Review article: hepatitis B core‐related antigen (HBcrAg): an emerging marker for chronic hepatitis B virus infection

Chronic hepatitis B (CHB) cannot be completely eradicated due to the presence of covalently closed circular DNA (cccDNA) in the nuclei of infected hepatocytes. While quantification of intrahepatic cccDNA requires liver biopsies, serological markers can be non‐invasive alternatives to reflect intrahepatic viral replicative activity. Recently, hepatitis B core‐related antigen (HBcrAg) has been advocated as a novel serum marker for disease monitoring and prognostication of CHB.


| INTRODUCTION
Chronic hepatitis B (CHB) affects approximately 240 million persons worldwide, 1 with an estimated 15%-40% progressing to cirrhosis, decompensation and/or hepatocellular carcinoma (HCC). 2 Currently, hepatitis B virus (HBV) cannot be completely eradicated due to the presence of covalently closed circular DNA (cccDNA) in the nuclei of infected hepatocytes. [3][4][5] Therefore, the present goal of monitoring is to ensure a high degree of virological suppression, ideally with hepatitis B surface antigen (HBsAg) seroclearance, leading to biochemical remission, histological improvement and reduction of the risk of complications. 6,7 Although liver biopsy for the quantification of intrahepatic cccDNA and intrahepatic HBV DNA remains the most accurate measurement for viral reservoir, it is limited by its invasive nature, the potential for sampling error and the lack of a standardised assay. Noninvasive serological markers can be used as surrogate markers of intrahepatic viral replicative activity. Established serological markers include serum HBV DNA levels and serum HBsAg titres, both of which have been shown to predict the risk of cirrhosis and HCC. [8][9][10][11] However, as cirrhosis and HCC can still occur in many patients despite undetectable HBV DNA 12 and HBsAg seroclearance, 13,14 any promising markers with predictive value in these scenarios are very much welcome. Moreover, given that almost all patients treated with antiviral therapy have undetectable HBV DNA, more accurate biomarkers for risk stratification are needed. Recently, linearised HBsAg (HQ-HBsAg) assay achieving an even lower limit of detection for HBsAg 15 has been shown to correlate with serum HBV DNA and HBsAg levels, which makes it a potential marker for stratifying patients with HBsAg seroclearance. 16 Another new marker, the hepatitis B core-related antigen (HBcrAg), has also been advocated as a serum marker for disease monitoring and prognostication of CHB. In this review, the clinical application of HBcrAg in CHB patients based on its virological features, the distinctive profiles in different disease stages and the profile under anti-viral treatment will be discussed.

| HBV REPLICATION
After initial viral entry into the infected hepatocyte and formation of stable cccDNA minichromosomes in the nucleus, the first step of HBV replication starts with HBV transcription. Five major messenger RNAs (mRNAs) species, namely the core mRNA/pregenomic RNA (3.5 kb), precore mRNA (3.5 kb), LHBs mRNA (2.4 kb), MHBs and SHBs mRNA (2.1 kb) and X mRNA (0.9 kb), are generated from the minus strand of the genome. These mRNAs encode for various structural and nonstructural proteins. The pregenomic RNA acts as a precursor for the synthesis of viral DNA genome by reverse transcription. The nascent viral genome, encapsidated by the hepatitis B core protein (HBcAg), is then packaged by the hepatitis B surface (HBs) proteins in the endoplasmic reticulum. The mature virions are then secreted into the blood stream, giving rise to a large number of progeny virions. The measurement of serum HBV DNA therefore can provide an estimate of the viral replicative activity, and widely used as a marker of anti-viral efficacy. However, nucleos(t)ide analogues (NAs) only act on limited steps of the viral replication cycle, and production of viral intermediate proteins may not be affected significantly. Therefore, measurement of viral proteins can be useful in monitoring HBV activities, especially in patients receiving NAs when HBV DNA levels are undetectable. One such candidate is HBsAg, which is found in mature virions as well as HBV DNA-negative empty particles, in either spherical or filamentous forms. Another candidate is HBcrAg, which consists of 3 species of related proteins sharing an identical 149 amino acid sequence: HBcAg, hepatitis B e antigen (HBeAg), and a truncated 22 kDa precore protein (p22Cr). HBcAg is the structural component of the viral capsids, which can be found in mature virions. HBeAg, a nonstructural HBV protein, is the N-terminal processed product of the precore protein. Like HBeAg, p22cr is also a processed product of the precore protein, but with protein processing at both the N-and C-terminals. 17 The viral replication cycle and the origins of HBV DNA, HBsAg, HBeAg and HBcrAg are illustrated in Figure 1.

