Article Text

Download PDFPDF
PWE-116 Detection of Colorectal Cancer from Urinary Volatile Organic Compounds Using a New Chromatograph/Electronic-Nose Instrument – Wolf System
  1. E Westenbrink1,
  2. N O’Connell2,
  3. C Bailey2,
  4. C Nwokolo2,
  5. K Bardhan3,
  6. R Arasaradnam4,
  7. J Covington1
  1. 1School of Engineering, University of Warwick
  2. 2Department of Gastroenterology, University Hospital Coventry and Warwickshire, Coventry
  3. 3Rotherham General Hospital, Rotherham
  4. 4Clinical Sciences Research Institute, University of Warwick, Coventry, UK


Introduction Colorectal cancer (CRC) remains one of the leading causes of cancer-related death in Europe and the USA. The gold standard diagnostic test of colonoscopy is highly invasive, expensive and has an associated morbidity. The current non-invasive option for CRC screening includes Faecal Immunochemical Testing (FIT) for haemoglobin, which shows a specificity of 87–96%, but has a wide variation in potential sensitivity (66–88%). One non-invasive method that is gaining interest for the diagnosis of a variety of cancers measures the volatiles/gases that emanate from human biological media.

Methods Urine samples were collected from 26 CRC and 23 controls (Irritable bowel syndrome patients; IBS) and stored at -80 oC. 5 mL aliquots were heated to 40oC for 5 minutes to develop sufficient headspace. The WOLF 3.1 gas chromatograph/electronic nose instrument, developed at Warwick University, was used to analyse the resultant headspace. The analysis method took a total of 25 minutes for each sample, with an air purge in between to avoid cross-contamination. Statistical evaluation by Linear Discriminant Analysis (LDA) was tested by repeated trials of single unknown sample by re-introduction and re-classification by a K-Nearest-Neighbour technique.

Results Figure 1 below shows the 2 group LDA classification of CRC and IBS samples using response time slices of 100 seconds and extraction of two response features from each. There is distinction between the disease groups, with no overlap seen in any of the samples in this population (P < 0.0001). The sensitivity and specificity of distinguishing CRC from IBS controls from KNN re-classification were 92% and 77% respectively.

Conclusion This pilot study affirms the utility of a custom made WOLF 3.1 gas chromatograph-electronic nose instrument to detect CRC using urine samples. Further validation in a larger sample set is underway but holds promise for a simple, economical tool in CRC detection.

Disclosure of Interest None Declared

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.