Sputum gene expression signature of 6 biomarkers discriminates asthma inflammatory phenotypes.

Background: Airway inflammation is associated with asthma exacerbation risk, treatment response, and disease mechanisms.

Objective: This study aimed to identify and validate a sputum gene expression signature that discriminates asthma inflammatory phenotypes.

Methods: An asthma phenotype biomarker discovery study generated gene expression profiles from induced sputum of 47 asthmatic patients. A clinical validation study (n = 59 asthmatic patients) confirmed differential expression of key genes. A 6-gene signature was identified and evaluated for reproducibility (n = 30 asthmatic patients and n = 20 control subjects) and prediction of inhaled corticosteroid (ICS) response (n = 71 asthmatic patients). Receiver operating characteristic curves were calculated, and area under the curve (AUC) values were reported.

Results: From 277 differentially expressed genes between asthma inflammatory phenotypes, we identified 23 genes that showed highly significant differential expression in both the discovery and validation populations. A signature of 6 genes, including Charcot-Leydon crystal protein (CLC); carboxypeptidase A3 (CPA3); deoxyribonuclease I-like 3 (DNASE1L3); IL-1β (IL1B); alkaline phosphatase, tissue-nonspecific isozyme (ALPL); and chemokine (C-X-C motif) receptor 2 (CXCR2), was reproducible and could significantly (P < .0001) discriminate eosinophilic asthma from other phenotypes, including patients with noneosinophilic asthma (AUC, 89.6%), paucigranulocytic asthma (AUC, 92.6%), or neutrophilic asthma (AUC, 91.4%) and healthy control subjects (AUC, 97.6%), as well as discriminating patients with neutrophilic asthma from those with paucigranulocytic asthma (AUC, 85.7%) and healthy control subjects (AUC, 90.8). The 6-gene signature predicted ICS response (>12% change in FEV1; AUC, 91.5%). ICS treatment reduced the expression of CLC, CPA3, and DNASE1L3 in patients with eosinophilic asthma.

Conclusions: A sputum gene expression signature of 6 biomarkers reproducibly and significantly discriminates inflammatory phenotypes of asthma and predicts ICS treatment response. This signature has the potential to become a useful diagnostic tool to assist in the clinical diagnosis and management of asthma.