗效果。因此，将进行以下研究，以调查跨平台和跨中心的可重复性，以验证这些算法在前瞻性预测组织结果方面的适用性：(1)分析来自多个卒中中心的回顾性急性MRI数据集，以确定不同医院和不同平台的成像结果差异的来源；(2)分析前瞻性收集的成像数据，以评估这些算法评估卒中新疗法疗效的能力。公共卫生相关性：中风是美国死亡和发病的主要原因。本研究开发了预测中风患者在治疗前对不同治疗策略的反应的组织命运的工具。这种对患者梗死风险的定量评估可用于客观评价新疗法的疗效，从而加快急需的脑卒中新疗法的开发。

DESCRIPTION (provided by applicant): Neuroimaging holds much promise for identifying markers of tissue salvageability after acute stroke and is therefore an ideal tool for accelerating assessment of new therapies. A limiting factor in the use of imaging for multi-center trials is that these techniques remain primarily qualitative, which complicates the aggregation of data across centers due to differing imaging protocols and platforms. The goal of our project is to develop quantitative imaging-based tools that can both be used with greater reliability within a single center and with greater reproducibility across multiple centers. Our proposal focuses on improving identification of tissue that can be salvaged by (1) reducing confounding influences of non-physiological factors and (2) improving objective monitoring of physiological changes in both natural history patients and patients receiving interventional therapies. Previously, we have shown that combining multiple acute MRI methods for measuring tissue and perfusion status can objectively and more accurately predict tissue infarction in both human and experimental stroke models than any single imaging modality. These algorithms have important clinical implications for reducing the sample size traditionally needed in stroke trials to demonstrate a therapeutic effect. The following studies for investigating cross-platform and cross-center reproducibility will therefore be performed to validate applicability of these algorithms for prospectively predicting tissue outcome: (1) analysis of retrospective acute MRI data sets from multiple stroke centers to identify sources of variability in imaging findings across hospitals and across platforms, and (2) analysis of prospectively collected imaging data to evaluate these algorithms' ability to evaluate the efficacy of new stroke therapies. PUBLIC HEALTH RELEVANCE: Stroke is a leading cause of death and morbidity in the US. This study develops tools for predicting the fate of tissue in stroke patients in response to different treatment strategies prior to therapy administration. This quantitative assessment of patient's infarction risk can be used for the objective evaluation of efficacy of novel therapies, thereby speeding the development of much-needed new stroke therapies.