Secondary Logo

Journal Logo

ARTICLE: LIVER

Hepatitis B Core-Related Antigen Stratifies the Risk of Liver Cancer in HBeAg-Negative Patients With Indeterminate Phase

Tseng, Tai-Chung MD, PhD1,2,3; Hosaka, Tetsuya MD4; Liu, Chun-Jen MD, PhD1,2,5; Suzuki, Fumitaka MD, PhD4; Hong, Chun-Ming MD1,6; Kumada, Hiromitsu MD, PhD4; Yang, Wan-Ting MPH2; Hsu, Chen-Yang MD, PhD7; Su, Tung-Hung MD, PhD1,2; Yang, Hung-Chih MD, PhD1,5,8; Liu, Chen-Hua MD, PhD1,2; Chen, Pei-Jer MD, PhD1,2,5; Chen, Hsiu-Hsi PhD7; Kao, Jia-Horng MD, PhD1,2,3,5

Author Information
The American Journal of Gastroenterology: May 2022 - Volume 117 - Issue 5 - p 748-757
doi: 10.14309/ajg.0000000000001691

Abstract

INTRODUCTION

Chronic hepatitis B virus (HBV) infection continues to be an important health issue worldwide, with a recent global estimate suggesting that more than 292 million individuals tested positive for hepatitis B surface antigen (HBsAg) (1). Patients with chronic hepatitis B (CHB) are at an increased risk of developing hepatocellular carcinoma (HCC) (2). It is important to identify high-risk patients and initiate timely antiviral therapy to lower the risk (2).

The natural history of CHB includes 2 distinct phases in hepatitis B e antigen (HBeAg)-negative patients: inactive CHB and immune-active CHB. According to the American Association for the Study of Liver Diseases (AASLD) guidelines (3), patients with inactive CHB (HBV DNA <2,000 IU/mL and alanine aminotransferase [ALT] < the upper limits of normal [ULN]) experience a low HCC risk and do not require antiviral therapy, whereas patients with immune-active CHB (HBV DNA ≥2,000 IU/mL and ALT ≥2× ULN) are recommended to receive antiviral therapy due to their high HCC risk. Approximately 40%–50% of HBeAg-negative patients fall outside the criteria and are categorized as indeterminate patients (4,5).

Heterogenous clinical outcomes have been shown in indeterminate patients, and frequent clinical follow-up is recommended (5–8). This recommendation becomes impractical in the face of global COVID-19 pandemic. This scenario highlights the need for a biomarker to stratify the HCC risk and to optimize the clinical management. However, none of the traditional biomarkers have been shown to predict HCC in indeterminate patients except older age (5).

Hepatitis B core-related antigen (HBcrAg) quantification is an emerging biomarker for evaluating the transcriptional activity of covalently closed circular DNA (9,10). Cohort studies from different research groups have shown that serum HBcrAg levels could serve as an HCC predictor in patients with CHB (11–13). For example, our data from a Taiwanese cohort study has already shown that a HBcrAg level of 10,000 U/mL could be used to stratify the HCC risk in HBeAg-negative patients with intermediate viral load (HBV DNA between 2,000 and 20,000 IU/mL) and normal ALT levels (11). However, this population is only approximately one-third of the indeterminate patients. It remains unclear whether this biomarker still works when we expand the study population into the whole indeterminate patients (14).

In this study, we first adopted the same retrospective Taiwanese cohort including treatment-naive HBeAg-negative patients without liver cirrhosis to explore the HCC risk in indeterminate patients. Second, we investigated whether viral markers including HBcrAg level could be used to stratify their HCC risk. Last, we validated our findings in an independent Japanese cohort.

MATERIALS AND METHODS

Patient cohort

The discovery cohort was based on patients from the “Elucidation of Risk fActors for DIsease Control or Advancement in Taiwanese hEpatitis B carriers” study (11,15–18). The inclusion and exclusion criteria are shown in Supplementary Digital Content (see Supplementary Figure 1A, https://links.lww.com/AJG/C452). In brief, we excluded 411 patients with liver cirrhosis at baseline, 412 patients who received antiviral therapy before HCC development or the end of follow-up, and 516 HBeAg-positive patients. Finally, 2,150 treatment-naive, HBeAg-negative carriers without liver cirrhosis were included. In the sensitivity analysis, we recalled 70 indeterminate patients who received antiviral therapy during the follow-up. All patients gave informed consent as approved by the research ethics committee of National Taiwan University Hospital.

