MR.

ENQIANG CHEN (Orcid ID : 0000-0002-8523-1689)

Accepted Article
Article type : Original Paper

Serum HBcrAg is better than HBV RNA and HBsAg in reflecting intrahepatic covalently
closed circular DNA

Running title：Serum HBcrAg can well reflect HBV cccDNA

En-Qiang Chen1,2, Meng-Lan Wang1,2, Ya-Chao Tao1,2, Dong-Bo Wu1,2, Juan Liao1,2, Min
He1,2, Hong Tang1,2*
1
Center of Infectious Diseases, West China Hospital, Sichuan University, Chengdu, Sichuan
610041, PR China
2
Division of Infectious Diseases, State Key Laboratory of Biotherapy, Sichuan University,
Chengdu, Sichuan 610041, PR China

*Corresponding author: Hong Tang, MD, PhD, Center of Infectious Diseases, West China
Hospital, Sichuan University, Chengdu, Sichuan 610041, PR China. Tel: 86-28-85422650;
Fax: 86-28-85423052; E-mail address: [email protected]

List of abbreviations:

HBV, hepatitis B virus; CHB, chronic hepatitis B; cccDNA, covalently closed circular DNA;
HBsAg, hepatitis B surface antigen; HBcrAg, hepatitis B core-related antigen; HBcAg,
hepatitis B core antigen; HBeAg, hepatitis B e antigen; pgRNA, pregenomic RNA; DR,
direct repeat; PC, precore; BCP, basal core promoter; NAs, nucleos(t)ide analogues

This article has been accepted for publication and undergone full peer review but has not been
through the copyediting, typesetting, pagination and proofreading process, which may lead to
differences between this version and the Version of Record. Please cite this article as doi:
10.1111/jvh.13061

This article is protected by copyright. All rights reserved.

 Conflict of interest: The authors do not have any conflict of interest.
Accepted Article
Funding Statement： No funding.

Abstract
The correlation between serum HBcrAg and HBV RNA is unclear; and correlations of
intrahepatic cccDNA with HBcrAg, HBV RNA and HBsAg are rarely reported in the same
cohort. This study aimed to assess the correlation of HBcrAg with HBV RNA and HBsAg, and
investigate whether serum HBcrAg is superior to serum HBV RNA and HBsAg in reflecting
intrahepatic HBV cccDNA in HBeAg-positive and HBeAg-negative CHB patients. In this
study, 85 HBeAg-positive and 25 HBeAg-negative patients who have never received
antiviral therapy, were included. Among HBeAg-positive patients, HBcrAg was correlated
positively with HBsAg(r=0.564, P<0.001) and HBV RNA(r=0.445, P<0.001), and HBV RNA
was also correlated positively with HBsAg(r=0.323, P=0.003). Among HBeAg-negative
patients, no significantly correlation was observed between HBcrAg, HBsAg and HBV RNA.
By multivariable linear regression, HBcrAg(=-0.563, P<0.001), HBsAg(=-0.328, P<0.001),
and HBV RNA( =0.180, P=0.003) were all associated with cccDNA levels among
HBeAg-positive patients, but only serum HBcrAg was associated with cccDNA
level(β=0.774, P=0.000) among HBeAg-negative patients. HBcrAg was better correlated with
cccDNA as compared to HBsAg and HBV RNA, irrespective of HBeAg status. Among
HBeAg-positive patients, though HBcrAg level was influenced by hepatic inflammatory
activity and HBV DNA levels, the good correlations of HBcrAg with cccDNA persisted after
stratification by inflammatory activity and HBV DNA levels. In conclusion, correlations
of serum HBcrAg, HBV RNA and HBsAg levels differ significantly between HBeAg-positive
and HBeAg-negative patients, but serum HbcrAg correlates with cccDNA levels better than
HBV RNA and HBsAg, irrespective of HBeAg status.

This article is protected by copyright. All rights reserved.

 Key words: hepatitis B core-related antigen; hepatitis B surface antigen; HBV RNA; covalently

closed circular DNA

Accepted Article
Introduction

Hepatitis B virus (HBV) infection is a worldwide public health problem. It is estimated that
240 million people are chronically infected and at least 650,000 people die each year due to
chronic hepatitis B (CHB) worldwide1. Though the currently approved nucleos(t)ide analogues
can effectively reduce serum HBV DNA of CHB patients, HBV is difficult to eliminate due to
the persistence of HBV covalently closed circular DNA (cccDNA) in the infected
hepatocytes2. As a template for transcription of all viral RNAs, intrahepatic cccDNA can
produce the offspring virion DNA and influence viral proteins synthesis 2. Additionally, low
levels of intrahepatic cccDNA also predict sustained virologic response after cessation of
antiviral therapy2. Thus dynamic monitoring of intrahepatic cccDNA level should be helpful to
accurately assess the efficacy of antiviral therapy and disease progression risk3. However,
because of the invasive nature of the procedure and potential for sampling error, dynamic
liver biopsy is not well tolerated , which greatly limits the use of intrahepatic cccDNA in
real-world clinical practice4. Therefore, many non-invasive convenient markers have been
investigated to reflect intrahepatic cccDNA level5. As a classic indicator, serum hepatitis B
surface antigen (HBsAg) levels have long been thought to be correlated with intrahepatic
cccDNA, but this correlation is not strong6.

