Despite differences in the underlying pathogenic mechanisms of asthma, the fundamental processes that drive lung biology are present across most, if not all, injury and repair responses. However, a major challenge is the identification of common patterns present in these conditions. Currently, asthma phenotypes have been defined in limited terms and imprecise clinical paradigms that rely on features such as allergic history, age of onset, lung function, and symptoms of "severity." These studies, including the Severe Asthma Research Program (SARP) characterization of asthma clusters, have been useful, but are driven by differences in demographic and physiologic variables, measures that are distal to many biologic aspects of the disease.4 In contrast to these approaches, integrative functional genomics has the potential to define asthma endophenotypes at a level reflective of true endophenotypes that are mechanistically important to the pathogenesis of asthma. To this end, our multidisciplinary research team has developed a molecular phenotyping protocol to evaluate transcriptomic asthma endophenotypes using genome-wide gene expression measured in the sputum and circulation of asthmatics and has identified 3 transcriptional endophenotypes of asthma (sputum TEA clusters). TEA cluster 1 has a low level of airway inflammation and the most reversible airway obstruction;TEA cluster 2 has a moderate amount of airway inflammation, reversible airflow obstruction, and high sputum IL-13 levels (Th2 cluster);and TEA cluster 3 has the highest level of airway inflammation, the least reversible airway obstruction (remodeled cluster), and high levels of sputum YKL-40 (a chitinase-like-protein we have shown to be associated with remodeling and severe asthma). In addition, using matched blood gene expression data from this cohort, we developed a predictive model using 69 genes that can determine an individual's sputum TEA cluster assignment with 85% accuracy. Taken together, these data demonstrate that transcriptomically-derived asthma endophenotypes are associated with airway inflammation, physiologic remodeling, and immunophenotype-associated cytokines. In this application, integrative functional genomics will be used to evaluate the stability of the TEA cluster model. A second independent cohort of asthma subjects will be studied longitudinally, and the innate and adaptive immune system responses associated with the TEA clusters will be determined. The generalizability of TEA clusters to other lung diseases will be evaluated to identify the fundamental expression networks associated with the TEA clusters. Ultimately, these studies will improve our understanding of asthma heterogeneity at the molecular level at the site of disease, and identify patients with similar modulation of gene networks. The results will generate new molecular diagnoses of asthma endophenotypes that can be used to sub-classify patients for pathogenetic and therapeutic studies using blood and sputum gene expression and identify novel targets for candidate gene studies.
Using an integrative functional genomics method to identify subgroups of patients, we will evaluate gene expression in sputum from asthmatics. These studies will improve our understanding of the molecular diversity of asthma and will identify patients with similar gene networks. The results will generate new molecular diagnoses of asthma that can then be used to sub-classify patients for future research and therapeutic studies.