Study on Post-Treatment Relapse in HBeAg Positive CHB Patients
Study on Post-Treatment Relapse in HBeAg Positive CHB Patients
Junfeng Lu 0 1
Jin'e Li 0 1
Yali Liu 0 1
Shan Ren 0 1
Zhenhuan Cao 0 1
Yi Jin 0 1
Lina Ma 0 1
Chengli Shen 0 1
Xinyue Chen 0 1
0 Editor: Joerg F. Schlaak, University Hospital of Essen , GERMANY
1 1 Beijing You'an Hospital, Capital Medical University , Beijing , China , 2 Department of Infectious Diseases and Microbiology, Graduate School of Public Health, University of Pittsburgh , Pittsburgh , United States of America
Many factors are associated with post-treatment relapse in CHB patients, and there are no effective factors to predict relapse. In this study, we investigate the influence factors associated with post-treatment relapse and their predictive value in HBeAg positive CHB (ePCHB).
The factors associated with post-treatment relapse were analyzed firstly by a retrospective
study in eP-CHB. Variables included age, sex, regimen, baseline HBeAg and HBV DNA
level, total course of treatment as well as duration of consolidation therapy after HBeAg
seroconversion. The predictive effects of the influence factors were evaluated in an
eP-CHB prospective cohort.
89 patients were enrolled in the retrospective study, 42(47.2%) relapsed after
discontinuation of treatment. Factors related to post-treatment relapse were total course of treatment,
duration of consolidation therapy and baseline HBV DNA level. Relapse rates in patients
with total course >36 months, consolidation duration >12 months and baseline HBV DNA
level < 1.0E+5IU/ml were lower than those of total course <24 months (P = 0.002),
consolidation duration 1 = 0.01) respectively. Patients with HBV DNA 1.0E+7IU/ml plus
HBeAg<200COI at baseline had the highest relapse rate and cumulative relapse rate than
the other three arms (P = 0.048 and 0.008 respectively). Logistic regression analysis
demonstrated that baseline HBV DNA level, duration of consolidation therapy and combination
of baseline HBV DNA and HBeAg (IgDNA/IgHBeAg) were independent factors to predict
posttreatment relapse. The model based on baseline IgDNA/IgHBeAg and consolidation duration
worked well in predicting post-treatment relapse in the prospective study and the accuracy,
specificity, sensitivity, PPV and NPV for prediction were 80.3%, 81.1%, 79.2%, 73.1% and
Competing Interests: The authors have declared
that no competing interests exist.
Virological factors including baseline HBV DNA, HBeAg and treatment course were major
influence factors associated with post-treatment relapse in eP-CHB. Patients with higher
HBV DNA and lower HBeAg levels at baseline, shorter total course as well as consolidation
therapy were more likely to develop relapse after discontinuation of therapy. The antiviral
therapy in eP-CHB patients should be individually managed at different levels. It is better to
treat those with higher viral load and lower HBeAg levels at baseline for a longer course,
especially longer consolidation duration so as to decrease the relapse rate.
Antiviral therapy is crucial for chronic hepatitis B (CHB) patients. With the rising rate of
patients withdrawing from antiviral treatment, a growing number of relapses are occurring.
HBeAg seroconversion was a widely used treatment endpoint for HBeAg positive CHB
(eP-CHB) [1–2]. However relapse is inevitable after discontinuing the antiviral agent. The
relapse rates in eP-CHB patients who acquired HBeAg loss/seroconversion with consolidation
therapy were approximately 40%-90% in nucleos(t)ide analogue (NA)-treated patients[3–5],
and 30%-40% in interferonα(IFNα)-treated patients[6–7], which might vary in different
studies. There are many factors influencing post-treatment relapse, including host factor, such as
age[8–9], treatment factors, such as regimen and course of treatment[4,6,8], and viral factors
such as genotype and viral load [10–11]. Reasons for the discrepancy may be due to different
study type (retrospective or prospective) and criterion of drug withdrawal. Moreover, there is
still lack of study on predictive effect of the influence factors by prospective study. Here, we
designed a retrospective and a prospective study to explore the influencing factors of
posttreatment relapse and their predictive effect in eP-CHB patients.
