From the Journals

Level of hepatitis B core–related antigen is risk factor for hepatocellular carcinoma



A high level of hepatitis B core–related antigen (HBcrAg) was a complementary risk factor for hepatocellular carcinoma, according to the results of a retrospective cohort study of more than 2,600 noncirrhotic adults with untreated hepatitis B virus (HBV) infection with a median of 16 years of follow-up.


“Patients with an intermediate viral load and high levels of HBcrAg had a risk for hepatocellular carcinoma that did not differ significantly from that of patients with a high viral load. [An] HBcrAg of 10 KU/mL may serve as a novel biomarker for the management of patients with intermediate viral load in our clinical practice,” wrote Tai-Chung Tseng, MD, PhD, of National Taiwan University Hospital in Taipei and associates in Gastroenterology.

Deciding whether to start antiviral therapy is controversial for some patients with HBV infection. Typically, monitoring without treatment is recommended for patients who have both low hepatitis B surface antigen levels (less than 1,000 IU/mL) and low levels of HBV DNA (less than 2,000 IU/mL), and early antiviral therapy is recommended for patients who have high levels of HBV DNA (20,000 IU/mL or more). However, there is no clear evidence that early antiviral therapy benefits patients who have intermediate levels of HBV DNA (2,000-19,999 IU/mL) and are negative for hepatitis B e antigen. Biomarkers for risk-stratifying these patients also are lacking, the researchers noted.

Therefore, they studied a cohort of 2,666 adults who had tested positive for hepatitis B surface antigen and were followed at National Taiwan University Hospital from 1985 through 2000. No patient had cirrhosis at baseline. In all, 209 patients developed hepatocellular carcinoma, yielding an incidence rate of 4.91 cases per 1,000 person-years.

Hepatitis B core–related antigen level remained an independent risk factor for hepatocellular carcinoma after accounting for age, sex, serum alanine aminotransferase (ALT) level, FIB-4 index, hepatitis B e antigen status, hepatitis B genotype (B, C, or undetermined), and HBV DNA level. Compared with patients whose HBcrAg level was less than 10 KU, a level of 10-99 KU/mL was associated with a nearly threefold increase in risk for hepatocellular carcinoma (HR, 2.93; 95% CI, 1.67-4.80), and this risk rose even further as HBcrAg levels increased.

In the subgroup of patients who tested negative for hepatitis B e antigen, had an intermediate HBV DNA load (2,000-19,999 IU/mL), and had a normal baseline ALT level (less than 40 U/L), a high HBcrAg level (10 KU/mL or more) was tied to a nearly fivefold greater risk for hepatocellular carcinoma (HR, 4.89; 95% CI, 2.18-10.93). This approximated the risk that is observed with high viral load (20,000 IU/mL), the researchers noted. In contrast, a low HBcrAg level was associated with a risk similar to that of minimal risk carriers (annual incidence rate, 0.10%; 95% CI, 0.04%-0.24%).

“To the best of our knowledge, this is the first study to report HBcrAg level as an independent viral biomarker to stratify hepatocellular risks in a large number of patients with intermediate viral load,” the researchers commented. Among the study limitations, 412 patients received antiviral therapy during follow-up. “This is a retrospective cohort study including Asian HBV patients with genotype B or C infection,” the investigators added. “It is unclear whether this finding could be extrapolated to populations with other HBV genotype infections. Nonetheless, we had a sound cohort, as several HBsAg-related clinical findings based on our cohort have already been validated by other prospective cohort studies, implying that our data were unlikely to be biased by the study design.”

Funders included National Taiwan University Hospital, the Ministry of Science and Technology, Executive Yuan in Taiwan, and National Health Research Institutes. The researchers reported having no conflicts of interest.

SOURCE: Tseng T-C et al. Gastroenterology. 2019 Aug 27. doi: 10.1053/j.gastro.2019.08.028.

Next Article: