Biomarker and Diagnostic Discovery for Inborn Errors of Metabolism

描述（由申请人提供）：在美国，由先天性代谢错误（IEM）引起的代谢性疾病的发病率估计高达1 / 3500婴儿，对未在新生儿中发现疾病的家庭的影响可能是毁灭性的。虽然新生儿筛查这类疾病的好处已得到证实，但技术挑战限制了其更广泛的应用。美国医学遗传学学院已经确定了两项可以显著改善新生儿筛查的具体挑战，即i)发现针对目前不存在的IEM疾病的新的生物标志物测试，ii)改进对目前假阳性率高的测试的生物标志物筛选。为了解决这两个挑战，我们建议利用目前可从高通量数据采集方法中获得的全范围代谢物测量，并使用我们专有的计算机代谢建模平台预测优于单一生物标志物筛选的生物标志物签名。新筛查的传统开发一直是数据驱动的，需要成千上万的患者数据点进行统计分析。这种自上而下的方法导致了前面提到的两个缺点。我们的计算平台提供基于机械的生物标志物计算，使用自下而上的基于途径的方法来重建人类细胞的全部代谢内容，然后确定IEM疾病的功能和生理影响。使用这种方法，我们可以直接计算人类生物体液中随给定IEM疾病而变化的多种候选代谢物生物标志物，并预测整个疾病的生物标志物特征。在我们的第一阶段工作中，我们开发了预测IEM疾病子集的生物标志物特征所需的计算模型和方法，并产生了非常有希望的结果（预测已知生物标志物收集的疾病集的准确率约为90%）。我们现在在第二阶段的努力中，i)扩展我们目前拥有的人类肝细胞代谢的计算机模型，以增加其在IEM疾病中的范围和应用，II)推进和验证生物标志物签名计算算法，以提高其准确性，iii.)为靶向IEM疾病生成新的生物标志物签名，并利用回顾性和前瞻性数据来确认新的生物标志物签名。然后，这些经过验证的生物标志物签名将通过与目前正在进行新生儿筛查的商业实验室和/或测量设备供应商的合作伙伴关系进行商业化。我们在代谢建模领域工作了十多年的科学家团队，以及我们的科学、临床和商业承包商，为诊断筛选成功生成新的生物标志物签名。该II期项目开发的生物标志物平台在癌症生物标志物的鉴定和验证（以及用于诊断、治疗选择和监测辅助的最终产品）、毒理学和安全性测试以及药物发现等领域也具有重要意义

DESCRIPTION (provided by applicant): The incidence in the United States of metabolic disease resulting from inborn errors of metabolism (IEM) is estimated to be up to 1 in 3500 infants, and the impact on families where diseases are undetected in newborns can be devastating. Although the benefits of newborn screening for such diseases has been demonstrated, technical challenges are limiting their broader application. Two specific challenges have been identified by the American College of Medical Genetics, that could significantly improve newborn screening, are i) the discovery of new biomarker tests for IEM diseases for which tests are currently nonexistent and ii) the improvement of biomarker screening for current tests that have high false-positive rates. To address these two challenges, we propose to leverage the full range of metabolite measurements that are currently available from high-throughput data acquisition methods and predict biomarker signatures that are superior to single biomarker screens using our proprietary computational in silico metabolic modeling platform. Classical development of new screens has been data-driven, requiring hundreds of thousands of patient data points for a statistical analysis. This top-down approach has led to the two shortcomings mentioned. Our computational platform offers a mechanistically-based calculation of biomarkers using a bottom-up pathway-based approach to reconstruct the full metabolic content of human cells and then determine the functional and physiological impacts of IEM diseases. Using this approach, we can directly calculate multiple candidate metabolite biomarkers in human biofluids that change with a given IEM disease and predict entire disease biomarker signatures. In our Phase I effort, we developed the computational models and methods needed to predict biomarker signatures for a subset of IEM diseases and produced extremely promising results (approximately 90% accuracy in predicting known biomarkers for the collected set of diseases). We now propose in a Phase II effort, to i) expand the in silico model we currently have of the human hepatocyte metabolism to increase its scope and application to IEM diseases, ii.) advance and validate the biomarker signature computational algorithm to increase its accuracy with focused enhancements, and iii.) generate new biomarker signatures for targeted IEM diseases and utilize retrospective and prospective data to confirm the new biomarker signatures. These validated biomarker signatures will then be commercialized through partnerships with commercial laboratories currently performing newborn screening and/or with vendors of the measurement equipment. Success in generating new biomarker signatures for diagnostic screens is supported by our team of scientists who have been working in the field of metabolic modeling for over a decade, as well as our scientific, clinical, and commercial contractors. The developed biomarker platform of this Phase II program also has significant implications in the areas of identification and validation of biomarkers for cancer (and resulting products for use as diagnostics, therapy selection, and monitoring aids), toxicology and safety testing, and drug discovery