帕金森病分子分型和早期预测模式的初步建立与随访验证

Parkinson's disease(PD) is a common neurodegenerative disorder. The clinical manifestations of PD include motor symptoms such as tremor,bradykinesia and rigidity, as well as non-motor symptoms such as sleep disturbance,depression,costipation and olfactory dysfunction.Classic diagnostic protocol usually causes misdiagnosis and missed diagnosis, and most cases diagnosed by this procedure are at middle or terminal phase of the disease. The current therapy for PD can merely ameliorate symptoms and can not cure the disease. Moreover, there has been lacking of the individualized treatment for PD. In our previuos study, by a combined analysis of the most common PD-associated genes using the largest Chinese cohort consisting of PD patients and aged people, we have demonstrated the interactions among multiple PD risk genes. More importantly, we have constructed the first risk map of PD using the genetic information provided by specific polygenic combinations. Based on these findings,the current project aims to establish a polygenic combination-based model for molecular classification of PD by Logstic regression analysis.In addition, the project aims to establish a model for early prediction of PD based on a joint effort of a two-step genetic screening of multiple PD risk genes and quantitative evaluation of non-motor symptoms.The reliability and efficacy of both models will be tested by longitudinal follow-up studies. In sum, the project is an important attempt for translational research on PD.The results may have significant implications in the early diagnosis, treatment and individualized intervention for the disease.

帕金森病（PD）是一种常见的神经变性疾病，临床上表现为震颤、运动迟缓、强直等典型运动症状及失眠、抑郁、便秘、嗅觉障碍等非运动症状。经典PD诊断方法容易导致误诊和漏诊，且经确诊的患者已为中、晚期。现有的PD治疗方法只能改善症状，而不能从根本上治疗本病，而且缺乏个体化治疗的指导方案。我们在前期工作中，利用国内最大的帕金森病和老年队列样本，首次对帕金森病的多个风险基因进行联合分析，阐明了多风险基因之间的相互作用关系，并利用特定多基因组合方式，首次构建了PD的多基因风险图。本项目将在前期研究的基础上，采用Logistic回归分析法建立一种基于多基因组合的帕金森病分子分型模型。同时联合采用多基因、"二步法"预测和非运动症状评估方法，建立一套早期预测模式，并采用纵向随访方法验证上述模式的有效性和可靠性。该项目是PD转化研究领域的重要尝试，研究结果对于制定PD早期预测、治疗和个体化干预方案具有重要意义。

帕金森病分子分型和风险预测是目前帕金森病诊断和防治领域的重要问题。本课题组通过努力，完成了PD病例和社区老年队列的建立，完成了所有病例、对照和老年社区队列样品的LRRK2， SNCA, MAPT, GBA, BST1, PARK16基因常见位点的分子定型，并在部分病例-对照中筛查了BST1基因编码区突变。在基因定型的基础上，分析了LRRK2，GBA和非-LRRK2、非-GBA型PD患者的基因型-表型相关性。同时，初步建立了基于LRRK2，GBA, MAPT, SNCA, BST1，PARK16基因主效应和二维交互作用、年龄、性别的多因素Logistic回归分析模型，并分析其特异性和敏感性。采用“六基因-二步法”在500名社区老年人中初步筛选出16人（3.2%）携带了中-高风险基因型组合的个体，并采用NMSS-时间加权法初步完成了其非运动症状的评估。研究结果首次阐明了“LRRK2-型”“GBA-型”和“非LRRK2、非GBA-型”PD在临床表型上的差异，从而证明利用特定分子遗传学信息，可将PD分成若干个分子亚型，并为更加精细化的PD分子分型提供了重要依据和简化模式。在对BST1基因9个外显子的突变筛查中，证实无PD致病性突变或与PD关联的基因多态，提示BST1基因与PD的相关性可能存在于BST1基因的非编码区，这对进一步研究BST1参与PD的发病机制及风险预测具有重要意义。同时，揭示了PD的6个常见关联基因（LRRK2、GBA、SNCA、MAPT、 BST1、PARK16）之间的相互修饰作用关系，为多基因风险预测模式提供了直接证据，并筛选出多个可望用于风险预测的多基因组合体。