Ketones and Terpenes in exhaled breath as prospective biomarkers for chronic liver diseases

Shreya Louis 1,6, Owen Birch 2, Shirin Siddiqi 3, Alejandro Rodarte 3, Daniella Allende 4, J. Mark Brown 5, Federico Aucejo 3, Jason Kinchen 2, Daniel Rotroff 1,6

1. Department of Quantitative Health Sciences, Lerner Research Institute, Cleveland Clinic, Cleveland, Ohio 2. Owlstone Medical Ltd., Cambridge, Cambridgeshire, UK 3. Department of General Surgery, Cleveland Clinic, Cleveland, Ohio 4. Department of Pathology, Cleveland Clinic, Cleveland, Ohio 5. Department of Cardiovascular and Metabolic Sciences, Cleveland Clinic, Cleveland, Ohio 6. Cleveland Clinic Lerner College of Medicine, Case Western Reserve University, Cleveland, Ohio

Poster PDF

Abstract:

Cirrhosis is the end-result of chronic hepatic injury, and is often associated with obesity, type 2 diabetes, or exposure to damaging agents (e.g. alcohol abuse, viral infection, etc). Early disease is often asymptomatic, with half of cases diagnosed at an advanced stage during an episode of decompensation, at which point damage is considered irreversible. There is a pressing need for diagnostic tests allowing early detection of hepatic disease, both to screen susceptible populations but also to enable drug discovery. 

Here, we use a cross-sectional, single-site design where breath is collected from healthy controls and subjects diagnosed with hepatic disease (cirrhosis, hepatocellular carcinoma, colorectal cancer with liver metastasis) using the ReCIVA® Breath Sampler.  At a high-level, observations in this pilot-scale study (using unadjusted p < 0.05 as the threshold for interest) replicated several findings in the literature, including associations of features tentatively identified as ketones and terpenoids with disease.  Evaluation of MELD score (proxy for severity) suggest biomarker candidates may have utility as surrogates for this marker. 

Sample size was limited for this pilot project.  Replication of findings in a larger cohort may provide additional insight into clinical utility for these findings, particularly if a broader range of subjects representative of early disease are included.