Using signatures of T-helper cell inflammation to phenotype human asthma

描述(申请人提供)：哮喘是一种以呼吸道炎症、可逆性呼吸道阻塞和呼吸道高反应性为特征的常见病。人们越来越多地认识到哮喘的表型异质性，包括严重程度和对药物的反应。这些差异很可能是由不同的潜在分子表型驱动的，这些表型的发现将为靶向治疗的发展提供信息。在先前使用具有良好特征的哮喘患者的研究中，人类呼吸道上皮细胞基因表达的差异确定了两组主要患者，一组患者的基因表达强烈受炎性T辅助细胞(Th2)细胞因子白介素13(IL-13)驱动，被称为Th2“高”哮喘；另一组患者IL-13驱动的基因表达不高于健康对照组的水平，即“Th2低”哮喘。这项研究表明，与Th2水平低的哮喘患者相比，Th2水平高的受试者对吸入皮质类固醇的反应更好。根据已发表的数据和我们自己的初步数据，我们假设Th1和Th17炎症途径定义了人类哮喘的其他分子表型。我们的目标是验证这一假设，将表型与临床特征(包括对皮质类固醇的反应)联系起来，并开发基于Th1、Th2和Th17信号的人类受试者分子表型诊断工具。Th1、Th2和Th17的签名将从基于微阵列的全基因组表达-细胞因子刺激的上皮细胞系的图谱中开发。这些特征将被用来根据获得的哮喘患者自身呼吸道上皮细胞的基因表达谱，将研究对象划分为T辅助细胞亚群表型。最具信息量的上皮细胞基因将被用来开发一种基于PCR的测试。通过使用机器学习算法，将开发一种针对这些分子表型的非侵入性测试，以在受试者外周血样本的整个基因组表达谱中找到模式。分子表型的识别，以及可靠的非侵入性分配Th2、Th1或Th17表型的测试的可用性，将提供预测哪种药物对个人最有效的可能性，同时避免无效治疗的不必要副作用，并通过在临床试验期间提供生物标记物，为针对特定表型的治疗的开发提供信息。 与公共卫生相关：哮喘是一种非常常见的肺部疾病，在美国有7%的成年人受到影响，尽管治疗费用很高，但多达30%的患者对药物没有反应。哮喘有多种形式，有不同的诱因、长期结果和对药物的反应。这项工作的目标是1)应用人类基因组学的技术进步来发现不同类别的哮喘，2)开发基于这些类别的测试来预测哪些人会对特定的治疗产生反应，从而导致更有效的护理和更少的无效药物的副作用。

DESCRIPTION (provided by applicant): Asthma is a common disease characterized by airway inflammation, reversible airway obstruction, and airway hyperresponsiveness. There is growing recognition of phenotypic heterogeneity in asthma, including in severity and response to medication. It is likely that these differences are driven by distinct underlying molecular phenotypes, the discovery of which would inform the development of targeted therapies. In prior studies using well-characterized patients with asthma, differences in gene expression from human airway epithelial cells identified two major groups of patients, one in which gene expression is strongly driven by the inflammatory the T-helper cell (Th2) cytokine interleukin-13 (IL-13), which has been called Th2 "high" asthma, and another in which IL-13 driven genes are not significantly expressed above that level found in healthy controls, or "Th2 low" asthma. This work demonstrated that compared to Th2-low asthmatics, Th2-high subjects a better response to inhaled corticosteroids. Based on published data and our own preliminary data, we hypothesize that Th1 and Th17 pathways of inflammation define additional molecular phenotypes of human asthma. Our aims are to test this hypothesis, associate phenotypes to clinical characteristics including response to corticosteroids, and develop diagnostic tools for molecular phenotyping of human subjects based on Th1, Th2 and Th17 signatures. Th1, Th2, and Th17 signatures will be developed from microarray- based whole genome expression-profiling of cytokine-stimulated epithelial cell lines. These signatures will be used to cluster research asthma subjects into T helper subset phenotypes based on the gene expression profiles of their own airway epithelial cells obtained. The most informative epithelial cell genes will be used to develop a PCR-based test. A non-invasive test for these molecular phenotypes will be developed through the use of machine learning algorithms to find patterns in the whole genome expression profiles of subjects' peripheral blood samples. The identification of molecular phenotypes, and the availability of reliable tests to assign Th2, Th1, or Th17 phenotypes non-invasively would provide the potential to predict which medication will work best for an individual while avoiding unnecessary side effects form ineffective therapies, and to inform the development of therapies targeted to specific phenotypes by providing biomarkers during clinical trials. PUBLIC HEALTH RELEVANCE: Asthma is a very common lung disease affecting 7% of adults in America, and despite high treatment costs, up to 30% of patients do not respond to medications. Asthma has many forms with varying triggers, long-term outcomes and responses to medications. The goals of this work are to 1) apply technological advances in human genomics to discover distinct classes of asthma, and 2) develop tests based on these classes to predict which individuals will respond to a specific therapy, leading to more efficient care and less side effects from ineffective medicines.