Integrative risk modeling for early prediction of endometriosis and its long-term health outcomes

每年有超过200，000名女性被诊断患有子宫内膜异位症，其中一半以上的女性 直到症状发作后8.5年才得到明确的诊断，并且许多时候 它们存在额外的合并症。虽然一些研究表明，基因组标记， 环境危险因素和炎症标志物在子宫内膜异位症病理学中起关键作用， 目前还没有有效的工具来预测个体发生子宫内膜异位症的风险， 预测其下游影响。长期目标是开发有效的和非侵入性的早期 筛选工具，以确定患者在发展子宫内膜异位症的风险，并预测长期影响。的 该项目的主要目标是开发模型来预测子宫内膜异位症的风险， 不同的临床表现和相关的长期健康结果。我们的核心假设是， 综合风险模型将成功地识别出有发生子宫内膜异位症风险的患者， 与子宫内膜异位症（育龄期）同时发生或 子宫内膜异位症的发展（长期健康结果），使早期诊断和预防。这 一般假设将通过以下具体目标进行检验：（1）建立一个综合风险模型， 预测患者发生子宫内膜异位症的高风险;（2）开发一个综合风险模型， 预测子宫内膜异位症妇女临床表现的遗传和非遗传危险因素; (3)创建子宫内膜异位症的终身时间顺序图，以确定处于发展风险的个体 相关合并症。在aim 1中，我们将整合从以下数据中提取的遗传和非遗传风险因素： 以线性和非线性方式生成EndoRisk模型的电子健康记录。在目标2中，我们 将产生与子宫内膜异位症的各种临床表现相关的其他风险因素的目录 并针对不同的表现形式建立风险模型。在目标3中，我们将评估 子宫内膜异位症对相关合并症的影响，并建立伴随疾病的中介风险预测模型 健康状况和长期健康结果。在成功完成拟议的研究后， 预期的结果将严格评估非侵入性计算方法进行筛选， 诊断子宫内膜异位症的各种临床表现及其长期影响， 遗传和非遗传因素。这项研究是创新的，因为我们的新方法 对于综合风险模型将产生直接的转化影响。这些结果将提供一个 为未来制定改善患者结局的战略和将 通过为识别高、中、轻度患者提供支持， 子宫内膜异位症的风险，预计将对患有子宫内膜异位症的妇女产生重大影响。 子宫内膜异位症或其长期影响，根据其相对风险定制个性化治疗。

More than 200,000 women are diagnosed with endometriosis every year and over half of those women do not receive a definitive diagnosis until 8.5 years after the onset of symptoms and many times when they present with additional comorbidities. While several studies have suggested that genomic markers, environmental risk factors and inflammatory markers play crucial roles in endometriosis symptomatology, there are no effective tools available to predict an individual's risk of developing endometriosis or to predict its downstream effects. The long-term goal is to develop effective and non-invasive early screening tools to identify patients at risk of developing endometriosis and predict long-term effects. The main objective of this project is the development of models to predict the risk of endometriosis across varied clinical manifestations and associated long-term health outcomes. Our central hypothesis is that integrative risk models will successfully identify patients at risk of developing endometriosis and associated diseases that occur either concurrently with endometriosis (reproductive age) or after endometriosis development (long-term health outcomes), enabling early diagnosis and prevention. This general hypothesis will be tested via the following specific aims:(1) Develop an integrative risk model to predict patients at high risk of developing endometriosis; (2) Develop an integrative risk model combining genetic and nongenetic risk factors to predict clinical manifestations among women with endometriosis ; (3) Create a lifelong chronological map of endometriosis to identify individuals at risk of developing associated comorbidities. In aim 1, we will integrate genetic and non-genetic risk factors extracted from Electronic Health Records in linear and non-linear fashion to generate an EndoRisk model. In aim 2, we will generate a catalog of additional risk factors linked to various clinical manifestations of endometriosis and develop risk model for varied manifestations. In aim 3, we will evaluate mediating risk of endometriosis on associated comorbidities and develop a mediator risk prediction model for concomitant conditions and long-term health outcomes. At the successful completion of the proposed research, the expected outcomes will be rigorously evaluated non-invasive computational methods for screening and diagnosing endometriosis across various clinical manifestations and its long-term effects based on genetic and non-genetic factors. The proposed research is innovative because our novel methodology for integrated risk models will have immediate translational implications. These results will provide a strong basis for future development of strategies for improving patient outcomes and translating the knowledge to clinical practice by providing support for identifying patients at high, moderate, and mild risk of endometriosis, which is expected to have a significant impact on women suffering from endometriosis or its long-term effects by tailoring personalized treatments based on their relative risk.