| HBcrAg MEASUREMEN T
The first prototype HBcrAg assay was reported in 2002. 18 As HBcAg, HBeAg and p22cr all share an identical 149 amino acid sequence, the same monoclonal antibodies can be used for the detection of all 3 proteins. The specimen is mixed with anti-HBcrAg-coated particles and allowed to form antigen-antibody immunocomplexes. Alkalaine phosphatase (ALP)-labelled anti-HBcrAg specifically binds to HBcrAg of the immunocomplexes on the particles, and additional immunocomplexes are formed. Luminescence is generated and the luminescent signal reflects the amount of HBcrAg. Quantitation of HBcrAg is made by comparing the chemiluminescence signal generated by known concentration of recombinant ProHBeAg. This assay is currently available in an automated format, using the Lumipulse G1200 CLEIA analyser (Fujirebio, Tokyo, Japan), with a lower limit of detection of 2.0 log U/mL, and a linear range of 3.0 log U/mL-7.0 log U/mL (1 kU/mL is equal to 3 logU/mL). Samples with HBcrAg above 7 log U/mL can be diluted and retested in order to calculate the quantitative HBcrAg level.
According to the Lumipulse â G HBcrAg Immunoreaction cartridges set product insert, the accuracy of the test is confirmed by assay values (calculated with the assay values in log U/mL) for 3 in-house controls range within AE 5% of their control values. The coefficient of variation for specimens (calculated with the assay values in log U/mL) is less than 5% when subjected to the assay 6 replicates. The analytical sensitivity of the reagent is 3.0 logU/mL. Serum-plasma equivalence is reflected by correlation coefficient of 0.999.

| Serum and intrahepatic HBV DNA
The serum HBcrAg concentration correlates strongly with the serum HBV DNA concentration in a positive and linear manner, regardless of the HBeAg status (Table 1). [19][20][21][22] Three out of 4 studies showed that the correlation coefficient is greater than .8. Intrahepatic total HBV DNA also correlates well with serum HBcrAg (correlation coefficient .67-.70) in patients who are either treatment-na€ ıve or treatment-experienced (Table 2). 21,22

| Intrahepatic cccDNA
It has been shown that serum cccDNA levels correlated only moderately (r = .481) with intrahepatic cccDNA (N = 39) and the assay involves multiple complicated steps 23 rendering it an unfavourable surrogate marker. HBcrAg has a much stronger correlation with intrahepatic cccDNA. 21,22,24 Except for a study with small number of patients (N = 31) 24 , the correlation coefficient is .664-.70 (Table 3). 21,22 In comparison, serum HBV DNA also correlates equally well with intrahepatic cccDNA (correlation coefficient .7, P<.001). 22 However, patients receiving anti-viral therapy often have undetectable serum HBV DNA, while 78% of these patients still have detectable serum HBcrAg. 22 Therefore, in the context of serum DNA undetectability, HBcrAg would be the preferred serum marker to estimate intrahepatic cccDNA quantity.

| HBsAg and linearised HBsAg (HQ-HBsAg)
Serum HBsAg was shown to correlate moderately with intrahepatic cccDNA (r = .46, P<.001) (N = 82). 25  With the above findings, HBcrAg seems to have the advantageous property of good correlation with both viral replicative activities, ie HBV DNA, and viral protein synthesis, ie HBsAg. In addition, HBcrAg is preferred over serum HBV DNA as a surrogate marker for intrahepatic cccDNA in the majority of on-therapy patients with undetectable serum HBV DNA.

HISTORY OF CHB
In For HBeAg-negative patients, the HBcrAg levels were significantly lower in HBeAg-negative chronic infection (also known as inactive carrier state) patients compared to HBeAg-negative chronic hepatitis (also known as HBeAg-negative active phase) patients (2.60 vs 4.92 log U/mL, respectively; P<.001) ( Figure 2). 16  intrahepatic cccDNA respectively. 38 Therefore, a decline in serum HBV DNA does not correlate well with a reduction in intrahepatic cccDNA for those receiving anti-viral therapy. In contrast, HBcrAg levels decline in a similar manner as cccDNA when patients are treated with NA (Table 4). 21 In patients treated with entecavir (ETV), the annual rate of HBcrAg decline was shown to be 0.244 kU/mL/year after 7 years of treatment. 39 More rapid decline was seen in patients who were HBeAg positive, had high baseline HBV DNA, and in the first year of ETV therapy compared to subsequent years ( Figure 3). 39 More importantly, the decline in HBcrAg demonstrates good correlation with the magnitude of change in intrahepatic cccDNA. 21 (Table 5). 43  These studies demonstrated that HBcrAg is a responsive viral marker to be measured whenever treatment can suppress viral DNA and protein synthesis. In particular for a situation where serum HBV DNA becomes undetectable, HBcrAg seems to remain as a measurable serum marker to correlate with the cccDNA.  43 FU, follow-up; HBcrAg, hepatitis B core-related antigen; PEG-IFN, pegylated interferon. performed no better than using HBsAg titre alone (AUROC 0.771). 48 It seems that HBcrAg may not be more advantageous than HBsAg titre in terms of predicting HBsAg loss in treatment-experienced patients.
Although the majority of patients treated with NA will remain on long-term therapy, a proportion of patients may opt to discontinue therapy. The decision to stop therapy has been traditionally based on viral serological markers, serum HBV DNA, ALT and more recently, HBsAg levels. The decline in HBcrAg observed under treatment with NA therapy, and the pattern of decline, might provide prognostic information on the risk of post-treatment reactivation of HBV. 43 For patients treated with LAM or ETV, a high end-of-treatment HBcrAg levels (reported range: 3.2-3.7 log U/mL) was shown to predict relapse after cessation of therapy despite undetectable HBV DNA for at least 6 months (

| UNR ESOLVE D ISSUES OF H BcrAg
There are several unresolved issues in HBcrAg measurement.
Firstly, the lower limit of detection for serum HBcrAg is 2 logU/ mL, and this is particularly an issue for HBeAg-negative patients.

APPLICATION OF HBcrAg
Before