Microbiota with AI for colorectal cancer

Olfat Khannous

Institut de Recerca Biomèdica (IRB Barcelona), Barcelona, Spain

Colorectal cancer is the second most common cause of cancer mortality worldwide. Some 1.9 million cases are diagnosed every year. Preventive measures and early diagnosis can save many lives and ensure a better quality of life for patients. Routine screening of populations over a certain age is being implemented globally. Current screening systems consist of a two-step procedure with a non-invasive test (which detects the presence of haemoglobin in the stool), followed by a colonoscopy if the test is positive. This approach is effective, but results in many false positives (> 60 %) and places a significant burden on the healthcare system, hospitals, doctors and patients.

This project applies an innovative approach that combines gut microbiota analysis with artificial-intelligence-based classification algorithms, with the aim of improving current screening systems and reducing the number of unnecessary colonoscopies. This new test has been designed to adapt to current EU screening protocols, in that the same stool sample is used to detect microbial markers. Its implementation in screening campaigns would reduce unnecessary colonoscopies by up to 30 %, as it would detect with high sensitivity cases of colorectal cancer and high-risk precancerous lesions. The team has already been able to validate the prototype test with more than 3,000 real patient samples for which colonoscopy test results were also available. In this stage of the project, the team will work to advance this promising technology towards clinical validation.

MICROSCREEN_CRC: Microbial biomarker test for early diagnosis of colorectal cancer (MiCRC)

Stage 2