MAY 30, 2019 06:19 AM PDT

This blood test uses machine learning to detect multiple cancers

A new study published online in Nature details the results of an investigation on a blood test that is capable of detecting multiple types of cancer. The test is called DELFI and stands for “DNA evaluation of fragments for early interception.” It looks for particular patterns of fragmentation from cancer cells’ DNA in the bloodstream in order to identify the presence of breast, colorectal, lung, ovarian, pancreatic, gastric and bile duct cancers.

The study looked at 208 patients with various stages of the above-listed cancers in the U.S., Denmark and the Netherlands and found that DELFI was able to detect the presence of cancers in 57-99% of the patients’ blood samples. It is also important to note that DELFI did result in some false positives, identifying cancer in four members of the control group of 215 that did not have actually have cancer.

DELFI is different from conventional liquid biopsy cancer detection tests because it uses machine learning to detect abnormal patterns of DNA fragments in cancer patients’ blood.

"For various reasons, a cancer genome is disorganized in the way it's packaged, which means that when cancer cells die they release their DNA in a chaotic manner into the bloodstream," says Jillian Phallen, Ph.D., a lead author on the study and a Johns Hopkins Kimmel Cancer Center postdoctoral fellow. "By examining this cell-free DNA (cfDNA), DELFI helps identify the presence of cancer by detecting abnormalities in the size and amount of DNA in different regions of the genome based on how it is packaged."

Photo: Pixabay

The researchers say that DELFI has the potential to identify the cancers' tissue of origin in up to 75% of cases, which is helpful in determining the source of the cancer. The test is also beneficial in that it is inexpensive and easy to administer.

"We're encouraged about the potential of DELFI because it looks at a completely independent set of cell-free DNA characteristics from those that have posed difficulties over the years, and we look forward to working with our collaborators worldwide to make this test available to patients," says senior study author Victor E. Velculescu, M.D., Ph.D., professor of oncology and co-director of the Cancer Biology Program at the Johns Hopkins Kimmel Cancer Center.

Sources: Science Daily, Nature

About the Author
  • Kathryn is a curious world-traveller interested in the intersection between nature, culture, history, and people. She has worked for environmental education non-profits and is a Spanish/English interpreter.
You May Also Like
JAN 17, 2020
Cardiology
JAN 17, 2020
What is Polycythemia Vera?
Polycythemia Vera (PV) is a slow-growing cancer of the blood. The cancer is characterized by an overproduction of blood cells within the bone marrow. Too m...
JAN 17, 2020
Immunology
JAN 17, 2020
Immunity to HPV Can Prevent Skin Cancer
Scientists discovered a surprise benefit of low-risk human papillomaviruses (HPV), a typical resident of human skin. New research shows that these harmless...
JAN 17, 2020
Immunology
JAN 17, 2020
Harnessing the Power of Natural Killer Cells to Fight Cancer
Manipulating the immune system’s population of natural killer cells could bolster therapies targeting cancer. A new study saw positive results invest...
JAN 17, 2020
Cancer
JAN 17, 2020
Inhalable nanoparticle-immunotherapy provides new hope for lung cancer
New research published in Nature Communications details the results of a study that tested inhalable nanoparticle-immunotherapy to determine its impact on ...
JAN 17, 2020
Cancer
JAN 17, 2020
Using AI to determine which patients are best suited for immunotherapy
A new study published in the journal Cancer Immunology Research suggests that we can use artificial intelligence to help determine which people with lung c...
JAN 17, 2020
Cancer
JAN 17, 2020
Physical exercise could halve the risk of prostate cancer
New research suggests physical activity could act as a tool to prevent increased risk for prostate cancer. The research comes from scientists at the U...
Loading Comments...