Supplementary Digital Content (see Supplementary Figure 1B, https://links.lww.com/AJG/C453) shows how the Japanese patients from the validation cohort were enrolled, which follows the inclusion and exclusion criteria of the discovery cohort (19). Finally, 1,312 treatment-naïve, HBeAg-negative carriers without liver cirrhosis were included. Informed consent for this study was obtained from each patient. The study protocol complied with the ethical guidelines of the Declaration of Helsinki and the ethical guidelines for medical and health research involving human subjects of the Ministry of Health, Labour and Welfare in Japan. The study was approved by the Toranomon Hospital Ethics Committee (ID: 2223).

Data collection and viral marker assay

Data on baseline serological markers (HBsAg, HBeAg, anti-HBe, anti-HCV, and anti-HDV), liver biochemical test results, and alpha-fetoprotein levels were collected. Blood tests, including liver biochemical tests, measurement of alpha-fetoprotein levels, and abdominal ultrasonography were performed for HCC surveillance every 6 months at least in both cohorts. Serum samples at each visit were collected and stored at −20°C (discovery cohort) or −80°C (validation cohort), which were used to determine HBV genotypes and levels of HBV DNA, HBsAg, and HBcrAg retrospectively (see Supplementary Material, https://links.lww.com/AJG/C464).

Definition of normal ALT, liver cirrhosis, and HCC

There are different definitions of normal ALT among different clinical practice guidelines (see Supplementary Table 1, https://links.lww.com/AJG/C465). In this study, the upper limit of normal ALT (35 U/L for male and 25 U/L for female individuals) defined by the AASLD guideline was adopted to present our main results (3). Inactive CHB or HBeAg-negative chronic HBV infection was defined as HBV DNA <2,000 IU/mL + ALT < 1× ULN. Immune-active CHB was defined as patients with HBV DNA level ≥ 2,000 IU/mL accompanied by ALT ≥ 2× ULN (3). We also analyzed our data using the definitions of European Association for the Study of the Liver (EASL) and The Asian Pacific Association for the Study of the Liver (APASL), respectively (20,21).

Liver cirrhosis was diagnosed via histology and ultrasonographic findings together with or without clinical features such as ascites thrombocytopenia, gastroesophageal varices (in both cohorts), or laparoscopy (in validation cohort) (22,23). HCC was diagnosed either by histology/cytology results or by typical image findings (arterial enhancement and venous washout by contrast-enhanced CT or MRI scanning) in hepatic nodules larger than 1 cm (24).

Statistical analysis

The mean and SD values were calculated for continuous variables and percentages were calculated for categorical variables.

The clinical follow-up started at the time of enrollment. The person-years were censored on the date of identifying HCC, death, or the last date of follow-up (June 30, 2017, in the discovery cohort and December 31, 2019, in the validation cohort), whichever came first. The cumulative incidence stratified by different variables was analyzed by the Kaplan-Meier curve analysis and the log-rank test. We adopted the Cox proportional hazards regression model to assess the crude and adjusted hazard ratios (HR) of each variable (11,15,16).

We adopted receiver operating characteristic (ROC) curve analysis to compute the area under the ROC curves for different factors in predicting HCC development within 10 years and 15 years, respectively. The optimal HBcrAg cutoff was determined using the Youden index (sensitivity plus the specificity minus one) (11).

We conducted a sensitivity analysis in the discovery cohort by reincluding indeterminate patients receiving antiviral treatment during the follow-up. An inverse probability of treatment weighting (IPTW) approach in conjunction with the marginal structural model was applied to control the potential effect of time-dependent antiviral treatment (see Supplementary Material, https://links.lww.com/AJG/C464) (11).

Statistical significance of all tests was defined as P < 0.05 by 2-tailed tests. All the analyses were performed using Stata statistical software (version 10.0; Stata, College Station, TX) except that IPTW-related analyses were performed using SAS v. 9.4 statistical software.

RESULTS

Comparison of baseline characteristics between discovery and validation cohorts

Table 1 compares the baseline characteristics between 2,150 patients from the discovery cohort and 1,312 patients from the validation cohort. The patients in both cohorts were HBeAg-negative and noncirrhotic at baseline and remained treatment free during the follow-up. Patients in the discovery cohort were younger (42.4 vs 45.4 years) and had a higher proportion of fibrosis-4 index (FIB-4) <1.29 (72.76% vs 65.93%) than those in the validation cohort. By contrast, there were more patients with HBV DNA level <2,000 IU/mL (59.22% vs 49.21%) and HBcrAg <10,000 U/mL (84.45% vs 58.51%) in the validation cohort than in the discovery cohort. The dominant viral strains were genotype B in the discovery cohort (83.70%) and genotype C in the validation cohort (65.86%).

