A method of analysing cancer patients’ blood for evidence of the disease could be up to 10 times more sensitive than previous methods, according to research funded by Cancer Research UK and published in Science Translational Medicine.
The technique uses personalized genetic testing of a patient’s tumour to search blood samples for hundreds of different genetic mutations in circulating tumour DNA. The researchers and their collaborators studied samples from 105 cancer patients, testing the method on small sets of patients with five different cancer types, with both early and late stage disease.
The method detected ctDNA at high sensitivity in patients with advanced breast and melanoma cancer, and in patients with glioblastoma, which is notoriously difficult to detect in blood. The test was also able to detect ctDNA in patients with earlier-stage disease, where the level of ctDNA in the blood is much lower and difficult to find. This included patients with lung or breast cancer, as well as patients with early-stage melanoma who had already had surgery, which makes detection even more difficult.
In the coming years, this method and others based on this approach could lead to tests that more accurately determine if a patient is likely to relapse after having treatment, and could pave the way for the development of pinprick home blood tests to monitor patients.
Combined with new methods to analyze this data to remove background noise and enhance the signal, the team were able to reach a level of sensitivity that in some cases could find one mutant DNA molecule amongst a million pieces of DNA—approximately ten times more sensitive than previous methods.
“Personalised tests that can detect if cancer is still present, or find it early if it is returning, are now being tested in clinical trials,” Nitzan Rosenfeld, senior group leader at the Cancer Research UK Cambridge Institute who led the team at the University of Cambridge that conducted this research, said in a statement.
“Whilst this may be several years away from clinical use, our research shows what is possible when we push such approaches to an extreme. It demonstrates that the levels of sensitivity we’ve come to accept in recent years in relation to testing for ctDNA can be dramatically improved. At present this is still experimental, but technology is advancing rapidly, and in the near future tests with such sensitivity could make a real difference to patients,” Rosenfeld said.
Liquid biopsies to monitor cancer can become much more sensitive. Until recently, personalized liquid biopsies have searched for around 10-20 mutations in the blood and up to around 100 at most. In the material from a tube of blood, these would be able to detect ctDNA to levels on the range of 1 mutant molecule amongst 30,000 pieces of DNA.
This new technique looks for hundreds and sometimes thousands of mutations in each blood sample, routinely achieving a sensitivity of one mutant molecule per 100,000, and under optimal conditions can reach a level measured in parts per million.
In ongoing studies funded by Cancer Research UK, the team and their collaborators plan to use this method to measure ctDNA levels in individuals who are at high risk of developing cancer to help refine future tests for cancer early detection.