Hepatitis B core-related antigen (HBcrAg) is a new valuable serum marker of HBV, and it
consists of 3 species of related proteins sharing an identical 149 amino acid sequence:
hepatitis B core antigen (HBcAg), hepatitis B e antigen (HBeAg), and a truncated 22 kDa
precore protein (p22Cr)7. Like HBeAg, p22cr is also a processed product of the precore
protein, but with protein processing at both the N- and C-terminals7. In our previous studies,
early on-treatment serum HBcrAg level was found to be a good biomarker for predicting

This article is protected by copyright. All rights reserved.

 off-treatment HBeAg seroconversion in patients receiving peginterferon therapy8; and as
compared to serum HBsAg level, serum HBcrAg has a better correlation with intrahepatic
Accepted Article
cccDNA6, 9. In addition, patients with low HBcrAg and HBsAg levels are also reported to have
a low relapse risk after cessation of antiviral therapy10, 11.

Recently, serum HBV RNA is also attracting attention as a useful biomarker12, 13. As early as
1996, serum HBV RNA was identified in HBV infected patients but its nature and origin was
unclear until recently when serum HBV RNA was confirmed to be pregenomic RNA
(pgRNA). Undetectable serum HBV RNA level has been reported to be associated with
sustained virological response to antiviral therapy, and it might serve as a new potential
surrogate marker to reflect the status of intrahepatic cccDNA14, 15. In addition, some scholars
have called for a new definition of sustained virological response (using undetectable HBV
DNA plus HBV RNA in place of undetectable HBV DNA)16. It is worth mentioning that the
majority of published studies are foccussed on HBeAg-negative patients, and the correlation
between serum HBV RNA and intrahepatic cccDNA is rarely reported in HBeAg-positive
patients.

At present, the correlation between HBcrAg and HBV RNA is unclear. The correlations of
intrahepatic cccDNA with serum HBcrAg, HBV RNA and HBsAg are rarely reported in the
same cohort. This study was designed to assess the correlation of serum HBcrAg with serum
HBV RNA and HBsAg, and further investigate whether serum HBcrAg is superior to serum
HBV RNA and HBsAg in reflecting intrahepatic HBV cccDNA in HBeAg-positive and
HBeAg-negative CHB patients.

Patients and Methods

Patients

This was a retrospective study of CHB patients who underwent percutaneous liver biopsy at
the West China Hospital of Sichuan University between January 2012 and December 2013. All
This article is protected by copyright. All rights reserved.
 patients had not received antiviral therapy before the biopsy. Patients were excluded if they had
any evidence of other concomitant liver diseases (including alcoholic liver disease,
Accepted Article
autoimmune liver disease, and hepatocellular carcinoma) or markers of hepatitis C virus or
human immunodeficiency virus co-infections. In addition, patients without serum for HBcrAg
measurement and eligible liver tissue for intrahepatic HBV cccDNA measurement were also
excluded. The detailed information of patient acquisition procedures is shown in Fig.1.

This study was conducted in accordance with the 1975 Declaration of Helsinki. The study
protocol was approved by the West China Hospital Ethics Committee, and verbal informed
consent was obtained from each patient.

General laboratory variables measurement

Serum biochemical indexes were measured according to standard procedures (Olympus

AU5400, Olympus Corporation, Tokyo, Japan). Serum HBeAg status was assessed using
electrochemiluminescence immunoassay (Roche Diagnostics, Indianapolis, IN, USA). Serum
HBsAg level was quantitatively measured using Elecsys® HBsAg II Quant Assay (Roche
Diagnostics, Penzberg, Germany). Serum HBV-DNA concentration was quantitatively
determined using Cobas Taqman assay kit (Roche Diagnostics, Branchburg, NJ), with a lower
limit of detection of 20 IU/mL. HBV genotypes were determined by direct S-gene sequencing.
The PC/BCP mutations of HBV (including A1762T and G1764A) were detected using
commercially available Line Probe Assays (INNOGENETICS, Belgium).

Serum HBcrAg measurement

The serum HBcrAg level was quantitatively measured using the fully automated CLEIA
system (Fujirebio Inc., Tokyo, Japan), and the detailed process of serum HBcrAg measurement
is as previously reported17. Since the assay's validated measurement range is from 1,000 U/ml

This article is protected by copyright. All rights reserved.

 (3 log10 U/ml) to 10,000,000 U/ml (7 log10 U/ml), serial dilutions of the serum sample is
required when serum HBcrAg level is above the detection limit.
Accepted Article
Serum HBV-RNA measurement

The HBV RNA was detected by RNA simultaneous amplification testing method (HBV-SAT)
based on real-time fluorescence detection of isothermal RNA amplification using HBV-SAT
kit (Shanghai Rendu Biotechnology Co., Ltd. China) according to the manufacturers
recommendations. Briefly RNA was extracted by magnetic microparticles with HBV specific
RNA oligonucleotides. The target RNA were reverse transcribed by MMLV enzyme,
transcribed by T7 RNA polymerase and detected by RNA beacon probe labeled by
fluorescence and quencher. The concentration of serum HBV RNA was calculated using
normalization to the internal control (IC) nucleic acid, which was distinct from HBV genome
and human genome. A fixed dose of IC was added to each same volume sample from nuclear
acid extraction step. All the reagents of the whole assay procedure were sufficient to IC
detection and the IC amplification results should be constant theoretically. The HBV RNA
amplification results were calibrated by IC amplification result and avoid the effects of the
variations in specimen processing, amplification and detection. The linear range was
established by testing panels of armored HBV RNA diluted in HBV negative human serum.
The linear ranged in concentration from 2 log copies/mL to 8 log copies/mL. The R2 value of
linear equation is more than 0.95. The limit of detection is 50 copies/mL.

