Redefining Chiari Type I Malformation through Genetically, Radiologically, and Clinically-Derived Endophenotypes that are Predictive of Long-Term Neurological Outcome

项目摘要 项目4：使用预测长期结局的多维内表型重新定义CM 基亚里1型畸形（CM）是一种常见的和潜在的衰弱性疾病，通常定义为尾部畸形， 小脑扁桃体移位到枕骨大孔以下。影像学CM仅1% 或者更多的人，这个群体是非常异质的。经典定义的CM患者可能 临床上无症状或存在严重症状，如衰弱性头痛或脑干症状 或者脊髓功能障碍此外，具有相似临床表现的患者可能具有不同的遗传学特征。 特征和共病诊断表明导致表面相似疾病的不同病因。 对这种异质性的日益认识强调了数据驱动方法的必要性， 重新定义传统意义上的“CM”此外，当代成像方法的进步， 基因组学和以患者为中心的临床措施大大扩大了表型分析的机会 复杂的疾病，如CM。利用华盛顿大学在这些领域的独特优势， 项目4的目的是在传统的- 定义了“CM”，以告知临床管理并优化长期患者结局。特别是，我们预计 这些多维的内表型将有助于结果预测， 关于疾病发病机制的见解，从而支持个性化的治疗策略。我们打算 通过三个具体目标实现这一目标。在目标1中，我们将应用成熟的无监督和半监督 有监督的聚类方法，以多维临床和放射学数据，以获得新的集群， 大型多中心数据集。然后，我们将从外部验证这些多维内在表型在三个 外部多中心数据集。在目标2中，我们将比较各组之间疾病表现的差异， 还使用最先进的遗传学和先进的成像技术，以获得新的见解，生物 差异潜在CM异质性和发病机制。在目标3中，我们将研究 这些集群和重要的CM结果（例如，生活质量的改变、术后并发症）。我们将 然后使用预测建模来检查集群身份和其他特征如何影响这些 结果。最后，我们将使用一种改进的德尔菲方法来创建一个新的治疗框架的基础上， 项目的发现沿着来自CM研究和临床管理专家的结构化反馈。 总之，该项目的结果将为CM分类提供一种全新的数据驱动方法。我们 预计这些发现将对临床管理实践产生直接影响，并将指导未来 在这个领域的研究。

PROJECT SUMMARY Project 4: Redefining CM Using Multidimensional Endophenotypes Predictive of Long-Term Outcome Chiari type 1 malformation (CM) is a common and potentially debilitating disease typically defined by caudal displacement of the cerebellar tonsils below the foramen magnum. While radiographic CM may be found in 1% or more of the population, this population is extremely heterogeneous. Patients with classically defined CM may be clinically asymptomatic or present with severe symptoms such as debilitating headaches or signs of brainstem or spinal cord malfunction. Furthermore, patients with similar clinical presentations may have distinct genetic profiles and comorbid diagnoses that suggest differing etiologies contributing to superficially similar disease. Increasing recognition of this heterogeneity has emphasized the need for a data-driven approach to reclassifying traditionally-defined “CM.” Furthermore, advances in contemporary methods in imaging, genomics, and patient-centered clinical measures have substantially expanded opportunities for phenotyping complex diseases, such as CM. Leveraging Washington University’s unique strengths in these areas, the overall objective of Project 4 is to derive clinically relevant, multidimensional endophenotypes within traditionally- defined “CM” to inform clinical management and optimize long-term patient outcomes. In particular, we expect these multidimensional endophenotypes will both aid in outcome prediction and also provide new biological insights regarding disease pathogenesis, thereby supporting personalized treatment strategies. We intend to pursue this objective through three specific aims. In Aim 1, we will apply well-established unsupervised and semi- supervised clustering methods to multidimensional clinical and radiological data to derive novel clusters in a large, multicenter dataset. We will then externally validate those multidimensional endophenotypes in three external, multicenter datasets. In Aim 2 we will compare differences in disease presentation across clusters and also use state-of-the-art genetics and advanced imaging techniques to obtain new insights into biological differences underlying CM heterogeneity and pathogenesis. In Aim 3 we will examine the association between these clusters and important CM outcomes (e.g., change in quality of life, postoperative complications). We will then use prediction modeling to examine how both cluster identity and other characteristics impact these outcomes. Finally, we will use a Modified Delphi Approach to create a novel treatment framework based on this Project’s findings along with structured feedback from experts in the research and clinical management of CM. Taken together, this Project’s results will offer a fundamentally new data-driven approach for classifying CM. We expect these findings will have an immediate impact on clinical management practices and will guide future research in this field.