ABC-Cancer

Background. Colorectal cancer (CRC) is Europe’s second most prevalent malignant disease. Due to its gradual development via premalignant lesions, its timely detection requires regular screenings. Routine screening programmes, as recommended by the Council of Europe and undertaken from age 50, are key to preventing CRC. One common screening method is the immunochemical faecal-occult-blood-test (iFOBT), but this suffers from high false-positive rates and requires endoscopic follow-up. As such, the current sole diagnostic procedure for CRC involves invasive colonoscopy. Therefore, there is a pressing need for an accurate, non-invasive diagnostic tool for CRC that offers a sufficiently high negative predictive value (NPV) such that it can be implemented in primary screening routines and thereby reduce the frequency of colonoscopy procedures. Recently, volatile organic compounds (VOCs) from biological fluids, including blood, urine and breath, have been proposed as potential biomarkers for various diseases. State-of-the-art analytical technologies based on mass spectrometry (MS) can identify thousands of VOCs in biological media, known as the volatilome, and past studies have linked certain VOCs with specific diseases like cancer. The driving hypothesis for this proposal is that CRC, as well as its pre-cancerous stages, leads to the generation of specific VOCs that manifest in faeces and exhaled breath, thereby offering the opportunity to identify potential biomarkers or patterns that can be implemented in CRC screening, diagnosis and monitoring.
Methods. The volatilome associated with CRC will be derived from analyses of VOCs in faecal headspace and exhaled breath by complementary MS-based technologies. Patient cohorts will include sufferers of diverse bowel disorders including CRC, colonic adenomas, inflammatory bowel disease (IBD), irritable bowel syndrome (IBS), and healthy controls without intestinal impairment. Multiple patient cohorts will be sourced from national screening programmes at clinical centres in the Netherlands, Germany and Poland. Machine learning will be applied to pooled volatilome data to search for unique CRC-specific VOC fingerprints that allow discrimination from other conditions. The predictive power of the screening tool will be independently validated in data subsets (bootstrapping) and compared with iFOBT data.
Expected results and potential impact. Successful identification of volatile CRC biomarkers in faeces and/or breath will offer a non-invasive, simple and low-cost CRC screening method that will contribute to reductions in associated morbidity, mortality and healthcare costs. Moreover, discovery of a CRC-specific volatilome will enable a targeted, future development of novel VOC sensing-based point-of-care (PoC) devices for non-invasive diagnosis and monitoring of CRC via faeces and/or exhaled breath.