Intrahepatic HBV cccDNA measurement

Intrahepatic HBV cccDNA levels in paraffin embedded liver tissue were measured with the
real-time PCR method, as described previously6, 18. For specifically amplification and
quantification of cccDNA, we designed cccDNA-selective primers and a probe targeting the
gap region between the viral genome direct repeat regions (DR1 and DR2). In the present study,
the cellular DNA was quantified by determining the copy number of cellular

This article is protected by copyright. All rights reserved.

 house-keeping gene -actin. To establish the standard curves for cccDNA quantitation,
ten-fold serial dilutions (102–109 copies/mL) of a plasmid containing the entire wild-type HBV
Accepted Article
genotype C genome were used. Human liver tissue without HBV infection was used as
negative controls. The amount of cccDNA was expressed as the number of copies per cell, with
the estimation of 6.667 pg of DNA/cell. The detailed primers information for intrahepatic HBV
cccDNA measurement are aas previously reported.

Statistical analyses
Continuous variables were expressed as median and range, and categorical variables were
expressed as counts and percentages. Student's t test or Mann–Whitney test was used to
analyze the differences between continuous variables; and paired samples t test was used to
analyze the continuous variables before and after antiviral therapy. The correlation between
two continuous variables was calculated using Spearman’s bivariate correlation analysis, and
the correlation is significant at the 0.01 level (2-tailed). Additionally, linear regression analysis
were also performed to determine factors associated with intrahepatic HBV cccDNA levels. A
P value less than 0.05 was considered to indicate statistical significance. All statistical analyses
were done with SPSS Version 18.0 (SPSS, Chicago, IL), and figures were drawn using
GraphPad Prism 6 (GraphPad Software Inc., California, USA).

Results

Patient’s characteristics

A total of 110 eligible CHB patients were analyzed, including 85 HBeAg-positive patients
(median age 31 years [range 20-57];57 male[67.05%]) and 25 HBeAg-negative patients
(median age 38 years [range 23-52];17 male[68.00%]). For HBeAg-positive patients, the
median level was 7.91 log10 IU/mL for serum HBV DNA, 6.83 log10 copies/mL for serum
HBV RNA, 4.59 log10 IU/mL for serum HBsAg, 2.81 log10 COI for serum HBeAg, 10.30 for
serum HBcrAg log10 U/mL, and 7.46 log10 copies/106 cell for intrahepatic HBV cccDNA. For

This article is protected by copyright. All rights reserved.

 HBeAg-negative patients, the median level was 3.68 log10 IU/mL for serum HBV DNA, 2.86
log10 copies/mL for serum HBV RNA, 3.49 log10 IU/mL for serum HBsAg,5.40 for serum
Accepted Article
HBcrAg log10 U/mL, and 6.03 log10 copies/106 cell for intrahepatic HBV cccDNA. The
detailed characteristics of patients and comparison between HBeAg-positive and
HBeAg-negative patients are shown in Table 1.

Regression analysis of factors associated with intrahepatic cccDNA level

Among HBeAg-positive patients, by univariable linear regression, factors associated with

intrahepatic cccDNA were age, inflammatory grade, serum HBeAg, HBcrAg, HBsAg, HBV
RNA and HBV DNA; while gender, fibrosis stage, ALT levels and HBV genotype were not
associated with intrahepatic cccDNA level (Table.2). By multivariable linear regression,
factors of serum HBcrAg, HBsAg and HBV RNA were all associated with intrahepatic
cccDNA, and the performance of serum HBcrAg (β=0.563, P<0.001) were superior to that of
serum HBsAg (β=0.328, P<0.001)and HBV RNA(β=0.180, P=0.003)( Table.2).

Among HBeAg-negative patients, by univariable linear regression, factors associated with

intrahepatic cccDNA were serum HBcrAg and HBsAg; while age, gender, inflammatory
grade, fibrosis stage, ALT levels, HBV genotype,HBV RNA and HBV DNA were all not
associated with intrahepatic cccDNA level. By multivariable linear regression, only serum
HBcrAg was associated with intrahepatic cccDNA level (β=0.774, P=0.000) ( Table.2).

Correlation analysis among serum HBcrAg, HBV RNA and HBsAg levels

The correlation of different HBV serum markers among HBeAg-positive and HBeAg-negative
patients are shown in Fig.2. Among HBeAg-positive patients, serum HBcrAg correlated
strongly with serum HBsAg (r=0.564, P<0.001). Both serum HBcrAg (r=0.445, P<0.001) and
HBsAg (r=0.323, P=0.003) correlated moderately with serum HBV RNA. It was worth to

This article is protected by copyright. All rights reserved.

 mention that the correlation was moderate for serum HBcrAg and HBV DNA(r=0.445,
P<0.001), strong for serum HBsAg and HBV DNA(r=0.654, P<0.001), but not significant
Accepted Article
correlation for serum HBV RNA and HBV DNA (r=0.271, P=0.012). Additionally, we also
analyzed the correlation of serum HBeAg level with serum HBcrAg, HBsAg HBV RNA and
HBV DNA. And we found that serum HBeAg strongly correlated with HBVDNA (r=0.540,
P<0.001) and HBsAg ((r=0.520, P<0.001), moderately with serum HBcrAg (r=0.491,
P<0.001) and weakly correlated with serum HBVRNA (r=0.299, P=0.005).