T1
Table 1.:
Comparison of baseline characteristics and follow-up data between 2,150 patients from the discovery cohort and 1,312 patients from the validation cohort

Comparison of HCC incidence between discovery and validation cohorts

The mean follow-up duration was 15.88 ± 4.62 years and 12.07 ± 8.51 years (mean ± SD) for the discovery and validation cohorts (P = 0.018), respectively, and 144 and 33 patients from the discovery and validation cohorts, respectively, developed HCC (Table 1). There was a higher cumulative incidence of HCC in the discovery cohort compared with that in the validation cohort (P < 0.001, see Supplementary Figure 2, https://links.lww.com/AJG/C454).

Cumulative incidence of HCC stratified by levels of HBV DNA and ALT in the discovery cohort

We first categorized the 2,150 patients from the discovery cohort into 8 groups by HBV DNA levels of 2,000 IU/mL and 20,000 IU/mL and ALT of 1× and 2× the ULN (Table 1) according to the AASLD guidelines (3).

The cumulative incidence of HCC of the 8 groups is shown in Figure 1a. In brief, group 0 (inactive CHB, N = 824) had the lowest HCC risk, whereas the highest HCC risk was shared by group 6 (HBV DNA level ≥20,000 IU/mL + ALT between 1× and 2× ULN, N = 161) and group 7 (immune-active CHB, N = 175). The HCC risk of the other groups (from groups 1 to 5) was heterogeneous, and all of them were grouped as indeterminate patients (N = 990). The detailed risk of each group is summarized in supplementary Digital Content (see Table 2, https://links.lww.com/AJG/C465).

F1
Figure 1.:
HCC risk stratification models. (a) In 2,150 HBeAg-negative patients from the discovery cohort, HCC risk is categorized into 8 groups by HBV DNA and AASLD-defined ALT levels. In 990 indeterminate patients (groups 1–5), HBcrAg serves as a better HCC predictor than HBV DNA and HBsAg within (b) 10 years (24 patients developed HCC) and (c) 15 years of follow-up (49 patients developed HCC) using ROC curve analysis. Their HCC risk could be stratified by (d) HBcrAg level at baseline and (e) HBcrAg kinetic defined by the levels at baseline and year 3 of follow-up: persistently low group (both <10,000 U/mL), fluctuated group (once elevated to > 10,000 U/mL), and persistently high group (both >10,000 U/mL). AASLD, American Association for the Study of Liver Diseases; ALT, alanine transaminase; HBcrAg; Hepatitis B core-related antigen; HBeAg, hepatitis B e antigen; HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; HCC, hepatocellular carcinoma; ROC, receiver-operating characteristic.

HBcrAg as an independent HCC predictor in indeterminate patients

Serum levels of HBcrAg, HBV DNA, and HBsAg were compared for predicting HCC prediction over 10 years of follow-up (Figure 1b) and 15 years of follow-up (Figure 1c) using area under the ROC curves. We found HBcrAg served as a better HCC predictor than HBV DNA and HBsAg levels in both settings.

The Youden index results showed that HBcrAg level of 10,000 U/mL was an optimal cutoff due to the highest value (sensitivity: 75.34%, specificity: 62.81%, see Supplementary Table 3, https://links.lww.com/AJG/C465).

The HBcrAg cutoff of 10,000 U/mL was tested for HCC risk stratification in groups 1–5, respectively. We found patients with an HBcrAg level above the cutoff were associated with a higher risk in most of the groups when compared with those below the cutoff despite a small number in each group (see Supplementary Figure 3A–E, https://links.lww.com/AJG/C455, https://links.lww.com/AJG/C456, https://links.lww.com/AJG/C457, https://links.lww.com/AJG/C458, https://links.lww.com/AJG/C459).

When stratifying all the indeterminate patients using an HBcrAg level of 10,000 U/mL, 396 patients (40%) had a high HBcrAg level and were associated with an increased risk of HCC when compared with patients who had a low level of HBcrAg. The HR was 4.47 (95% confidence interval [CI]: 2.62–7.63, Figure 1d), and the 10-year HCC cumulative incidence was 0.51% (95% CI: 0.16%–1.57%) and 5.33% (3.51%–8.06%) for low HBcrAg and high HBcrAg groups, respectively. It remained an independent risk factor using multivariable analysis (Table 2).

