IMPROVED MONITORING OF PREMANIFEST AND EARLY HUNTINGTONS USING 7T MULTIMODAL MRI

项目摘要 亨廷顿氏病（HD）是一种遗传性神经退行性疾病，潜伏期长，通常持续 直到运动、认知和精神症状明显影响功能能力， 然后随着时间的推移逐渐推进。发病年龄、症状严重程度和进展速度各不相同 在受影响的个人之间存在显著差异，并且迫切需要制定客观的指标来描述 疾病状态，以便适当地为患者提供咨询，设计临床试验，并评估 假定的治疗干预。现有的基于年龄和基因检测的预后标记缺乏 预测能力和疾病进展的指标使用临床和神经心理学评估， 灵敏度和再现性差。纹状体体积的纵向变化是确定的标志物 疾病进展，但未能捕获在疾病过程中观察到的患者异质性。 本研究的目的是增加对表现前和早期脑局部变化的敏感性。 基于定量磁敏感性标测（QSM）的症状性HD多对比度7T MRI，定量 结构MRI的形态测量学，以及扩散张量成像的脑连接分析，以开发 能够预测HD病程的空间变化、时间依赖性、多模态模型。 我们建议纵向研究患者使用解剖，灵敏度，和扩散加权 核磁共振使用纹状体体积和形状的测量，定量磁化率的区域值， 从7 T MRI检查中获得的特定束白色物质扩散，我们旨在检测和表征 神经元丢失、疾病相关铁沉积和白色物质的区域分布和时间进程 在这种混乱中受伤。高场7T MRI将用于增强微观组织的测量 敏感性，以及最近开发的纹状体形状分析和白色物质扩散技术 应用改变以增加对疾病进展的敏感性。将在45个个体中收集数据 45例有早期症状的HD患者，以及30例健康对照者。 目的1将通过以下方法评价多对比度7T MRI用于监测亚临床和早期HD进展的效果： 估计成像参数的区域横截面差异和患者内纵向变化。 目标2将确定在疾病负担的多变量模型中包括QSM是否会改善 通过生成患者水平的多变量模型来预测症状表现的准确性， 受影响脑区的体素级空间图，可区分每个疾病阶段，以及相关成像 疾病负担、认知功能和临床损害的度量指标。 这项拟议中的研究将最终导致对铁和蛋白质之间复杂关系的理解。 沉积和萎缩，增强预测接近临床症状发作的能力， 开发HD多变量风险预测模型的框架。

PROJECT ABSTRACT Huntington's disease (HD) is a genetic neurodegenerative disorder with a long latent period that usually lasts into early adulthood, until motor, cognitive and psychiatric symptoms overtly impact functional capacity and then gradually advance with time. The age of onset, severity of symptoms, and rate of progression vary significantly across affected individuals, and there is a strong need to develop objective metrics to characterize disease status in order to appropriately counsel patients, design clinical trials, and evaluate the efficacy of putative therapeutic interventions. Existing markers for prognosis based on age and genetic testing lack predictive power, and indicators of disease progression use clinical and neuropsychological evaluation that suffer from poor sensitivity and reproducibility. Longitudinal changes in striatal volume are established markers for disease progression, but fail to capture the heterogeneity seen during the disease course across patients. The goal of this study is to increase the sensitivity to regional brain changes in premanifest and early symptomatic HD multi-contrast 7T MRI based on quantitative susceptibility mapping (QSM), quantitative morphometry of structural MRI, and brain connectivity analysis with diffusion tensor imaging in order to develop spatially varying, time-dependent, multi-modal models capable of predicting disease course in HD. We propose to longitudinally study patients using anatomic, susceptibility-sensitive, and diffusion-weighted MRI. Using measurements of striatal volume and shape, regional values of quantitative susceptibility, and tract-specific white matter diffusion derived from 7T MRI examinations, we aim to detect and characterize the regional distribution and temporal course of neuronal loss, disease-related iron deposition, and white matter injury in this disorder. High-field 7T MRI will be used to enhance measurements of microscopic tissue susceptibility, and recently developed techniques for striatal shape analysis and white matter diffusion alterations are applied to increase the sensitivity to disease progression. Data will be collected in 45 individuals with premanifest HD and 45 individuals with early symptomatic HD, as well as 30 healthy controls. Aim 1 will evaluate multi-contrast 7T MRI for monitoring progression of subclinical and early HD by estimating regional cross-sectional differences and within-patient longitudinal changes in imaging parameters. Aim 2 will determine whether including QSM in a multivariate model of disease burden improves predictive accuracy of the manifestation of symptoms by generating a patient-level multivariate model and voxel-level spatial map of affected brain regions that discriminates each disease stage, and relating imaging metrics to measures of disease burden, cognitive function, and clinical impairment. The proposed study will ultimately result in an understanding of the complicated relationship between iron deposition and atrophy in HD, enhance the ability to predict proximity to clinical onset of symptoms, and create a framework for developing multivariate risk prediction models in HD.