Among HBeAg-negative patients, serum HBcrAg correlated strongly with serum HBsAg
(r=0.552, P=0.007), but not significantly correlated with serum HBV RNA (r=-0.017,
P=0.937) and HBV DNA (r=-0.187, P=0.370). Additionally, no significant correlation was
observed between serum HBsAg and HBV RNA(r=0.156, P=0.457), serum HBsAg and HBV
DNA(r=-0.038, P=0.856), and serum HBV RNA and HBV DNA(r=0.398, P=0.049).

Serum HBcrAg, HBV RNA and HBsAg distribution stratified by demographic and other
clinical characteristics

The distribution of serum HBcrAg, HBV RNA and HBsAg levels among HBeAg-positive
patients are shown in Table.3. In present study, there was no statistically significant difference
in serum level of HBV RNA stratified by age (P=0.707), gender(P=0.631), inflammatory
grade (P=0.143), fibrosis stage(P=0.183), ALT(P=0.419), viral genotype(P=0.363), BCP
mutation (P=0.805) and HBV DNA levels (P=0.110). The level of serum HBsAg was
significant higher in patients with young age (P=0.005), low inflammatory grade (P=0.013),
and high serum HBVDNA levels (P<0.001); but similar between patients with different
gender(P=0.545), fibrosis stage (P=0.060), ALT(P=0.186), viral genotype (P=0.057), BCP
mutation (P=0.063). Interesting, the distribution of serum HBcrAg was significant different
among patients with different inflammatory grade (P<0.001) and serum HBV DNA(P=0.004),

This article is protected by copyright. All rights reserved.

 but not influenced by age(P=0.073), gender(P=0.433), fibrosis stage(P=0.154), ALT
level(P=0.829), viral genotype (P=0.066) and BCP mutations(P=0.071).
Accepted Article
Intrahepatic cccDNA in relation to serum HBcrAg, HBsAg, HBV RNA and HBV DNA

Among HBeAg-positive patients, the level of intrahepatic cccDNA strongly correlated with
serum HBcrAg (r=0.843, P<0.001) and serum HBsAg (r=0.710, P<0.001), and moderately
correlated with serum HBV RNA (r=0.541, P<0.001), serum HBV DNA (r=0.507, P<0.001)
and serum HBeAg (r=0.510, P<0.001)(Fig.3). The correlations of intrahepatic cccDNA with
serum HBcrAg, HBsAg and HBV RNA, stratified by inflammatory grade and serum HBV
DNA levels, are shown in Table.4. The correlation of serum HBcrAg with intrahepatic
cccDNA was comparable for patients with different inflammatory grade (G<2 vs. G≥2 ) or
serum HBV DNA levels(<8 vs. ≥8 log10 IU/mL). The correlation of serum HBV RNA with
intrahepatic cccDNA were also not influenced by inflammatory grade and serum HBV DNA
levels. However, the correlation of serum HBsAg with intrahepatic cccDNA was HBV
DNA-dependent, indicated by a moderate correlation within HBV DNA<8 log10
IU/mL(r=0.575, P<0.001), but strong correlation within HBV DNA≥8 log10 IU/mL(r=0.926,
P<0.001).

Among HBeAg-negative patients, the level of intrahepatic cccDNA strongly correlated with
serum HBcrAg (r=0.865, P<0.001) and moderately correlated with serum HBsAg (r=0.579,
P=0.002), but not significantly correlated with serum HBV RNA (r=-0.028, P=0.892) and
HBV DNA(r=-0.086, P=0.681)(Fig.3).

Discussion

In the past decades, the fundamental role of intrahepatic cccDNA, as a template for
transcription of all viral RNAs and further synthesis of viral proteins, has been recognized17.
Thus, monitoring intrahepatic cccDNA levels could reflect the real activity of HBV replication

This article is protected by copyright. All rights reserved.

 in patients, and low levels of intrahepatic cccDNA could also predict sustained virologic
response after cessation of antiviral therapy17. In recent years, several HBV-associated serum
Accepted Article
markers (including serum HBcrAg, HBsAg and HBV RNA ) were all reported to be a potential
indicator of intrahepatic cccDNA activity in different studies5, 16, and low level of serum
HBcrAg or loss of serum HBV RNA also might indicate exhausting or transcription silencing
of cccDNA reservoirs7, 16. To our knowledge, the present study represents a truly “real-life”
correlation analysis of serum HBcrAg and HBV RNA levels, and a first head-to-head
comparison of serum HBcrAg, HBsAg and HBV RNA levels in reflecting intrahepatic
cccDNA levels. The main findings from the present study are as follow: serum HBcrAg level is
positively correlated with serum HBsAg or HBV RNA among HBeAg-positive patients, but
not among HBeAg-negative patients; serum HBcrAg and HBsAg levels are both significantly
influenced by inflammatory grade and serum HBV DNA levels among HBeAg-positive
patients, but serum HBV RNA level is not; serum HBcrAg, HBsAg and HBV RNA levels are
all associated with intrahepatic cccDNA levels among HBeAg-positive patients, but only
serum HBcrAg was associated with cccDNA level among HBeAg-negative patients.