T2
Table 2.:
Univariable and multivariable analyses of factors associated with HCC development in 990 indeterminate patients from the discovery cohort by Cox proportional hazards regression model

HBcrAg kinetics associated with HCC development

In addition to baseline HBcrAg levels, we determined the HBcrAg levels at year 3 of follow-up in 883 patients with available serum samples: 479 patients (54.3%) had persistently low HBcrAg level (<10,000 U/mL at both time points); 168 patients (19.0%) had persistently high HBcrAg level (≥10,000 U/mL); 236 patients (26.7%) had fluctuated HBcrAg levels with HBcrAg levels increasing in 52 and decreasing in 184 patients. HCC risk was the lowest in the patients who maintained HBcrAg <10,000 U/mL, followed by those with fluctuated HBcrAg levels (HR: 3.98, 95% CI: 1.80–8.78), whereas those who maintained ≥10,000 U/mL had the highest HCC risk (HR: 10.18, 95% CI: 4.89–21.19) (Figure 1e).

HBcrAg to stratify HCC risk in indeterminate patients defined by APASL and EASL guidelines, respectively

We further identified 930 and 1,112 indeterminate patients from the discovery cohort according to the EASL and APASL guidelines, respectively (see Supplementary Table 1, https://links.lww.com/AJG/C465). Serum HBcrAg level of 10,000 U/mL was also shown to split HCC risk in both indeterminate patients (Figure 2a,b). The age-adjusted and sex-adjusted HR of HBcrAg ≥10,000 U/mL (vs <10,000 U/mL) were 5.43 (95% CI: 3.16–9.34) and 5.54 (95% CI: 3.39–9.06) for indeterminate patients defined by the EASL and APASL guidelines, respectively.

F2
Figure 2.:
Baseline HBcrAg of 10,000 U/mL stratifies the HCC risk in indeterminate patients defined by (a) EASL and (b) APASL guidelines in the discovery cohort. APASL, The Asian Pacific Association for the Study of the Liver; EASL, European Association for the Study of the Liver; HBcrAg, Hepatitis B core-related antigen; HCC, hepatocellular carcinoma.

Internal validation by 2 subgroup analyses was as follows: 905 patients remaining indeterminate for 1 year and 570 patients with FIB-4 <1.29.

The first subgroup analysis includes 905 indeterminate patients who did not transition to immune active during the first year of follow-up (see Supplementary Figure 4A, https://links.lww.com/AJG/C460). The second subgroup analysis includes 570 indeterminate patients with baseline FIB-4 <1.29 (15), an indicator for minimal fibrosis (see Supplementary Figure 4B, https://links.lww.com/AJG/C461). HBcrAg level >10,000 U/mL (vs <10,000 U/mL) was shown to be associated with a higher HCC risk in both analyses (Both P < 0.001).

Sensitivity analysis by reincluding indeterminate patients receiving antiviral treatment during the follow-up

We recalled 70 HBeAg-negative indeterminate patients who received antiviral treatment during the follow-up (see Supplementary Figure 1A, https://links.lww.com/AJG/C452). Data were analyzed using structural marginal Cox proportional hazards regression model in conjunction with IPTW to address the issues of time-dependent status of antiviral treatment. The results consistently supported that HBcrAg level of 10,000 U/mL was effective to stratify the HCC risk (see Supplementary Table 4, https://links.lww.com/AJG/C465) in 1,060 indeterminate patients with or without antiviral treatment during the follow-up.

External validation of HCC risk stratified by HBcrAg in indeterminate patients of Japanese cohort

We aimed to validate our findings in a Japanese cohort of 1,312 patients. All patients were categorized into 8 groups according to the AASLD guidelines, and the HBcrAg levels are summarized in Supplementary Digital Content (see Supplementary Table 5, https://links.lww.com/AJG/C465). The risk pattern is similar to what we have observed in the discovery cohort except groups 2 and 5 (see Supplementary Figure 5A, https://links.lww.com/AJG/C462). Notably, the HCC risk of groups 6 and 7 was also close; thus, we defined the patients of groups 1–5 as indeterminate patients.

