This article is original by Transformational Medicine Network. Please indicate the source when reprinting. Author: Lily Introduction: About 350 million people worldwide suffer from chronic viral hepatitis infection and 50 million people suffer from cirrhosis; for these people, th

This article is original from Translational Medicine Network. Please indicate the source when reprinting

Author: Lily

Introduction : About 350 million people worldwide suffer from chronic viral hepatitis infection, and 50 million people suffer from cirrhosis ; for these people, the risk of liver cancer will be greatly increased. Currently, screening methods for high-risk populations include: abdominal ultrasound imaging with or without alpha-fetoprotein (AFP), with screening sensitivity ranging from 47%-84%, and specificity between 67%-90%. is currently in urgent need of developing sensitive and efficient non-invasive hepatocellular carcinoma (HCC) screening methods. Recently, a research team at Johns Hopkins University in the United States developed a blood-based genome-wide cfDNA fragmented feature detection method, providing a new and high-performance and cost-effective choice for HCC detection.

The incidence and mortality rate of liver cancer rank among the top among cancers, with more than 900,000 new cases worldwide and more than 800,000 deaths each year. Among all liver cancer cases, hepatocellular carcinoma (HCC) accounts for about 90%, and the patient's survival rate depends largely on the disease stage at diagnosis: when the tumor is in the localized stage, the five-year survival rate of is 34%, accounting for 44% of the total number of patients; the five-year survival rate of the regional stage is 12%, accounting for 27% of the total number of patients; the five-year survival rate of the distant patients is 3%, accounting for 18% of the total number of patients. Therefore, it is particularly important to conduct effective and highly sensitive screening of people at high risk of hepatocellular carcinoma (HCC).

In recent years, cell free DNA (cfDNA) biomarkers have provided another possibility for cancer detection. Recently, a research team at Johns Hopkins University in the United States developed a blood-based genome cfDNA fragmentation feature detection method, providing a high-performance and cost-effective new option for HCC detection. The research results related to were published in the authoritative international academic journal Cancer Discovery (IF: 38.27).

https://aacrjournals.org/cancerdiscovery/article/doi/10.1158/2159-8290.CD-22-0659/711014/Detecting-liver-cancer-using-cell-free-DNA

cfDNA genome analysis

01

researchers tested plasma samples from 501 individuals—75 were from HCC patients; 426 were from non-cancer patients. In the non-cancer cohort, 133 people suffered from diseases that increased the risk of HCC (including cirrhosis caused by various causes or cirrhosis-free viral hepatitis ). The research team used 0.5-5ml of plasma to generate a genome library and performed low coverage whole genome sequencing of cfDNA fragments. At the same time, the researchers also tested 223 whole-genome sequence data from Hong Kong patients to serve as verification cohorts—including resectable early HCC ((n=90), HBV (n=66), hepatitis B-related cirrhosis (n=35), and healthy individuals without liver disease (n=32).

Whole-genome cfDNA fragment spectrum determined by chromatin structure

02

researchers evaluated the cfDNA fragment profile, and generated fragment profiles of the entire genome in 473 non-overlapping 5MB regions, each containing about 80,000 fragments; and spanned approximately 2.4GB of genome using the DELFI method. The results showed that there was no significant difference in cfDNA fragment profile in cancer-free individuals; there was a large difference in cfDNA fragment profile in HCC patients. Compared with HCC patients, the characteristics of patients with cirrhosis were closer to those of non-cancer individuals without cirrhosis. The fragment profile of patients with toxic hepatitis is almost the same as that of non-cancer individuals without liver disease. The study also found that the cfDNA fragmentation pattern of healthy individuals is highly correlated with lymphoblasts; the cfDNA fragment group from HCC patients represents a mixture of cfDNA profiles of peripheral blood cells and chromatin compartments of hepatocellular carcinoma cells.

genome-wide fragment profile reflects the potential chromatin structure

DELFI model validation

03

Through machine learning methods, the researchers explored the question of "whether changes in cfDNA fragment groups can distinguish HCC patients from non-cancer patients" and built a DELFI model.The results showed that among the average risk population, the DELFI model's sensitivity to HCC detection was 88% and the specificity was 98%; among the high-risk population, the DELFI model's sensitivity to HCC detection was 85% and the specificity was 80%.

researchers examined the relationship between DELFI scores and HCC occurrence and staging in high-risk populations. The DELFI scores of 133 non-cancer individuals were low, while the DELFI score of patients with viral hepatitis or cirrhosis was 0.078 or 0.080, respectively. In contrast, 75 HCC patients scored significantly higher median DELFI at all stages (Figure 4A). Characteristics (ROC) curves of the DELFI method used to identify HCC patients showed that in high-risk individuals, the area under the curve (AUC) was 0.90 (Fig. 4B). The performance of early HCC was still robust, and individuals with late HCC were almost completely detected (AUC0.97) (Fig. 4C). In the Asian validation cohort, the DELFI model distinguished HCC patients with AUC of 0.97 from high-risk individuals (Fig. 4D), indicating that the basic characteristics of cfDNA fragmentation are similar in this cohort, and that DELFI is a reliable method for detecting HCC and is expected to be generalized in high-risk populations of HCC.

Research significance

04

cfDNA fragmentation group analysis method "DELFI" is the first genome-wide fragmentation analysis independently verified in a single high-risk population, with stable and powerful performance in detecting HCC. In this study, the training cohort of the DELFI scoring model came from Europe and the United States, while the verification cohort was from Hong Kong, with racial differences, and the whole genome sequencing methods used by the two cohorts were also different. However, the resulting DELFI scoring model in the training queue still performed well in the verification queue, which strongly demonstrated the robustness of the model. The biological basis of the open/close state of the cfDNA fragment group and chromatin and the activity of transcription factors revealed by the

study further enriches people's understanding of the essential properties of cfDNA, greatly expands the information load of liquid biopsy, and provides new ways and ideas for real-time dynamic monitoring of the functional status of tumors.

Reference:

https://aacrjournals.org/cancerdiscovery/article/doi/10.1158/2159-8290.CD-22-0659/711014/Detecting-liver-cancer-using-cell-free-DNA

Note: This article aims to introduce the progress of medical research and cannot be used as a reference for treatment plans. If you need health guidance, please go to a regular hospital for treatment.

Recommended·Live/Activity

D 08 14:00-17:00 Online

High-deep proteomics application in precision medicine Online seminar

D 10 08:00-12:10 Online

CSCO 2022 Lung Point-Same Frequency with the World