For a long time, the quantitative measurement of serum HBV DNA was widely used to
estimate the activity of viral replication and anti-viral efficacy of nucleos(t)ide analogues(NAs)
treatment. However, NAs only act on limited steps of the viral replication cycle, and
production of viral intermediate proteins may not be affected significantly19. Therefore,
measurement of viral proteins can be useful in monitoring HBV activities, especially in
patients receiving NAs when HBV DNA levels are undetectable. Currently, the most attractive
viral proteins should be serum HBcrAg and HBsAg, and both of them can be found in mature
virions as well as HBV DNA-negative empty particles. The effectiveness of either serum
HBcrAg or HBsAg in reflecting intrahepatic cccDNA level had been identified in cohorts of
patients with different races and viral genotypes7, 20-22. Though a positive correlation was
reported between HBcrAg and HBsAg in HBeAg-positive patients without antiviral therapy,
the intensity of correlation was moderate (r=0.564), which indicated that serum HBcrAg and
HBsAg could not replace each other6. Additionally, serum HBcrAg and HBsAg had unique

This article is protected by copyright. All rights reserved.

 patterns of distribution throughout the five disease phases of CHB, including high detectability
rates of serum HBcrAg after HBsAg seroclearance, which decided the different possibilities
Accepted Article
for their applicability in clinical practice7. Thus, a difference may exist between serum HBcrAg
and HBsAg in reflecting intrahepatic cccDNA level. In fact, we previously investigated the
correlations of serum HBcrAg and HBsAg level with intrahepatic cccDNA among 139 patients
with liver biopsy, and found that serum HBcrAg had a much stronger correlation with
intrahepatic cccDNA than serum HBsAg, either before or during NAs treatment6. Importantly,
the better performance of serum HBcrAg than HBsAg in reflecting intrahepatic cccDNA level
was also observed in this cohort. As we know, serum HBsAg also could originate from the
expression of the integrated HBV S gene in patients with S gene integration23, 24, and this may
partly cause the performance of HBsAg to be inferior to that of HBcrAg in reflecting
intrahepatic cccDNA level and activity.

In theory, intrahepatic HBV RNA could well correlate with intrahepatic cccDNA, but the
biggest problems facing intrahepatic HBV RNA application is the difficulty in measurement,
as the latter relies on liver biopsy. As we know, serum HBV RNA is also transcribed from
intrahepatic cccDNA and thus it may be a potential alternative marker for intrahepatic
cccDNA. Though a recent study showed that serum HBV RNA level could reflect intrahepatic
cccDNA transcriptional activity, it was nevertheless inferior to serum HBV DNA in reflecting
the level of intrahepatic cccDNA before treatment14. In the present study, as compared to the
strong correlation of serum HBcrAg (r=0.843) and HBsAg (r=0.710) with intrahepatic
cccDNA, the similar moderate correlation intensity for serum HBV RNA (r=0.541) and HBV
DNA (r=0.507) with intrahepatic cccDNA suggested that serum HBV RNA was not a good
indicator of intrahepatic cccDNA. Though the weak correlation of serum HBV RNA with both
serum HBcrAg and HBsAg indicated a different possibility for its applicability in future, serum
HBV RNA should be at least not an ideal indicator for the activity of HBV replication in liver
tissue. In fact, the poor efficacy of serum HBV RNA in reflecting intrahepatic cccDNA activity
was also reported in another reently published study15.

This article is protected by copyright. All rights reserved.

 A large number of studies have showed that the hepatic inflammation and fibrosis are
prominent features in chronic viral hepatitis, and the severity of inflammation and fibrosis play
Accepted Article
important roles in the decision making of antiviral treatment and risk assessment of disease
progression. In present study, the levels of serum HBV RNA were similar between patients
with different inflammation grade and fibrosis stage. However, Prof. Zhang WH and his team
reported an association between serum HBV-RNA levels and liver histological changes
(r=0.665 for grading and r= 0.722 for staging) in patients receiving NAs therapy13. Due to the
fact that there were limited studies and small samples were provided, more data is still required
to clarify whether the level of serum HBV RNA could effectively predict severe inflammation
and advanced fibrosis. In the present study, we found that serum HBcrAg and HBsAg levels
were influenced by liver histological changes, and both serum HBcrAg and HBsAg correlated
strongly with intrahepatic cccDNA, regardless of inflammation grade. Among patients with
severe inflammation, though serum HBcrAg showed a relative good correlation with
intrahepatic cccDNA(r=0.743), its accuracy for predicting intrahepatic cccDNA levels may be
further improved if combined with serum HBsAg. It is worth mentioning that the correlation
between serum HBsAg and intrahepatic cccDNA would be affected by serum HBV DNA
levels; and the correlation in patients with low serum HBV DNA levels was notably less than
that in patients with high serum HBV DNA levels. On the contrary, the correlations between
serum HBcrAg and intrahepatic cccDNA levels were not only excellent but also not influenced
by serum HBV DNA levels. Thus, to a certain extent, serum HBcrAg may be more stable than
serum HBsAg in reflecting intrahepatic cccDNA, especially among patients with high serum
HBV DNA levels.

In the present study, multivariable linear regression analysis also showed that serum HBcrAg
had the highest performance of association with intrahepatic cccDNA levels, which were
followed by serum HBsAg and HBV RNA. Indeed, this finding was also highly consistent with
the result of the correlation analysis. In previous studies, HBV genotypes, serum HBV DNA
levels and liver histological changes were reported to be associated with the levels of serum
HBcrAg, HBsAg or HBV RNA7, 21, 25. However, they had no significant association with

This article is protected by copyright. All rights reserved.

 intrahepatic cccDNA levels among HBeAg-positive patients in this study. Theoretically, the
prediction efficiency should be better for the combination of several correlated variables than
Accepted Article
single variable. Thus, if serum HBcrAg,HBsAg and HBV RNA were combined together, the
intrahepatic cccDNA levels should be more reliably and more accurately reflected.

