Liquid biopsy is a widely used method inas an early measure for the diagnosis of cancer. The spinal fluid, blood, and urine are taken as the test material. Regular research by American scientists led them to develop a blood test that identifies the most common and deadly malignant neoplasms.
DELFI test presented by scientists from the UniversityJohns Hopkins, It is based on the identification of cell-free DNA (cfDNA) cancer cells in the bloodstream using artificial intelligence techniques. The DELFI method can detect breast and pancreas, lung, ovarian, stomach, intestinal and bile duct cancer at various stages.
In a cancer cell, the organization of the genome is messy, so when the cell dies, it randomly sends its DNA into the bloodstream. Here it is detected and analyzed by DELFI.
Machine learning DELFI allows you to simultaneouslyanalyze millions of sequences from hundreds to thousands of regions in the genome, and even on small amounts of cfDNA, reveal abnormalities characteristic of a tumor.
208 cancer patients at different stages have passedtesting according to the developed analysis of DELFI before the start of any treatment. The study also included 215 healthy people. DELFI diagnosed the presence of a tumor in 73% of cases and gave four false positive results in healthy people.
When, on the basis of mutations, DELFI and cfDNA analysis were combined, it was already possible to detect cancer in 91% of cases, the authors emphasized.
Now the team improves the test and will continueresearch on more samples. Noting that the analysis is very simple and does not require the use of expensive laboratory methods, the team, meanwhile, reports on the need to further improve the test and conduct research on more samples.