News

Microbe DNA in blood samples could reveal clues about cancer – study

Bacterial and viral DNA associated with cancer could help detect and treat the disease at its early stage, scientists have said.
Bacterial and viral DNA associated with cancer could help detect and treat the disease at its early stage, scientists have said. Bacterial and viral DNA associated with cancer could help detect and treat the disease at its early stage, scientists have said.

A blood test that detects the DNA of the microbes associated with cancer could help detect and treat the disease at its early stage, scientists have said.

Researchers in the US have found a way to identify the “signatures” of bacterial and viral DNA across multiple cancers by using machine learning techniques to analyse the blood samples of patients.

The team believe their work, which is still in its preliminary stages, could in future provide a “new therapeutic avenue” for treating different types of cancer.

Dr Sandip Pravin Patel, a medical oncologist at Moores Cancer Centre at the University of California, San Diego – and one of the study authors, said: “The ability, in a single tube of blood, to have a comprehensive profile of the tumour’s DNA (nature) as well as the DNA of the patient’s microbiota (nurture), so to speak, is an important step forward in better understanding host-environment interactions in cancer.

“With this approach, there is the potential to monitor these changes over time, not only as a diagnostic, but for long-term therapeutic monitoring.

“This could have major implications for the care of cancer patients, and in the early detection of cancer, if these results continue to hold up in further testing.”

The researchers used The Cancer Genome Atlas database, an online catalogue of genetic mutations responsible for cancer, to analyse data for 33 cancer types – totalling more than 17,000 samples from 10,000 patients.

Along with known associations between microbes and cancer, such as the connection between human papillomavirus (HPV) and cervical cancer, the team said they also identified previously unknown associations, such as a link between the Faecalibacterium species and colon cancer.

After collecting microbe profiles of the cancer samples, the team trained machine learning algorithms to associate certain microbial patterns with the presence of specific cancers.

The researchers then tested their algorithms using data from the blood samples collected from 59 patients with prostate cancer, 25 with lung cancer and 16 with melanoma.

They said the machine-learning models were able to tell which participants had which of the three cancer types, correctly distinguishing between prostate and lung cancer “with 81% sensitivity”.

However, the researchers said there is still the possibility their test could miss signs of cancer and return a false-negative result, adding that their method will become more accurate as they refine their machine learning models with more data.

Dr Sandrine Miller-Montgomery, a professor of practice in the Jacobs School of Engineering at the University of California, San Diego, and one of the study authors, said: “This new understanding of the way microbial populations shift with cancer could open a completely new therapeutic avenue.”

The lead authors on the study have filed a provisional patent application in the US bases on their work, which is published in the journal Nature.