A computer-based blood test that detects cancer and locates where it is in the body, has been developed by researchers. Professor Jasmine Zhou from the University of California at Los Angeles, and her team, created the computer program, called CancerLocator.
It works by searching for specific molecular patterns in cancer DNA that is free flowing in the patients’ blood. It then compares the patterns against a database of tumor epigenetics, from different cancer types, collated by the authors.
DNA from tumor cells is known to end up in the bloodstream in the earliest stages of cancer so it offers a unique target for early detection of the disease.
Non-invasive Cancer Diagnosis
Professor Zhou said:
“Non-invasive diagnosis of cancer is important, as it allows the early diagnosis of cancer, and the earlier the cancer is caught, the higher chance a patient has of beating the disease. We have developed a computer-driven test that can detect cancer, and also identify the type of cancer, from a single blood sample. The technology is in its infancy and requires further validation, but the potential benefits to patients are huge.”
Using blood tests to screen for cancer is not a new strategy. But with an over 80 per cent success rate in detecting breast, liver and lung cancer, researchers believe their test is one of the most comprehensive available.
The new computer program, and two other methods, called Random Forest and Support Vector Machine, were tested with blood samples from 29 liver cancer patients, 12 lung cancer patients and 5 breast cancer patients. Tests were run 10 times on each sample to validate the results.
The Random Forest and Support Vector Machine methods had an overall error rate (the chance that the test produces a false positive) of 0.646 and 0.604 respectively, while the new program obtained a lower error rate of 0.265.
Ready In One Year?
25 out of 29 liver cancer patients and 5 out of 12 lung cancer patients tested had early stage cancers, which the program was able to detect in 80% of cases. The level of tumor DNA present in blood is much lower during the early stages of these cancers. But the program was nevertheless able to make a diagnosis showing the potential of this method for the early detection of cancer, say the researchers.
Speaking to The Independent, Professor Zhou said:
“I hope it will be available within a year. It depends on training data, testing and machine learning. With enlarged training and testing data we are confident to achieve much higher performance.”
Other experts are taking a more cautious attitude about that hope. In an interview with the Independent, the University of Cambridge’s Professor Paul Pharoah said:
“If you were going to use this test as a screening test, ie. to detect cancer in otherwise healthy people, it would need to be evaluated as a screening test. Any study of any screening modality takes years to do. For example, to show you could detect pancreatic cancer early, before symptoms develop, would take a trial of hundreds of thousands of people over years.”
Shuli Kang et al
CancerLocator: non-invasive cancer diagnosis and tissue-of-origin prediction using methylation profiles of cell-free DNA
Genome Biology (2017). DOI: 10.1186/s13059-017-1191-5