Retrospective study. In retrospective study, eight-nine Chinese CHB patients were enrolled
in Beijing You’an Hospital. The inclusion criteria were as follows: (1) Patients were diagnosed as
eP-CHB according to China’s guideline of prevention and treatment for chronic hepatitis B
(2010 version) with ALT>2ULN and HBV DNA>1.0E+4IU/ml; (2) IFNα(Peg IFNα or standard
IFNα), NAs or combination of the both had been used for antiviral treatment; (3) Consolidation
therapy was at least for 6 months after achieving HBV DNA<20IU/ml and HBeAg
seroconversion; (4) Regular follow-up was conducted after drug withdrawal.
Prospective study. Sixty-one eP-CHB patients with ALT>2ULN and HBV DNA>1.0E+
4IU/ml who were referred to Beijing You’an Hospital from January 2010 were recruited. The
patients received IFNα (Peg IFNαor standard IFNα), NAs or combination of the both
according to patient' s preference. Patients were excluded if (1) there was evidence of decompensated
cirrhosis, either diagnosed or suspected HCC; (2) they were co-infected with HAV, HCV,
HDV, HEV or HIV; (3) they accompanied with autoimmune hepatitis, alcoholic liver disease
and inherited metabolic liver disease. Consolidation treatment was at least for 6 months in
IFNα-treated and 1 year in NA-treated patients after achieving HBV DNA<20IU/ml and
Liver function, HBV markers and HBV DNA levels were conducted at an interval of 3 months
from the start of treatment to the end of follow-up. Liver function was assayed by routine
automated analysis system. HBV DNA levels were assayed using real-time PCR with the lower
detection limit of 20 IU/ml (COBAS Taqman, Roche diagnostics). HBeAg was measured by
the use of Elecsys MODULAR ANALYTICS E-170(Roche diagnostics).
Virological response was defined as an undetectable serum HBV DNA level (<20IU/ml).
Serological response was defined as HBeAg seroconversion. Relapse was defined as HBV DNA
level >1.0E+3IU/ml in any two consecutive visits with or without HBeAg reccurence.
Nonrelapsers were followed-up to over 6 months. Patients would withdraw from the study once
relapse were confirmed.
The analyses were performed using the Statistical Package for the Social Sciences (SPSS),
version 16.0. Normally distributed data were displayed as mean±SD, while non-normally
distributed data were displayed as medium(range). Student t tests and Wilcoxon's Sign rank tests
were used to perform grouped pair-wise comparison. Enumeration data were compared
usingχ2 tests. Potentially related factors were also tested by univariate and multivariate logistic
regression analysis. The Kaplan-Meier method was used to calculate the cumulative relapse
rate. Receiver operating characteristic curve (ROC curve) was used to analysis the relevant
predictive factor for relapse. Differences were considered significant at P<0.05 level.
The study was approved by the Human Ethics Committee of Beijing Youan Hospital, Capital
Medical University, Beijing, China. Informed written consent was obtained from each patient
in the study.
Factors associated with post-treatment relapse in retrospective study
Demographic characteristics. A total of 89 patients aging from 16–63 years were enrolled
in the study. Among them, 61 were male, 28 were female, 27 were NA-treated and 62 were
IFN-treated (including IFNαplus NA and IFNαmonotherapy). The total treatment course
ranged from 12 to 58 months with a median of 26 months and the period of followed-up ranged
from 6 to 43 months with a median of 11 months. 47.2% (42/89) relapsed after discontinuation
of treatment. Sex, age, total course, consolidation therapy, baseline HBV DNA and HBeAg
level are no significantly different between NA-treated and IFN-treated patients.
Relevant factors for post-treatment relapse. All possible factors associated with
posttreatment relapse were listed in Table 1. Firstly, stratified analysis revealed that total course,
consolidation treatment and baseline HBV DNA level were associated with relapse. The longer
the total course and consolidation treatment were, the lower the rate of relapse was. Relapse
rates in patients with total course >36 months, consolidation therapy >12 months were lower
than those of total course <24 months (23.1% vs 63.2%, P = 0.002) and consolidation
therapy 12 months (25.0% vs 55.4%, P = 0.011) respectively. Relapse rate was higher in
patients with baseline HBV DNA level>1.0E+07 IU/ml than those with <1.0E+05IU/ml
(59.0% VS 22.2%, P = 0.01). Patients with low baseline HBeAg level tend to have higher relapse
rate than those with high HBeAg level, but no significant difference was detected. Patients with
high HBV DNA plus low HBeAg levels at baseline had significantly higher relapse rate
compared with other HBV DNA and HBeAg combination groups (P = 0.048). However, there were
no significant differences in relapse rate between patients with different age, sex and regimens.