Several limitations exist in this study. First, the sample size is relatively small, and there was a
heterogeneous sample size between HBeAg positive and HBeAg negative patients. So further
large sample-size cohort studies are required to confirm the present findings. Second, the
methods for serum HBV RNA and intrahepatic cccDNA quantification are not standardized,
which may lead to inconsistent results with other studies, thus standardized testing methods
and agents are urgently needed. Third, as not all intrahepatic cccDNA molecules are equally
transcriptional active, present findings just suggest a correlation of serum HBV serum markers
with total amount of intrahepatic cccDNA but not the transcriptional activity of intrahepatic
cccDNA.

In summary, the present study is the first head-to-head comparison of serum HBcrAg, HBsAg
and HBV RNA levels in reflecting intrahepatic cccDNA levels. We find that serum HBcrAg,
HBsAg and HBV RNA levels are significantly correlated with each other among
HBeAg-positive patients but not among HBeAg-negative patients; and serum HBcrAg is better
than HBV RNA and HBsAg in correlation with cccDNA level, irrespective of HBeAg status.
Thus, based on the findings of the present study and previous reports, serum HBcrAg is likely
to be the most useful marker for disease monitoring, predicting treatment response and disease
outcome of CHB.

Acknowledgments
We appreciate the kind work from Dr. Dong-Mei Zhang, Cong Liu and Ling-Bo Liang, West
China Hospital of Sichuan Hospital in collection and separation of blood and liver samples.

This article is protected by copyright. All rights reserved.

 References
Accepted Article
1 World Health Organization. Guidelines for the prevention, care and treatment of persons with chronic
hepatitis B infection. http://wwwwhoint/hiv/pub/hepatitis/hepatitis-b-guidelines/en/.

2 Sung JJ, Wong ML, Bowden S, et al. Intrahepatic hepatitis B virus covalently closed circular DNA can be a
predictor of sustained response to therapy. Gastroenterology, 2005; 128:1890-7.

3 Yang HC, Kao JH. Persistence of hepatitis B virus covalently closed circular DNA in hepatocytes: molecular
mechanisms and clinical significance. Emerg Microbes Infect, 2014; 3:e64.

4 Thampanitchawong P, Piratvisuth T. Liver biopsy:complications and risk factors. World J Gastroenterol,

1999; 5:301-4.

5 Lin CL, Kao JH. New perspectives of biomarkers for the management of chronic hepatitis B. Clin Mol
Hepatol, 2016; 22:423-31.

6 Chen EQ, Feng S, Wang ML, et al. Serum hepatitis B core-related antigen is a satisfactory surrogate
marker of intrahepatic covalently closed circular DNA in chronic hepatitis B. Sci Rep, 2017; 7:173.

7 Mak LY, Wong DK, Cheung KS, Seto WK, Lai CL, Yuen MF. Review article: hepatitis B core-related antigen
(HBcrAg): an emerging marker for chronic hepatitis B virus infection. Aliment Pharmacol Ther, 2018; 47:43-54.

8 Wang ML, Liao J, Wei B, et al. Comparison of hepatitis B virus core-related antigen and hepatitis B surface
antigen for predicting HBeAg seroconversion in chronic hepatitis B patients with pegylated interferon therapy.
Infect Dis (Lond), 2018:1-9.

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 Drafting Committee for Hepatitis Management G, the Japan Society of H. JSH Guidelines for the
Management of Hepatitis B Virus Infection. Hepatol Res, 2014; 44 Suppl S1:1-58.

11 Wang ML, Chen EQ, Tao CM, Tang H. Letter: Serum HBcrAg is a useful marker for disease monitoring,
predicting treatment response and disease outcome of CHB. Alimentary Pharmacology & Therapeutics, 2018.

12 Wang J, Shen T, Huang X, et al. Serum hepatitis B virus RNA is encapsidated pregenome RNA that may be
associated with persistence of viral infection and rebound. J Hepatol, 2016; 65:700-10.

13 Wang J, Yu Y, Li G, et al. Relationship between serum HBV-RNA levels and intrahepatic viral as well as
histologic activity markers in entecavir-treated patients. J Hepatol, 2017.

This article is protected by copyright. All rights reserved.

 14 Gao Y, Li Y, Meng Q, et al. Serum Hepatitis B Virus DNA, RNA, and HBsAg: Which Correlated Better with
Intrahepatic Covalently Closed Circular DNA before and after Nucleos(t)ide Analogue Treatment? J Clin
Microbiol, 2017; 55:2972-82.
Accepted Article
15 Huang H, Wang J, Li W, et al. Serum HBV DNA plus RNA shows superiority in reflecting the activity of
intrahepatic cccDNA in treatment-naive HBV-infected individuals. J Clin Virol, 2018; 99-100:71-8.

16 Lu F, Wang J, Chen X, Xu D, Xia N. Potential use of serum HBV RNA in antiviral therapy for chronic
hepatitis B in the era of nucleos(t)ide analogs. Front Med, 2017; 11:502-8.

17 Nassal M. HBV cccDNA: viral persistence reservoir and key obstacle for a cure of chronic hepatitis B. Gut,
2015; 64:1972-84.

