Early Detection of Cystic Fibrosis Acute Pulmonary Exacerbations by Exhaled Breath Condensate Metabolomics.
Xiaoling ZangMaría Eugenia MongeDavid A GaulNael A McCartyArlene StecenkoFacundo M FernándezPublished in: Journal of proteome research (2019)
The most common cause of death in cystic fibrosis (CF) patients is progressive lung function decline, which is punctuated by acute pulmonary exacerbations (APEs). A major challenge is to discover biomarkers for detecting an oncoming APE and allow for pre-emptive clinical interventions. Metabolic profiling of exhaled breath condensate (EBC) samples collected from CF patients before, during, and after APEs and under stable conditions (n = 210) was performed using ultraperformance liquid chromatography (UPLC) coupled to Orbitrap mass spectrometry (MS). Negative ion mode MS data showed that classification between metabolic profiles from "pre-APE" (pending APE before the CF patient had any signs of illness) and stable CF samples was possible with good sensitivities (85.7 and 89.5%), specificities (88.4 and 84.1%), and accuracies (87.7 and 85.7%) for pediatric and adult patients, respectively. Improved classification performance was achieved by combining positive with negative ion mode data. Discriminant metabolites included two potential biomarkers identified in a previous pilot study: lactic acid and 4-hydroxycyclohexylcarboxylic acid. Some of the discriminant metabolites had microbial origins, indicating a possible role of bacterial metabolism in APE progression. The results show promise for detecting an oncoming APE using EBC metabolites, thus permitting early intervention to abort such an event.
Keyphrases
- cystic fibrosis
- mass spectrometry
- lung function
- liquid chromatography
- pseudomonas aeruginosa
- ms ms
- end stage renal disease
- multiple sclerosis
- chronic obstructive pulmonary disease
- ejection fraction
- machine learning
- prognostic factors
- newly diagnosed
- chronic kidney disease
- high resolution mass spectrometry
- deep learning
- pulmonary hypertension
- randomized controlled trial
- intensive care unit
- liver failure
- peritoneal dialysis
- high resolution
- electronic health record
- big data
- tandem mass spectrometry
- gas chromatography
- drug induced
- single cell
- respiratory failure
- artificial intelligence
- patient reported outcomes
- data analysis