DELFI (blood test for cancer detection)

2022: Announcement

In mid-November 2022, artificial intelligence blood testing technology developed and used by Johns Hopkins Kimmel Cancer Center researchers to successfully detect cancer now identifies more than 80% of cases.

A blood test called DELFI (evaluation of DNA fragments for early interception) reveals changes in the fragmentation of DNAcancer cells that have entered the bloodstream and are known as cell-free DNA (cfDNA). In the latest study, scientists used DELFI technology on blood plasma samples obtained from 724 people from the US, the European Union (EU) and Hong Kong to detect hepatocellular cancer (HCR), a type of liver cancer.

New high-precision test diagnoses liver cancer by blood test

The researchers believe this is the first genome fragmentation analysis independently confirmed in two high-risk populations and in different racial and ethnic groups with different liver cancer-related causes. According to various estimates, 400 million people worldwide are at increased risk of developing HCC due to cirrhosis caused by chronic liver diseases, including chronic viral hepatitis or non-alcoholic fatty liver disease.

Of the 724 plasma samples examined, 501 were collected in the US and the EEC and included samples from 75 people with HCC for training and validation of a machine learning (AI) model that uses data and algorithms to improve accuracy, explains Zacharia Foda. For validation, another 223 blood plasma samples from people from Hong Kong were analyzed, including samples from 90 people with HCC, 66 people with hepatitis B virus (HBV), 35 people with HBV-associated liver cirrhosis, and 32 people without major risk factors.

DELFI technology uses a blood test to measure the way DNA is packaged inside a cell nucleus by studying the size and amount of cell-free DNA present in the bloodstream from different regions of the genome. Healthy cells pack DNA like a well-organized suitcase in which different regions of the genome are neatly placed in different compartments. Cancer cell nuclei, by contrast, resemble more disorganized suitcases in which elements from different regions of the genome are thrown haphazardly. When cancer cells die, they eject fragments of DNA in a chaotic order into the bloodstream.

DELFI determines the presence of cancer by examining millions of kfDNA fragments for abnormal patterns, including the size and amount of DNA in different genomic regions. The DELFI approach only requires low-coverage sequencing, allowing the technology to be used in screening settings, the researchers say.

In the latest study, the scientists conducted a test - which has been shown to accurately classify lung cancer - on kfDNA fragments isolated from plasma samples. They analyzed the pattern of fragmentation in each sample to develop a DELFI score. Rates were low for cancer-free individuals with viral hepatitis or cirrhosis (median DELFI 0.078 and 0.080, respectively), but 5-10 times higher on average for 75 HCC patients in US/EU samples, with high rates observed in all stages of cancer, including early stage disease (DELFI for step 0 = 0.46, step A = 0.61, step B = 0.83, and step C = 0.92). In addition, the test detected fragmentary changes in the content and packaging of liver cancer genomes, including from regions of the genome associated with liver-specific activity.

The technology revealed liver cancer in its earliest stages, with the overall sensitivity - or ability to accurately detect cancer - at 88% and specificity at 98%, meaning it almost never produced false positive results, among those at average risk. In samples collected from people at high risk of HCC, the test had a sensitivity of 85% and a specificity of 80%. The next steps include testing this approach in larger studies for clinical application, the researchers said.^[1]

Notes