In addition, the factors related to relapse for NA and IFN treatment regimen were evaluated
separately (S1 Table). Consistent with the combined analysis results, the longer total treatment
course and longer consolidation treatment were associated with low relapse rate and high HBV
DNA plus low HBeAg levels was associated with high relapse rate in both NA and IFN
treatment groups. Because the sample size was relatively small, we did not do the formal statistical
Secondly, the seven factors above were individually introduced in univariate logistic
regression model. Serum HBeAg and HBV DNA levels were log10 transformed for analysis. Only
consolidation duration and baseline HBV DNA level (IgDNA) were statistically significant.
Although there was no statistical difference, baseline HBeAg level was negatively related to
relapse. Based on the consideration for viral replication and protein expression, we introduced
the concept of IgDNA/IgHBeAg in univariate regression model, which was referred to a combined
marker for baseline HBeAg and HBV DNA level. The result indicated that this factor was
statistically significant (Table 2). Further, IgDNA/IgHBeAg together with other factors was
introduced in multivariate logistic regression model, showing that consolidation therapy and
IgDNA/IgHBeAg were significantly correlated with relapse (Table 3).
Cumulative relapse rate. As displayed in Fig 1A, 1B and 1C, cumulative relapse rates in
patients with high baseline HBV DNA, low baseline HBeAg and consolidation treatment 12
months were higher than those with low HBV DNA, high HBeAg and consolidation therapy
>12 months, but the differences were not statistically significant between them. However,
patients with high HBV DNA level ( 1.0E+07IU/ml) plus low HBeAg level (<200COI) at
baseline had the highest cumulative relapse rate among all the four arms, as shown in Fig 1D
(P = 0.008).
Assessment of the predictive value of the relevant factors with relapse
by the prospective study
Demographic characteristics. A total of 61 eP-CHB patients aging from 17–61years were
enrolled in the prospective study and followed up over 6 months. Among them, 31 were male,
30 were female, 38 were IFN-treated (including IFNαplus NA and IFNαmonotherapy) and 23
were NA-treated. The mean age was 31.8±8.82 years. The median course of treatment was 25
(18–48) months and the median consolidation duration was 15(6–30) months. 39.3% (24/61)
relapsed after discontinuation of treatment. Also, there were no statistical differences in the
sex, age, total course, consolidation therapy, baseline HBV DNA and HBeAg level between
IFN-treated and NA-treated groups.
Prediction of post-treatment relapse. Based on retrospective data, Baseline IgDNA,
baseline IgDNA/IgHBeAg and duration of consolidation therapy were three independent factors
influencing post-treatment relapse. Subsequently, ROC curves of the IgDNA, IgDNA/IgHBeAg and
IgDNA/IgHBeAg combined with consolidation duration in differentiating relapse were drawn.
Area under the curves (AUC) was 0.678, 0.754 and 0.824 respectively (Fig 2A, 2B and 2C). The
IgDNA/IgHBeAg combined with consolidation duration had the biggest AUC, which had the
most significant value in predicting relapse. Date from the prospective study was introduced in
the ROC generated by IgDNA/IgHBeAg combined with consolidation duration. The optimized
cut-off value of 0.590 produced a sensitivity of 79.2%, specificity of 81.1%, positive predictive
value (PPV) of 73.1%, negative predictive value (NPV) of 85.7%, possibility of correct
classification of 80.3% (Table 4). The results support that baseline IgDNA/IgHBeAg combined with
consolidation duration are good prediction markers for post-treatment relapse.