Among the 574 indeterminate patients, we first showed a positive relationship between HBcrAg levels and HCC development (P for trend< 0.001), and HBcrAg served as a better predictor for HCC development than HBsAg and HBV DNA levels (see Supplementary Figure 5B, https://links.lww.com/AJG/C463). The optimal cutoff of HBcrAg was approximately 10,000 U/mL using Youden index analysis (see Supplementary Table 6, https://links.lww.com/AJG/C465), which is consistent with the value adopted by the discovery cohort.

We then categorized the indeterminate patients using HBcrAg of 10,000 U/mL. There was a higher HCC risk in the high HBcrAg group compared with that in the low HBcrAg group (Figure 3a, HR: 8.20; 95% CI: 2.78–24.14). We also redefined the indeterminate patients by the EASL and APASL guidelines, respectively (see Supplementary Table 1, https://links.lww.com/AJG/C465). The HCC risk was consistently stratified by HBcrAg level of 10,000 U/mL in both analyses (Figure 3b,c).

F3
Figure 3.:
Baseline HBcrAg of 10,000 U/mL stratifies the HCC risk in indeterminate patients of the validation cohort defined by (a) AASLD, (b) EASL, and (c) APASL guidelines in the validation cohort. APASL, The Asian Pacific Association for the Study of the Liver; EASL, European Association for the Study of the Liver; HBcrAg, Hepatitis B core-related antigen; HCC, hepatocellular carcinoma.

Restratification of HCC risk using HBV DNA, ALT, and HBcrAg in HBeAg-negative patients

We aimed to simplify the risk stratification model using ALT, HBV DNA, and HBcrAg in both cohorts. All the HBeAg-negative patients were categorized into 4 groups: group I: inactive CHB; group II: indeterminate CHB with HBcrAg level <10,000 U/mL; group III: indeterminate CHB with HBcrAg level ≥10,000 U/mL; and group IV: patients with immune-active CHB and those with comparable risk (groups 6 and 7). We found HCC risk was similar between groups I and lI, both of which were lower than the high HCC risk of groups III and IV either in the discovery cohort (Figure 4a) or in the validation cohort (Figure 4b). The multivariable analysis, including age and sex as adjusting variables, showed a similar pattern in both cohorts (see Supplementary Table 7, https://links.lww.com/AJG/C465). Therefore, we pooled groups I and II as low-risk patients and groups III and IV as high-risk patients. The high-risk group, when compared with the low-risk group, was associated with an increased HCC risk with a HR of 6.70 (95% CI: 4.56–9.85) and 9.24 (95% CI: 4.39–19.46) for the discovery and validation cohorts, respectively.

F4
Figure 4.:
Reclassify HCC risk using levels of HBV DNA, ALT, and HBcrAg in HBeAg-negative patients in the (a) discovery cohort (N = 2,150) and the (b) validation cohort (N = 1,312). ALT, alanine transaminase; HBcrAg, Hepatitis B core-related antigen; HBeAg, hepatitis B e antigen; HCC, hepatocellular carcinoma.

DISCUSSION

The natural history of the HBeAg-negative patients with indeterminate phase remains variable; thus, the optimal management is unclear (25). Our study showed that the long-term HCC risk of patients with AASLD-defined indeterminate CHB lies between inactive CHB and immune-active CHB, except for patients with viral load ≥20,000 IU/mL + 1–2× ULN of ALT. Serum HBcrAg level is a better HCC predictor than other viral markers in indeterminate CHB patients. In particular, HBcrAg level of 10,000 U/mL was shown to stratify HCC risk in indeterminate CHB patients defined by AASLD, EASL, or APASL guidelines. HBcrAg level or kinetic remained as an HCC predictor in different subgroup analyses. Finally, the role of HBcrAg was validated in an independent cohort of Japanese patients, which provides strong evidence that HBcrAg level of 10,000 U/mL is a useful biomarker for optimizing the management.

To the best of our knowledge, we are the first group to show that HBcrAg is effective to complement HBV DNA in categorizing the HCC risk (11). Initially, we found that HBcrAg of 10, 000 U/mL, a biomarker to define minimal disease activity (26), is effective to stratify HCC risk in HBeAg-negative patients with intermediate viral load. In this study, we expanded the application of this biomarker to the whole HBeAg-negative patients with indeterminate phase not only in our original cohort but also in an independent Japanese cohort. When stratifying the indeterminate patients by HBcrAg of 10,000 U/mL, the HCC risk was similar between the low HBcrAg group and the inactive CHB patients, whereas the high HBcrAg group had HCC risk close to the immune-active CHB patients. In other words, HBeAg-negative patients could be simply divided into 2 different groups by HBV DNA, ALT, and HBcrAg levels. This finding is useful because it may guide the physicians to prioritize CHB patients for HCC surveillance, especially when facing the global COVID-19 pandemic, which is associated with limited medical resources and concerns over potential nosocomial transmission (27).

