A new artificial intelligence blood test can detect liver cancer

Researchers at Johns Hopkins Kimmel Cancer Center developed and used a new AI-powered blood test technology that successfully detected lung cancer in a 2021 study and now detects more than 80 percent of liver cancers
in a new study of 724 people.

This blood test is called DELFI (DNA Assessment for Early Interception of DNA Fragments), and it detects changes in cancer cells' DNA fragments, known as cell-free DNA (cfDNA),
that enter the bloodstream.
In a recent study, researchers used DELFI technology to test plasma samples from 724 individuals from the United States, the European Union and Hong Kong to detect hepatocellular carcinoma (HCC), a type of liver cancer
.

The researchers believe this is the first genome-wide fragmentation analysis independently validated in two high-risk populations involving different causes of liver cancer in different races and ethnicities
.

Their findings were presented Nov.
18 in Cancer Discovery and the American Association for Cancer Research Special Meeting: Precise Prevention, Early Detection and Interception
of Cancer.

According to a global liver disease burden analysis (J.
Hepatology, 2019), it is estimated that 400 million people worldwide are at higher risk of developing HCC due to cirrhosis caused by chronic liver disease, including chronic viral hepatitis or nonalcoholic fatty liver disease
.

"Increasing early detection of liver cancer can save lives, but the screening tests currently available are underutilized and miss many cancers," said Victor Velculescu, MD, professor of oncology at the Kimmel Cancer Center at Johns Hopkins University and co-director of the Cancer Genetics and Epigenetics Program, along with Zachariah Foda, MD, Gastroenterology Researcher, Akshaya Annapragada, MD.
The Ph.
D.
co-led the study
.
She is an assistant professor of medicine at Johns Hopkins University School of Medicine and Amy King
, MD.

Foda explained that of the 724 plasma samples studied, 501 were from the United States and Europe, including samples from 75 HCC patients, for training and validating machine learning models, an artificial intelligence
that uses data and algorithms to improve accuracy.
To validate, we analyzed an additional 223 plasma samples from Hong Kong, including samples from 90 HCC patients, 66 hepatitis B virus (HBV) patients, 35 hepatitis B-associated cirrhosis patients, and 32 patients with no underlying risk factors
.

DELFI technology uses blood tests to measure how
DNA is packaged within the nucleus by studying the size and amount of free cellular DNA circulating in different regions of the genome.
Healthy cells pack DNA like a well-organized suitcase, and different regions of the genome are carefully placed in
different compartments.
In contrast, the nuclei of cancer cells are like more cluttered suitcases, with items in the genome thrown in haphazardly
.
When cancer cells die, they release pieces of DNA into the bloodstream
in a chaotic way.

DELFI identifies the presence of cancer by examining abnormal patterns in millions of cfDNA fragments, including the size and amount of
DNA in different genomic regions.
The researchers say the DELFI method only requires low-coverage sequencing, making the technique cost-effective
in screening settings.

In the latest study, the researchers tested cfDNA fragments isolated from plasma samples — previously proven to accurately classify
lung cancer.
They analyzed the fragmentation patterns of each sample to arrive at a DELFI score
.

Patients with viral hepatitis or cirrhosis without cancer had lower DELFI scores (median DELFI scores of 0.
078 and 0.
080, respectively), but high scores
were observed on average 5- 10 times higher in 75 HCC patients in the US/EU at all stages of cancer, including early-stage disease (stage 0 DELFI score = 0.
46, stage A = 0.
61, stage B = 0.
83, stage C = 0.
92).
In addition, the test detected fragmented changes in the contents and packaging of the liver cancer genome, including genomic regions
associated with liver-specific activity.

DELFI technology is detectable at an early stage of liver cancer, with an overall sensitivity (or ability to accurately detect cancer) of 88% and specificity of 98%, meaning that it rarely provides false-positive results
falsely in people at average risk.
In samples collected from populations at high risk of HCC, the test had 85% sensitivity and 80% specificity
.

"Currently, less than 20% of high-risk groups are screened
for liver cancer due to accessibility and poor testing performance.
This new blood test could double the number of liver cancer cases detected and increase early detection of cancer compared to existing standard blood tests," said
Kim, co-senior author of the study.

The researchers say the next step includes validating the approach
in larger clinical studies.

According to the American Cancer Society, more than 800,000 people worldwide are diagnosed with liver cancer each year, the leading cause
of cancer death worldwide.

Journal Reference:

1. Zachariah H.
   Foda, Akshaya V.
   Annapragada, Kavya Boyapati, Daniel C.
   Bruhm, Nicholas A.
   Vulpescu, Jamie E.
   Medina, Dimitrios Mathios, Stephen Cristiano, Noushin Niknafs, Harry T.
   Luu, Michael G.
   Goggins, Robert A.
   Anders, Jing Sun, Shruti H.
   Mehta, David L.
   Thomas, Gregory D.
   Kirk, Vilmos Adleff, Jillian Phallen, Robert B.
   Scharpf, Amy K.
   Kim, Victor E.
   Velculescu.
   Detecting liver cancer using cell-free DNA fragmentomes.
   Cancer Discovery, 2022; DOI: 10.
   1158/2159-8290.
   CD-22-0659