Fig 1. Cumulative relapse rate among subgroups with different baseline factors. Cumulative relapse
rates in patients with baseline HBV DNA level > 1.0E+7IU/ml were higher than HBV DNA level between 1.0E
+5 IU/ml and 1.0E+7 IU/ml, and HBV DNA level < 1.0E+5IU/ml (A), with baseline HBeAg level < 200COI
higher than HBeAg level between 200COI and 1000COI, and HBeAg>1000COI (B), consolidation
treatment > 12 months lower than 12 months group (C), and HBV DNA level 1.0E+07IU/ml plus HBeAg
level < 200COI at baseline higher than the other three groups with statistically significant differences (D).
In recent years, post-treatment relapse has become the biggest challenge in clinical practice
over the world. Identifying the most efficient predictive factors for post-treatment relapse may
be a promising way to improve antiviral efficacy and reduce relapse.
In our study, we analyzed the possible factors associated with post-treatment relapse in
eP-CHB in the retrospective study. The results showed that total course of antiviral treatment,
duration of consolidation treatment after HBeAg seroconversion, baseline HBV DNA and
baseline IgDNA/IgHBeAg were related to post-treatment relapse, but age, sex and regimens had
no relationship with relapse.
Patients with longer total course (>36 months) and consolidation duration (>12 months)
were less likely to relapse after discontinuation of treatment. The latter was an independent
predictive factor for relapse, which was reported in previous studies. In the study from Pan, a
total of 136 NA-treated eP-CHB patients achieving treatment endpoints were enrolled. It was
Fig 2. ROC curves of the independent factors based on retrospective study for predicting relapse. ROC
curve of the IgDNA in differentiating post-treatment relapse with AUC 0.678 (A), ROC curve of IgDNA/IgHBeAg with
AUC 0.754 (B) and IgDNA/IgHBeAg combined with consolidation duration with AUC 0.824 (C).
reported that those with duration of consolidation therapy >11 months had lower relapse rate
than those with consolidation duration 11 months (26.0% VS 58.7%). In another study
on Peg IFNαin eP-CHB, extended course of treatment was associated with high sustained
response rate and lower relapse rate. As recommended in Chinese CHB guideline,
consolidation treatment for over 1 year and total treatment course for over 2 years are needed
before discontinuing NAs for eP-CHB patients achieving complete response, and the
recommended treatment course of Peg IFNαis only 1 year. In our study, 23.1% patients relapsed with
a total course of more than 36 months, and 25.0% relapsed with consolidation duration of
more than 12months. Thus, it is difficult to define a certain total course and consolidation
IgDNA/IgHBeAg combined with
treatment. We consider that the antiviral treatment for eP-CHB patients should be individually
optimized with set goals and different course. Total course, especially consolidation duration
after HBeAg seroconversion should be extended under the premise of safety and effectiveness
to achieve better treatment outcomes.
In our study, high HBV DNA level at baseline was associated with high relapse rate and
cumulative relapse rate. High baseline HBV DNA was an independent risk factor for
posttreatment relapse, which was consistent with the findings of the previous study on eP-CHB
patients treated with LAM [11,15]. The possible mechanism includes: 1) HBV DNA level is
positively correlated with intrahepatic cccDNA level and higher HBV DNA reflects the greater
amount of intrahepatic cccDNA storage, which is the source of relapse;2) High HBV DNA
level is indicative of active viral replication and insufficient host immunity for viral clearance.
The implementation of antiviral therapy at the moment increased the risk of drug resistance. It
is recommended by both CHB guidelines of EASL and AASLD that NA with potent
antiviral efficacy and low risk of resistance should be used. In the Chinese experts’
recommendation, initial combination therapy should be considered to lower the risk of post-treatment
relapse in patients with high viral load.