There are some differences between the 2 cohorts. First, baseline characteristics are very different (Table 1). In addition to distinct HBV genotypes, the Taiwanese patients are younger and have lower FIB-4 levels but have higher levels of HBV DNA and HBcrAg than Japanese patients. It is likely that more active viral replication in Taiwanese patients outperforms the other favored host factors, which leads to a higher risk of HCC compared with that in Japanese patients. However, when the patients were stratified by HBV-DNA, ALT, and HBcrAg levels, the HR (high risk vs low risk) was higher in the Japanese patients than in the Taiwan patients (9.4 vs 6.7). We believe the age, genotype, and fibrosis stage may play a role in the stratified analysis. Second, we noted the HCC incidence of group 2 and group 5 were close to groups 6 and 7 in the Japanese cohort, which is different from the Taiwanese cohort (see Supplementary Figure 5A, https://links.lww.com/AJG/C462). To be noted, the HCC risk increases rapidly after 9 and 18 years of follow-up in group 2 and group 5, respectively, and the corresponding number at risk then was 29 and 31. We believe a small number of the patients may lead to an inaccurate estimation and enrolling more patients is needed to address this issue.

There are 3 unique findings in this study. First, it is challenging to define a valid HBcrAg cutoff, although HBcrAg has been shown as an HCC predictor (11–13). In this study, HBcrAg of 10,000 U/mL was first identified in Taiwanese patients infected by genotype B virus and was validated in Japanese patients infected by genotype C virus. To the best of our knowledge, this is the first study to define an HBcrAg cutoff with external validation, especially by a different ethnic population infected with a distinct HBV genotype. Second, we excluded the highly viremic patients with ALT 1–2× ULN from the indeterminate group because their HCC risk was similar to those with ≥2× ULN of ALT in both cohorts. This finding echoed the data from a recent clinical trial showing liver fibrosis progressed in approximately 50% of the highly viremic patients with minimally elevated ALT after a 3-year follow-up (28). Because antiviral treatment has been shown to reduce the risk of fibrosis progression in this clinical trial, these highly viremic patients with mildly elevated ALT are recommended for antiviral treatment (25). Finally, both cohorts enrolled patients before the antiviral therapy was introduced into clinical practice. That is why we have a chance to explore the long-term disease progression in HBeAg-negative patients with indeterminate phase.

There are a few limitations in this study. First, we initially excluded the patients receiving antiviral therapy during the follow-up, which could introduce bias. This issue has been addressed using IPTW to simulate posttreatment follow-up, which is useful in controlling the disease modification effect elicited by antiviral treatment. Second, ALT varies rapidly; hence, indeterminate patients may fulfill the treatment criteria after a short period of time. We thus excluded 85 indeterminate patients who developed HBeAg-negative hepatitis within the first year of follow-up. In this subgroup analysis, HBcrAg level continued to serve as a useful HCC predictor. Finally, both cohorts included Asian patients only, and we need more data from White patients with genotype A or D infection to validate our findings.

In summary, baseline HBcrAg level of 10,000 U/mL was shown to be a valuable biomarker to stratify the HCC risk in HBeAg-negative patients with indeterminate CHB in both Asian cohorts. The HCC risk of HBeAg-negative patients could be simply reclassified into 2 groups by serum HBV DNA, ALT, and HBcrAg levels. This validated finding indicates that HBcrAg level of 10,000 U/mL serves as a novel biomarker to guide the clinical management of indeterminate CHB.

CONFLICTS OF INTEREST

Guarantor of the article: Tetsuya Hosaka, MD and Jia-Horng Kao, MD, PhD.

Specific author contributions: Study concept and design: T.-C.T. and J.-H.K. Acquisition of data: W.-T.Y., C.-J.L., H.-C.Y., T.-H.S., C.-H.L., P.-J.C., J.-H.K., T.H., F.S., and H.K. Analysis and interpretation of data: W.-T.Y., T.-C.T., T.H., and J.-H.K. Drafting of the manuscript: T.-C.T. and T.H. Critical review of the manuscript for important intellectual content: J.-H.K. Statistical analysis: C.-Y.H., W.-T.Y., H.-H.C., T.H., and T.-C.T. Obtained funding: T.-C.T. and J.-H.K. Technical or material support: C.-M.H. and T.-C.T. Study supervision: J.-H.K. All the authors approved the final version of the manuscript.

