Colorectal cancer (CRC) is the 2nd most common UK cause of cancer death. The bowel cancer screening programme (BCSP) has saved lifes through early cancer diagnosis. However, it only targets those aged 60-74, and leads to many potentially unnecessary colonoscopies. A breath test could be used as a more specific non-invasive screening tool, or to triage/reassure symptomatic patients. Prior studies have shown promising results using exhaled propanal for detecting CRC (96% sensitivity/76% specificity). The Colorectal BReath Analysis (COBRA) study aims to determine the diagnostic accuracy of breath volatile organic compounds for detecting CRC, first by analysing breath of known colorectal cancer patients, then by testing the resultant diagnostic model on a prospective BCSP population, target 2000 patients.
Clinical data was collected from patients attending for colonoscopies or CRC surgery across 4 London centres July 2017 to May 2018. Exhaled breath (500mls) was collected using the ReCIVA™ breath sampling device, onto thermal desorption tubes. Analysis by gas chromatography mass spectrometry and proton transfer reaction (PTR) mass spectrometry identified and quantified breath compounds.
426 patients were recruited for this preliminary dataset. 80 samples were excluded due to inadequate colonoscopies, instrument faults or inadequate quality measures. This left 346 patients; 20 with known CR adenocarcinoma were sampled pre-operatively, the rest from endoscopy units pre-colonoscopy. All were nil by mouth having had bowel preparation. Colonoscopy results showed: 29 CR adenocarcinoma, 84 normal, 35 benign pathology, 31 IBD, 83, 20 and 57 with low, intermediate and high risk polyps, and 7 with polyposis syndromes. PTR data analysis so far indicates that there are discriminatory breath compounds between pathology groups, but analysis is ongoing to investigate potential confounders.
Further analysis of this preliminary dataset is expected to reveal compounds of interest for colorectal cancer diagnosis. The true diagnostic accuracy of breath testing is expected to be revealed once these compounds are tested on the larger population dataset for COBRA (recruitment ongoing since January 2018).
Clinical trial identification
Legal entity responsible for the study
Imperial College London.
Imperial College Charity - Rosetrees.
All authors have declared no conflicts of interest.