Baseline HBeAg level has long been recognized to be associated with HBeAg seroclearence
with IFN  or NAs treatment in eP-CHB. The lower HBeAg level at baseline was, the
more likely HBeAg seroclearence could be achieved and vice versa. In our study, we were
astounded to discover that low HBeAg level at baseline was associated with high
post-treatment relapse. Usually, low viral load produced low HBeAg level indicative of balance between
viral replication and protein expression. However, in our study, HBV DNA level seems to be
not consistent with HBeAg level. Therefore, we performed a analysis in patients with different
baseline HBV DNA and HBeAg levels. The results showed that patients with HBV DNA 1.0E
+07IU/ml plus HBeAg<200COI had the highest relapse and cumulative relapse rates among
all the four groups. IgDNA/IgHBeAg defined as HBV DNA combined with HBeAg level and
better reflecting the baseline virological characteristics was demonstrated to be an independent
predictive factor for post-treatment relapse. Higher HBV DNA level and lower HBeAg level
were associated with greater risk of relapse. The reason may be due to the mutations at pre-C/
BCP in HBV genome, which is a common strategy for HBV to escape from host immune
In the latter part of this study, we drawn three ROC curves based on the results of the
retrospective study and tested the predictive effect in the prospective study. AUC from baseline
IgDNA/IgHBeAg plus consolidation duration was greater than that from baseline IgDNA/IgHBeAg
and baseline HBV DNA level alone, which had a sensitivity of 79.2%, specificity of 81.1%, PPV of
73.1%, NPV of 85.7% in predicting relapse in prospective study. This indicated that it would be
better to take both virological factors and treatment factors into account in prediction of
posttreatment relapse. In this study, NPV which predicted the probability of not developing relapse
after drug withdrawal provided a basis that partial patients could discontinue treatment in
limited treatment course and avoid adverse events from extended course of treatment and waste of
medical resource. On the other hand, PPV can be used to find patients with high risk of relapse,
preventing relapse by extending the treatment course. Among the three factors, baseline IgDNA/
IgHBeAg is related to viral replication and viral expression. However, consolidation duration is
more essential factor predictive of post-treatment relapse, which is not only the predictive factor
of relapse but also the method of reducing relapse by extending the consolidation duration.
In summary, for eP-CHB patients, low HBeAg level and high HBV DNA level at baseline
are associated with high post-treatment relapse rate. Patients with longer total course of
treatment as well as consolidation treatment have lower post-treatment relapse rate. Baseline
IgDNA/IgHBeAg combined with duration of consolidation therapy has a good predictive value of
post treatment relapse. We hereby suggest that antiviral therapy for eP-CHB patients should be
individualized upon baseline HBeAg and HBV DNA levels and for those with high viral load
and low HBeAg levels, extending treatment course and duration of consolidation treatment
after HBeAg seroconversion should be considered.
Conceived and designed the experiments: JFL XC. Performed the experiments: JEL SR.
Analyzed the data: JFL JEL YL. Contributed reagents/materials/analysis tools: ZC YJ LM. Wrote
the paper: JFL. Revised the manuscript: CS XC.
1. European Association for the Study of the Liver ( 2012 ) EASL clinical practice guidelines: Management of chronic hepatitis B virus infection . J Hepatol 57 : 167 - 185 . doi: 10.1016/j.jhep. 2012 . 02.010 PMID: 22436845
2. Liaw YF , Kao JH , Piratvisuth T , Chan HLY , Chien RN , Liu CJ , et al. ( 2012 ) Asian-Pacific consensus statement on the management of chronic hepatitis B: a 2012 update . Hepatol Int 6 : 531 - 561 . doi: 10. 1007/s12072- 012 - 9365 - 4 PMID: 26201469
3. Sohn HR , Min BY , Song JC , Seong MH , Lee SS , Jang ES , et al. ( 2014 ) Off-treatment virologic relapse and outcomes of re-treatment in chronic hepatitis B patients who achieved complete viral suppression with oral nucleos(t)ide analogs . BMC Infect Dis 14 : 439 doi: 10.1186/ 1471 - 2334 - 14 -439 PMID: 25125320
