Multimodal longitudinal imaging of brain and cervical cord as an ALS disease biomarker using microstructure statistics and morphometry

项目摘要 磁共振成像（MRI）、生物物理建模和计算方面的最新发展 提高了成像指标检测中枢神经系统疾病相关变化的灵敏度， 神经系统疾病这种提高的灵敏度为利用这些指标作为潜在的 本发明还涉及疾病的生物标志物，特别是用于测量短持续时间内的疾病进展。我们假设 来自脑和脊柱的MRI生物标志物（微观结构和形态学）的多模式分析将得到改善 在短至3至6个月的持续时间内检测疾病相关变化的灵敏度。我们的假设是基于 我们先前的工作是在ALS队列中检测6个月内大脑微观结构的纵向变化， 在这段时间内，功能指标发生了适度的变化，并且结合大脑和脊柱的多模式分析 MRI检查可提高疾病诊断的准确性。在这个项目中，我们将建立可扩展性，灵敏度， 在较短的持续时间内，通过两个地点的研究，这些指标的总体临床试验准备情况。我们也 建议通过组合来自 脑和脊柱的纵向多模态统计框架。此外，我们将展示这些 成像指标与液体生物标志物和功能进展指标相关。 我们将从40名ALS参与者和10名对照者中获取结构（T1和T2）和弥散MRI数据 参加者在两个地点：明尼苏达大学（主办机构）和佛罗里达大学。我们将扫描 基线和3次随访访视（3、6和12个月）时参与者的大脑和颈椎。我们将 在入组时完成神经系统检查、ALSFRS-R和UMN评分，并获得纵向 ALSFRS-R和血浆神经丝光（NfL）测量。我们将提取微结构和 使用专用的计算方法和建模从MRI数据获得形态信息。我们还将 应用新颖的统计工具来将这些互补的成像度量联合收割机组合成多模态分析。最后， 我们将分析NfL、ALSFRS-R变化和多模态MRI指标之间的相关性。 在我们的项目完成后，我们预计，我们提出的纵向MRI的增强灵敏度 生物标志物将通过提供新的替代标志物作为潜在结果而对ALS治疗产生影响 临床试验的措施。预期增加的效应量也将减少进行研究所需的队列规模。 试验，从而提高其可行性。在临床试验范围之外，我们的多模式MRI生物标志物将 作为单个患者水平上运动神经元变性的客观量度。我们的核磁共振测量 还将与液体生物标志物交叉验证，并将有助于将ALS患者分层为临床 同质性队列。 1

PROJECT SUMMARY Recent developments in Magnetic Resonance Imaging (MRI), biophysical modeling, and computing have improved the sensitivity of imaging metrics to detect disease-related changes in the central nervous system in neurological disorders. This improved sensitivity has paved the way for utilizing these metrics as potential biomarkers of disease, in particular, to measure disease progression over short durations. We hypothesize that the multimodal analysis of MRI biomarkers (microstructure and morphology) from the brain and spine will improve sensitivity to detect disease-related changes over durations as short as 3 to 6 months. Our hypothesis is based on our prior work detecting longitudinal changes in brain microstructure over 6 months in an ALS cohort with modest change in functional measures over that time, and that a multimodal analysis combining brain and spine MRI measures can improve disease diagnosis accuracy. In this project, we will establish the scalability, sensitivity over shorter durations, and overall clinical trial readiness of these metrics through a two-site study. We also propose to improve the sensitivity of imaging metrics by combining multiple complementary measures from the brain and spine in a longitudinal multimodal statistical framework. Additionally, we will demonstrate how these imaging metrics correlate with fluid biomarkers and functional progression measures. We will acquire structural (T1 and T2) and diffusion MRI data from 40 participants with ALS and 10 control participants at two sites: the University of Minnesota (host institution) and the University of Florida. We will scan the brain and cervical spine of participants at baseline and 3 follow-up visits (3, 6 and 12 months). We will complete a neurological examination, ALSFRS-R, and UMN score at enrollment and obtain longitudinal ALSFRS-R and plasma neurofilament light (NfL) measurements. We will extract microstructural and morphological information from MRI data using dedicated computational methods and modeling. We will also apply novel statistical tools to combine those complementary imaging metrics into a multimodal analysis. Finally, we will analyze correlations between NfL, change in ALSFRS-R, and multimodal MRI metrics. Upon completion of our project, we anticipate that the enhanced sensitivity of our proposed longitudinal MRI biomarkers will have an impact on ALS treatment by providing novel surrogate markers as potential outcome measures for clinical trials. The expected increased effect size will also reduce the cohort size needed to conduct trials, thereby increasing their feasibility. Beyond the scope of clinical trials, our multimodal MRI biomarkers will serve as an objective measure of upper motor neuron degeneration at the single patient level. Our MRI measures will also be cross validated with fluid biomarkers and will contribute to efforts to stratify ALS patients into clinically homogeneous cohorts. 1