18 Liang LB, Zhu X, Yan LB, et al. Quantitative intrahepatic HBV cccDNA correlates with histological liver
inflammation in chronic hepatitis B virus infection. Int J Infect Dis, 2016; 52:77-82.

19 Wu D, Ning Q. Toward a Cure for Hepatitis B Virus Infection: Combination Therapy Involving Viral
Suppression and Immune Modulation and Long-term Outcome. J Infect Dis, 2017; 216:S771-S7.

20 Wong DK, Seto WK, Fung J, et al. Reduction of hepatitis B surface antigen and covalently closed circular
DNA by nucleos(t)ide analogues of different potency. Clin Gastroenterol Hepatol, 2013; 11:1004-10 e1.

21 Tseng TC, Kao JH. Clinical utility of quantitative HBsAg in natural history and nucleos(t)ide analogue
treatment of chronic hepatitis B: new trick of old dog. J Gastroenterol, 2013; 48:13-21.

22 Werle-Lapostolle B, Bowden S, Locarnini S, et al. Persistence of cccDNA during the natural history of
chronic hepatitis B and decline during adefovir dipivoxil therapy. Gastroenterology, 2004; 126:1750-8.

23 Hu B, Wang R, Fu J, et al. Integration of hepatitis B virus S gene impacts on hepatitis B surface antigen
levels in patients with antiviral therapy. J Gastroenterol Hepatol, 2018; 33:1389-96.

24 Wooddell CI, Yuen MF, Chan HL, et al. RNAi-based treatment of chronically infected patients and
chimpanzees reveals that integrated hepatitis B virus DNA is a source of HBsAg. Sci Transl Med, 2017; 9.

25 van Campenhout MJH, van Bommel F, Pfefferkorn M, et al. Host and viral factors associated with serum
hepatitis B virus RNA levels among patients in need for treatment. Hepatology, 2018.

This article is protected by copyright. All rights reserved.

 Figures

Fig.1 The acquisition process of registered patients with liver biopsy in present retrospective
Accepted Article
study.

Fig.2 Correlations among different serum markers of HBV in HBeAg-positive (A~F) and
HBeAg-negative patients(G~L). (A, G) : HBcrAg and HBsAg; (B, H) : HBcrAg and HBV
RNA;(C, I) : HBcrAg and HBV DNA; (D, J): HBsAg and HBV RNA;(E, K) HBsAg and HBV
DNA; (F, L): HBV RNA and HBV DNA

Fig.3 Correlations of serum viral proteins with intrahepatic cccDNA in HBeAg-positive (A~D)
and HBeAg-negative patients (E~H). (A, E): Serum HBcrAg; (B, F): Serum HBsAg; (C, G):
Serum HBV RNA; (D, H): Serum HBV DNA.

This article is protected by copyright. All rights reserved.

ccepted Articl
Tables

Table 1 Characteristics of patients included in this study

Variables HBeAg positive HBeAg negative P-value

(n=85) (n=25)
Age, median(range), years 31.00(20.00-57.00) 38.00(23.00-52.00) 0.000
Gender, Male/ Female, n(%) 57(67.05)/28(32.94) 17(68.00)/8(32.00) 0.930
Inflammation grade*, G<2/G≥2, n(%) 62(72.94)/23(27.05) 9(36.00)/16(64.00) 0.213
Fibrosis stage*, S<2/S≥2, n(%) 69(81.2)/16(18.8) 6(24.00)/19(76.00) 0.240
HBV genotype, B/C, n(%) 55(64.7)/30(35.3) 13(52.00)/12(48.00) 0.250
HBV BCP mutations, Yes/No, n(%) 21(24.7)/64(75.3) 9(36.00)/16(64.00) 0.265
ALT, median(range), IU/L 30.00(6.00-97.00) 21.00(8.00-56.00) 0.002
ALT, elevated/normal, n(%) 28(32.9)/57(67.1) 5(20.00)/20(80.00) 0.215
Serum HBV-DNA, median (range), log10 IU/mL 7.91(2.90-8.80) 3.68(2.70-6.08) 0.000
Serum HBV RNA, median (range), log10 copies/mL 6.83(4.20-8.15) 2.86(1.70-6.60) 0.000
Serum HBsAg, median (range), log10 IU/mL 4.59(0.82-5.10) 3.49(0.99-4.01) 0.000
Serum HBeAg, median (range), log10 COI 2.81(0.11-3.13) 0.41(0.14-0.84) 0.000
Serum HBcrAg, median (range), log10 U/mL 10.30(6.00-12.30) 5.40(3.28-7.20) 0.000
HBV cccDNA, median (range), log10 copies/106 cell 7.46(5.11-8.17) 6.03(5.00-6.85) 0.000
Note: * pathological assessment of liver tissue using METAVIR score

This article is protected by copyright. All rights reserved.

ccepted Articl
Table.2 Univariable and multivariable linear regression analysis of factors associated with intrahepatic cccDNA