4. Song MJ , Song do S , Kim HY , Yoo SH , Bae SH , Choi JY , et al. ( 2012 ) Durability of viral response after off-treatment in HBeAg positive chronic hepatitis B . World J Gastroenterol 18 : 6277 - 6283 . doi: 10. 3748/wjg.v18. i43.6277 PMID: 23180949
5. Chaung KT , Ha NB , Trinh HN , Garcia RT , Nguyen HA , Nguyen KK , et al. ( 2012 ) High frequency of recurrent viremia after hepatitis B e antigen seroconversion and consolidation therapy . J Clin Gastroenterol 46 : 865 - 870 . doi: 10.1097/MCG.0b013e31825ceed9 PMID: 22941429
6. van Nunen AB , Hansen BE , Suh DJ , Löhr HF , Chemello L , Fontaine H , et al. ( 2003 ) Durability of HBeAg seroconversion following antiviral therapy for chronic hepatitis B: relation to type of therapy and pretreatment serum hepatitis B virus DNA and alanine aminotransferase . Gut 52 : 420 - 424 . PMID: 12584227
7. Flink HJ , Buster EH , Merican I , Nevens F , Kitis G , Cianciara J , et al. ( 2007 ) Relapse after treatment with peginterferon alpha-2b alone or in combination with lamivudine in HBeAg positive chronic hepatitis B . Gut 56 : 1485 - 1486 . PMID: 17872587
8. Lee HW , Lee HJ , Hwang JS , Sohn JH , Jang JY , Han KJ , et al. ( 2010 ) Lamivudine maintenance beyond one year after HBeAg seroconversion is a major factor for sustained virologic response in HBeAg-positive chronic hepatitis B . Hepatology 51 : 415 - 421 . doi: 10.1002/hep.23323 PMID: 19902424
9. Wang L , Liu F , Liu YD , Li XY , Wang JB , Zhang ZH , et al. ( 2010 ) Stringent cessation criterion results in better durability of lamivudine treatment: a prospective clinical study in hepatitis B e antigen-positive chronic hepatitis B patients . J Viral Hepat 17 : 298 - 304 . doi: 10.1111/j.1365- 2893 . 2009 .01178.x PMID: 19758278
10. Chien RN , Yeh CT , Tsai SL , Chu CM , Liaw YF ( 2003 ) Determinants for sustained HBeAg response to lamivudine therapy . Hepatology 38 : 1267 - 1273 . PMID: 14578866
11. Yeh CT , Hsu CW , Chen YC , Liaw YF ( 2009 ) Withdrawal of lamivudine in HBeAg-positive chronic hepatitis B patients after achieving effective maintained virological suppression . J Clin Virol 45 : 114 - 118 . doi: 10.1016/j.jcv. 2009 . 04.006 PMID: 19451024
12. Pan X , Zhang K , Yang X , Liang J , Sun H , Li X , et al. ( 2013 ) Relapse rate and associated-factor of recurrence after stopping NUCs therapy with different prolonged consolidation therapy in HBeAg positive CHB patients . PLoS One 8: e68568. doi: 10.1371/journal.pone.0068568 PMID: 23844222
13. Cao ZH , Ma LN , Zhang HW , Liu YL , Chen XY ( 2013 ) Extended treatment with peginterferon α-2a in combination with lamivudine or adefovir for 96 weeks yields high rates of HBeAg and HBsAg seroconversion . J Dig Dis 14 : 446 - 450 . doi: 10.1111/ 1751 - 2980 . 12065 PMID: 23615131
14. Chinese Society of Hepatology and Chinese Society of Infectious Disease, Chinese Medical Association ( 2011 ) The guideline of prevention and treatment for chronic hepatitis B (2010 version) . Chin J Hepatol 19 : 13 - 21 . [Article in Chinese]
15. Song BC , Suh DJ , Lee HC , Chung YH , Lee YS ( 2000 ) Hepatitis B e antigen seroconversion after lamivudine therapy is not durable in patients with chronic hepatitis B in Korea . Hepatology 32 : 803 - 806 . PMID: 11003626
16. Lok AS , McMahon BJ ( 2009 ) Chronic hepatitis B: update 2009 . Hepatology 50 : 661 - 662 . doi: 10.1002/ hep.23190 PMID: 19714720
17. Fried MW , Piratvisuth T , Lau GK , Marcellin P , Chow WC , Cooksley G , et al. ( 2008 ) HBeAg and hepatitis B virus DNA as outcome predictors during therapy with peginterferon alfa-2a for HBeAg-positive chronic hepatitis B . Hepatology 47 : 428 - 434 . doi: 10.1002/hep.22065 PMID: 18220290
18. Xie DY , Lin BL , Xu QH , Chen YM , Lu WL , Li JG , et al. ( 2008 ) The predictive value of ALT, HBeAg and HBV DNA levels at baseline and the degree of HBV suppression at week 12 adefovir dipivoxil treatment to the efficacy of it at week 52 in patients with HBeAg-positive chronic hepatitis B . Zhonghua Gan Zang Bing Za Zhi 16 : 341 - 344 . [Article in Chinese] PMID: 18510844