Financial support: The work from Taiwanese research team was supported by the grants from National Taiwan University Hospital (106-N3626, 107-S3816, and 107-N4041), the Ministry of Science and Technology, Executive Yuan, Taiwan (MOST 105-2314-B-303-008 and MOST 106-2314-B-002-136), National Health Research Institutes (NHRI-EX108-10807BC), and Gilead Sciences (IN-TW-988-5987). The work from Japanese research team was supported in part by the Program for Basic and Clinical Research on Hepatitis, the Japan Agency for Medical Research and Development (AMED) (No. 19HC1001) (No. JP21fk0210084).

Potential competing interests: T.-C.T. has served on speaker's bureaus for Fujirebio, Bristol-Myers Squibb, and Gilead Sciences and received grant support from Gilead Sciences. H.K. served as a speaker for Gilead Sciences, AbbVie Inc., MSD K.K., Eisai Co., Ltd., and Dainippon Sumitomo Pharma.

Study Highlights

WHAT IS KNOWN

  • ✓ The optimal management is unknown for hepatitis B e antigen-negative patients with indeterminate phase.
  • ✓ Serum level of hepatitis B virus (HBV) core-related antigen (HBcrAg) predicts hepatocellular carcinoma (HCC) development.

WHAT IS NEW HERE

  • ✓ Serum HBcrAg levels serve as a better HCC predictor in indeterminate patients compared with serum HBV DNA or hepatitis B surface antigen levels.
  • ✓ Serum HBcrAg level of 10,000 U/mL is an effective cutoff to stratify the HCC risk in indeterminate patients.
  • ✓ Combing this HBcrAg cutoff with HBV DNA and alanine aminotransferase levels reclassifies the hepatitis B e antigen-negative patients into high-risk and low-risk groups for HCC development, which simplifies the clinical management.

ACKNOWLEDGEMENTS

We thank Fujirebio Company for providing quantitative HBcrAg kits. We thank the staff of the Department of Medical Research, National Taiwan University Hospital, for the Integrated Medical Database (NTUH-iMD), who provided the follow-up data. We also thank Cancer Registry, Cancer Administration and Coordination Center, NTUH, to provide the cancer registration data to confirm our diagnosis. Finally, the grant supports from Gilead Sciences and Liver Disease Prevention & Treatment Research Foundation, Taiwan, are also appreciated.

