基于空间协方差分析建立的AD和SIVD脑灌注模式：生物标志物和机制研究

Although cerebral hypoperfusion is the common pathogenesis for Alzheimer’s disease (AD) and subcortical ischemic vascular dementia (SIVD), its clinical implication and pathophysiological actions and mechanisms are still unclear. Arterial spin labeling (ASL) with MRI has been investigated in patients with AD or SIVD for detecting regional changes in cerebral perfusion. However, the findings from previous studies are neither consistent nor appropriate for single-case evaluation in clinical practice. An age-related perfusion pattern measured with ASL has been obtained using a novel spatial covariance analysis (SCA) method of scaled subprofile model/principal component analysis (SSM/PCA) in our previous research. Compared with traditional univariate methods, the pattern derived from multivariate SCA is more consistent with altered large-scale network connection among functional brain regions inherent in diseased states. We further quantified its expression in individual subject by calculating the subject score of this pattern, which correlated significantly with age and could robustly distinguish younger group and older group. In this study, we plan to establish AD-related and SIVD-related perfusion patterns measured with ASL using the method of SCA in a prospectively enrolled patient cohort with both FDG and PiB PET assessments. Our primary aims are (1) to evaluate the use of these perfusion patterns for early diagnosis, differential diagnosis, and monitoring of disease progression in AD and SIVD; (2) to compare the association between perfusion pattern, structural pattern, metabolic pattern, white matter lesions, and amyloid deposition. Our secondary aim is to explore the feasibility of identifying perfusion, structural and metabolic patterns associated with deficits in different cognitive domains in each of these disease conditions. The findings of this research will facilitate the application of perfusion patterns measured with ASL in clinical diagnosis and management of AD and SIVD, reveal pathophysiological actions and mechanisms underlying perfusion changes, and shed light on neurobiological basis of functional and anatomical brain network correlates of cognitive dysfunction.

尽管脑灌注减低是阿尔茨海默病（AD）和皮层下缺血性血管性痴呆（SIVD）重要的共同发病机制，其临床意义及病理生理作用和机制并不明确。以往动脉自旋标记（ASL）MRI的AD和SIVD脑灌注研究结果不一致，也不便用于临床个体评价。申请人前期通过多因素的空间协方差分析成功获得年龄相关的ASL脑灌注模式，比传统方法更符合脑功能的网络机制；并进一步通过计算个体表达值对单个个体进行定量分析，可有效区分年轻组和年老组。本课题将以双探针PET辅助诊断的队列为研究对象，建立AD和SIVD相关的脑灌注模式，评价其对疾病早期诊断、鉴别诊断和预测进展的价值，比较其与脑结构和脑代谢模式、白质病变和淀粉样蛋白沉积的关系。同时，探索与不同认知领域相关的脑结构、脑灌注和脑代谢模式。本研究将为ASL检测的脑灌注模式在AD和SIVD诊治中的临床应用提供依据，并揭示脑灌注改变的病理生理作用和机制，以及认知障碍的神经网络基础。

阿尔茨海默病（AD）和血管性痴呆是造成老年人认知损伤的最常见原因。其中，皮层下缺血性血管性痴呆（SIVD）是血管性痴呆中最为常见的一个临床亚型，早期临床表现与AD相似。探寻反映病理生理改变的可靠生物标志物，将有助于两种疾病的早期诊断和鉴别诊断，为制定疾病特异性的治疗策略创造时机。脑血流量改变是AD和SIVD重要的共同发病机制，但其临床意义及病理生理作用和机制并不明确。本研究以双探针PET生物标志物诊断的AD和SIVD人群作为研究对象，应用多因素的空间协方差分析方法确立基于动脉自旋标记（ASL）MRI的疾病相关脑灌注模式，明确该模式对AD和SIVD早期诊断的价值和临床相关性，进一步探索疾病相关脑灌注改变的病理生理作用和机制。本研究首先揭示了AD和SIVD各自的认知损伤特征，建立了以情景记忆、信息加工速度、语言和视空间功能为主的神经心理学评估工具，对AD和SIVD具有最佳的鉴别价值。其次，本研究发现了AD和SIVD特征性的局部脑血流量变化，以多变量的SSM/PCA方法分别构建了AD和SIVD相关的ASL脑灌注模式，并验证了其诊断性能和临床相关性。AD相关脑灌注模式表现为双侧中、后扣带回和楔前叶、顶下小叶和额叶等区域的相对负荷减低，右侧小脑和双侧基底节等区域的相对负荷增加；该模式的个体表达值具有较高的诊断准确性，曲线下面积（AUC）在创建队列和验证队列分别为0.87和0.85，且与AD患者的总体认知功能、注意力和信息加工速度、执行功能、语言功能、视空间功能和记忆力均呈显著负相关。SIVD相关脑灌注模式的特点是额叶、顶叶和颞叶脑区的相对负荷减低；其个体表达值对SIVD诊断的AUC在创建队列和验证队列分别为0.96和0.89，并与SIVD患者的全脑血流量呈负相关，与白质高信号体积呈正相关，而与不同认知领域的评分均无显著相关性。本研究证实以空间协方差方法构建的脑灌注模式是一种有前途的MRI生物标志物，有希望作为AD和SIVD早期诊断和病情监测的工具。