Univariable Multivariable
Parameter
B 95%CI p-value B 95%CI β p-value
HBeAg positive
Age, year -0.023 -0.041~-0.005 0.019 0.001 -0.008~0.011 0.016 0.779
Gender -0.009 -0.272~0.274 0.943
Inflammation grade -0.486 -0.779~-0.213 0.004 -0.031 -0.130~0.068 -0.041 0.530
Fibrosis stage -0.326 -0.654~-0.016 0.053 0.001 -0.081~0.084 0.003 0.971
ALT, IU/mL -0.004 -0.011~0.002 0.240
HBV genotype -0.250 -0.558~0.034 0.098 -0.017 -0.154~0.120 -0.014 0.803
Serum HBV-DNA , log10 IU/mL 0.272 0.158~0.461 0.002 -0.007 -0.086~0.071 -0.014 0.924
Serum HBV RNA, log10 copies/mL 0.416 0.275~0.600 0.001 0.139 0.050~0.227 0.180 0.003
Serum HBeAg, log10 COI 0.460 0.290-0.629 0.000 0.016 0.156~0.269 0.017 0.794
Serum HBsAg, log10 IU/mL 0.578 0.358~1.084 0.022 0.267 0.136~0.399 0.328 0.000
Serum HBcrAg, log10 U/mL 0.318 0.261~0.380 0.001 0.212 0.156~0.269 0.563 0.000
HBeAg negative
Age, year -0.015 -0.050~0.020 0.397
Gender 0.050 -0.451~0.550 0.839
Inflammation grade -0.146 -0.386~0.094 0.221
Fibrosis stage -0.070 -0.260~0.119 0.450
ALT, IU/mL -0.005 -0.023~0.013 0.585
HBV genotype 0.104 -0.361~0.570 0.648
Serum HBV-DNA , log10 IU/mL -0.054 -0.324~0.216 0.681
Serum HBV RNA, log10 copies/mL -0.014 -0.190~0.161 0.869
Serum HBsAg, log10 IU/mL 0.493 0.194~0.793 0.002 0.149 -0.062~0.361 0.175 0.157
Serum HBcrAg, log10 U/mL 0.476 0.357~0.595 0.000 0.425 0.289~0.562 0.774 0.000

This article is protected by copyright. All rights reserved.

ccepted Articl
Table.3 Serum HBcrAg, HBV RNA and HBsAg distribution stratified by demographic and other clinical characteristics among HBeAg-positive
patients

HBcrAg HBV RNA HBsAg

Variable P-Value P-Value P-Value
(Log10 U/mL) (Log10 copies/mL) (Log10 IU/mL)

Age, ys

< 30(n=35) 6.30(6.00-12.30) 0.073 6.90(4.20-7.80) 0.707 4.68(2.99-5.10) 0.005

30(n=50) 9.95(6.70-11.90) 6.79(4.33-8.15) 4.45(0.82-5.10)

Gender, n(%)

Male(n=57) 10.10(6.00-12.30) 0.433 6.78(4.20-8.15) 0.631 4.59(0.82-5.10) 0.545

Female(n=28) 10.65(6.70-12.00) 6.89(5.32-7.80) 4.58(3.08-5.10)

Inflammatory grade

<2(n=62) 10.55(6.00-12.30) 0.000 6.88(4.20-7.80) 0.143 4.67(0.82-5.10) 0.013

≥2(n=23) 9.20(6.30-11.80) 6.74(4.33-8.15) 4.29(1.26-4.80)

Fibrosis stage

<2(n=69) 10.50(6.00-12.30) 0.154 6.86(4.33-8.15) 0.183 4.61(0.82-5.10) 0.060

≥2(n=16) 9.85(6.30-11.80) 6.80(4.20-7.52) 4.34(1.26-4.80)

ALT

Elevated(n=57) 10.40(6.00-12.00) 0.829 6.78(4.20-8.15) 0.419 4.61(3.31-5.10) 0.186

Normal(n=28) 10.20(6.30-12.30) 6.95(5.40-7.74) 4.46(0.82-5.10)

This article is protected by copyright. All rights reserved.

ccepted Articl
HBV genotype

B(n=55) 10.40(6.70-12.30) 0.066 6.81(4.20-7.86) 0.363 4.66(1.26-5.10) 0.057

C(n=30) 9.85(6.00-11.90) 6.87(5.32-8.15) 4.31(0.82-5.10)

HBV BCP mutations

No(n=64) 10.50(6.70-12.00) 0.071 6.81(4.20-8.15) 0.805 4.61(1.26-5.10) 0.063

Yes(n=21) 9.80(6.00-12.30) 6.94(4.33-7.86) 4.00(0.82-5.10)

Serum HBV-DNA, log10 IU/mL

<8(n=49) 9.80(6.00-11.80) 0.004 6.69(4.33-7.79) 0.110 4.35(0.82-5.02) 0.000

≥8(n=36) 10.70(6.30-12.30) 6.95(4.20-8.15) 4.73(3.65-5.10)

This article is protected by copyright. All rights reserved.

ccepted Articl
Table.4 Correlation of intrahepatic cccDNA with HBcrAg, HBsAg and HBV RNA stratified by inflammatory grade and HBV DNA level among
HBeAg-positive patients

cccDNA cccDNA

HBV-DNA HBV-DNA
Parameter G<2 G≥2
<8 log10 IU/mL ≥8 log10 IU/mL

r p-value r p-value r p-value r p-value

Serum HBcrAg 0.852 0.000 0.743 0.000 0.848 0.000 0.828 0.000

Serum HBsAg 0.650 0.000 0.750 0.000 0.632 0.000 0.926 0.000

Serum HBV RNA 0.513 0.000 0.551 0.006 0.575 0.000 0.417 0.011

This article is protected by copyright. All rights reserved.

Accepted Article

This article is protected by copyright. All rights reserved.

Accepted Article

This article is protected by copyright. All rights reserved.