REFERENCES

1. Nguyen MH, Wong G, Gane E, et al. Hepatitis B virus: Advances in prevention, diagnosis, and therapy. Clin Microbiol Rev 2020;33:33.
2. Tseng TC, Kao JH. Elimination of hepatitis B: Is it a mission possible? BMC Med 2017;15:53.
3. Terrault NA, Lok ASF, McMahon BJ, et al. Update on prevention, diagnosis, and treatment of chronic hepatitis B: AASLD 2018 hepatitis B guidance. Hepatology 2018;67:1560–99.
4. Di Bisceglie AM, Lombardero M, Teckman J, et al. Determination of hepatitis B phenotype using biochemical and serological markers. J Viral Hepat 2017;24:320–9.
5. Huang DQ, Li X, Le MH, et al. Natural history and hepatocellular carcinoma risk in untreated chronic hepatitis B patients with indeterminate phase. Clin Gastroenterol Hepatol 2021 (doi:10.1016/j.cgh.2021.01.019).
6. Zhou K, Wahed AS, Cooper S, et al. Phase transition is infrequent among north American adults with e-antigen-negative chronic hepatitis B and low-level viremia. Am J Gastroenterol 2019;114:1753–63.
7. Bonacci M, Lens S, Mariño Z, et al. Anti-viral therapy can be delayed or avoided in a significant proportion of HBeAg-negative Caucasian patients in the Grey Zone. Aliment Pharmacol Ther 2018;47:1397–408.
8. Lee HW, Kim SU, Baatarkhuu O, et al. Progression of untreated minimally active chronic HBV infection compared to inactive infection. Clin Gastroenterol Hepatol 2019;17:2808–10. e2.
9. Wong DK, Seto WK, Cheung KS, et al. Hepatitis B virus core-related antigen as a surrogate marker for covalently closed circular DNA. Liver Int 2017;37:995–1001.
10. Suzuki F, Miyakoshi H, Kobayashi M, et al. Correlation between serum hepatitis B virus core-related antigen and intrahepatic covalently closed circular DNA in chronic hepatitis B patients. J Med Virol 2009;81:27–33.
11. Tseng TC, Liu CJ, Hsu CY, et al. High level of hepatitis B core-related antigen associated with increased risk of hepatocellular carcinoma in patients with chronic HBV infection of intermediate viral load. Gastroenterology 2019;157:1518–29. e3.
12. To WP, Mak LY, Wong DK, et al. Hepatitis B core-related antigen levels after HBeAg seroconversion is associated with the development of hepatocellular carcinoma. J Viral Hepat 2019;26:1473–80.
13. Tada T, Kumada T, Toyoda H, et al. HBcrAg predicts hepatocellular carcinoma development: An analysis using time-dependent receiver operating characteristics. J Hepatol 2016;65:48–56.
14. Wu JW, Kao JH, Tseng TC. Three heads are better than two: Hepatitis B core-related antigen as a new predictor of hepatitis B virus-related hepatocellular carcinoma. Clin Mol Hepatol 2021;27:524–34.
15. Tseng TC, Liu CJ, Su TH, et al. Fibrosis-4 index helps identify HBV carriers with the lowest risk of hepatocellular carcinoma. Am J Gastroenterol 2017;112:1564–74.
16. Tseng TC, Liu CJ, Yang HC, et al. High levels of hepatitis B surface antigen increase risk of hepatocellular carcinoma in patients with low HBV load. Gastroenterology 2012;142:1140–4.
17. Tseng TC, Liu CJ, Yang WT, et al. Serum hepatitis B core-related antigen level stratifies risk of disease progression in chronic hepatitis B patients with intermediate viral load. Aliment Pharmacol Ther 2021;53:908–18.
18. Tseng TC, Liu CJ, Yang HC, et al. Serum hepatitis B surface antigen levels help predict disease progression in patients with low hepatitis B virus loads. Hepatology 2013;57:441–50.
19. Kobayashi M, Hosaka T, Suzuki F, et al. Seroclearance rate of hepatitis B surface antigen in 2,112 patients with chronic hepatitis in Japan during long-term follow-up. J Gastroenterol 2014;49:538–46.
20. European Association for the Study of the Liver. Electronic address eee, European Association for the Study of the L. EASL 2017 Clinical Practice Guidelines on the management of hepatitis B virus infection. J Hepatol 2017;67:370–98.
21. Sarin SK, Kumar M, Lau GK, et al. Asian-Pacific clinical practice guidelines on the management of hepatitis B: A 2015 update. Hepatol Int 2016;10:1–98.
22. Tseng TC, Liu CJ, Yang HC, et al. Higher proportion of viral basal core promoter mutant increases the risk of liver cirrhosis in hepatitis B carriers. Gut 2015;64:292–302.
23. Tseng TC, Liu CJ, Chang CT, et al. HEV superinfection accelerates disease progression in patients with chronic HBV infection and increases mortality in those with cirrhosis. J Hepatol 2020;72:1105–11.
24. Heimbach JK, Kulik LM, Finn RS, et al. AASLD guidelines for the treatment of hepatocellular carcinoma. Hepatology 2018;67:358–80.
25. Jeng WJ, Lok AS. Should treatment indications for chronic hepatitis B be expanded? Clin Gastroenterol Hepatol 2021;19:2006–14.
26. Testoni B, Lebossé F, Scholtes C, et al. Serum hepatitis B core-related antigen (HBcrAg) correlates with covalently closed circular DNA transcriptional activity in chronic hepatitis B patients. J Hepatol 2019;70:615–25.
27. Boettler T, Newsome PN, Mondelli MU, et al. Care of patients with liver disease during the COVID-19 pandemic: EASL-ESCMID position paper. JHEP Rep 2020;2:100113.
28. Hsu YC, Chen CY, Chang IW, et al. Once-daily tenofovir disoproxil fumarate in treatment-naive Taiwanese patients with chronic hepatitis B and minimally raised alanine aminotransferase (TORCH-B): A multicentre, double-blind, placebo-controlled, parallel-group, randomised trial. Lancet Infect Dis 2021;21:823–33.

Supplemental Digital Content

© 2022 by The American College of Gastroenterology