Construction and validation of a bronchoalveolar lavage cell-associated gene signature for prognosis prediction in idiopathic pulmonary fibrosis

Yuechong Xia,Cheng Lei,Danhui Yang,Hong Luo
DOI: https://doi.org/10.1016/j.intimp.2021.107369
IF: 5.714
2021-03-01
International Immunopharmacology
Abstract:<h3 class="u-h4 u-margin-m-top u-margin-xs-bottom">Background</h3><p>Idiopathic pulmonary fibrosis (IPF) is a chronic and progressive interstitial lung disease. It is urgent to identify biomarkers to precisely predict mortality. <em>Methods:</em> Gene expression data of bronchoalveolar lavage (BAL) cells and clinical information were downloaded from the Gene Expression Omnibus database. We identified key modules associated with prognosis using weighted gene co-expression network analysis (WGCNA). Then we screened genes with the least absolute shrinkage and selection operator Cox regression. Finally, we constructed a prognostic gene signature using multivariate Cox regression. The risk model was evaluated using the time-dependent receiver operating characteristic (ROC) curve and the concordance index. Additionally, the risk model was validated using an external independent dataset. <em>Results:</em> Two key modules, strongly associated with inflammation and immune response, were identified by WGCNA. Four genes, including <em>TLR2</em>, <em>CCR2</em>, <em>HTRA1</em>, and <em>SFN</em>, were screened to construct the prognostic model. The patients with a high-risk score had a significantly worse prognosis than patients with a low-risk score. Time-dependent ROC analysis showed that the risk model had a moderate predictive performance for overall survival in the training and external validation datasets. <em>Conclusions:</em> Our study provides new insights into the prognostic value of BAL cells in IPF and it may be helpful to assist clinicians in making treatment decisions for the personalized management of IPF.</p>
pharmacology & pharmacy,immunology
What problem does this paper attempt to address?