Exhalation metabolomics for noninvasive staging biomarker exploration in silicosis

Silicosis is a prevalent chronic occupational disease, causing incurable damage to the lungs. The conventional methods for diagnosing silicosis are costly and complex. Noninvasive biomarker studies based on exhalation metabolomics have potential in the early diagnosis of silicosis, but existing studies remain scarce and especially lack the biomarkers for staging diagnosis. Exhaled breath from 74 healthy controls and 112 patients, including 28 stage Ⅰ silicosis patients (SILs), 22 stage Ⅱ SILs, 52 stage ࡲ SILs and 10 coal workers' pneumoconiosis patients, were detected using solid-phase microextraction incorporating gas chromatography-mass spectrometry for the identification of the volatile metabolites. The univariate statistical analysis and orthogonal partial least squares-discriminant analysis were employed to screen potential biomarkers of SILs, with diagnostic performance assessed with the receiver-operating characteristic (ROC) curve and decision tree model. Fourteen volatile metabolites were found to distinguish different stages of SILs from healthy controls, and 8 metabolites differentiating stage Ⅰ and ࡲ as well as 3 metabolites distinguishing stage Ⅱ and ࡲ. The ROC analysis of silicosis based on the biomarkers exhibited an area under curve (AUC) of more than 0.9, with the largest AUC of 0.986 in stage Ⅰ SILs and healthy controls. Mechanistic exploration showed that these biomarkers were associated with inflammatory response, oxidative stress, and fibrosis in silicosis, respectively. This study preliminarily screened out the biomarkers of exhaled breath for different stages of SILs, and explored the metabolic pathways of biomarkers, which can provide basic data for the early, staging and specific diagnosis of SILs.