Chronic hepatitis B virus (HBV) infection is a major health problem, as approximately 240 million people are infected worldwide [
Long-term NA therapy can control chronic hepatitis B, but several problems remain. The first is the economical burden, especially in developing countries where health care and life insurance are poor. The second are poor adherence and retention in care of the patients, resulting from this high cost. The third is the side effect of the drugs, including renal and bone toxicity. The fourth are viral mutant and drug resistance which are further hindrances to drug efficacy. Finite NA therapy has the advantages of solving the above problems. How to make finite NA therapy is what physicians and patients concern.
Guidelines of prevention and treatment of chronic hepatitis B (CHB) from China [
This prospective study recruited 92 Chinese HAN outpatients with chronic hepatitis B in Third Affiliated Hospital of Sun Yat-sen University from December 2013 to January 2017. All patients were treated with nucleos(t)ide analogues, such as entecavir (ETV), telbivudine (LDT), lamivudine (LAM), adefovir dipivoxil (ADV), or a combination of LAM and ADV. Informed consent from the patients was obtained.
All patients were monitored every month for the first 3 months after cessation and every 3 months thereafter. Symptoms (e.g., fatigue, poor appetite, and jaundice), occurrence of liver cirrhosis or HCC, and mortality were all recorded for the study. Blood cells (white blood cells, red blood cells, hemoglobin, and platelets), biochemical parameters (serum aspartate transaminase [AST], alanine transaminase [ALT], total bilirubin, blood urea nitrogen [BUN], and creatinine), virologic parameters (quantitative HBsAg, HBeAg, HBeAb, and HBV DNA), T lymphocytes (CD4 positive T lymphocytes [CD4+T], CD8 positive T lymphocytes [CD8+T], Type 1 helper T lymphocytes [Th1], and Type 2 helper T lymphocytes [Th2]), and ultrasound results were assessed at every visit.
Routine automated techniques were used for all biochemical tests at our clinical laboratories. Serum HBsAg levels were measured using Roche Elecsys HBsAg II quant assay (range 0.05–52000 IU/mL, Roche Diagnostics, Mannheim, Germany). Serum HBeAg and HBeAb were assayed using the EIA kit (Abbott Diagnostics, North Chicago, IL). Serum HBV DNA levels were measured with real-time PCR using the COBAS AmpliPrep/COBAS TaqMan HBV Test, version 2.0 (detection limit: 20 IU/mL, Roche Molecular Systems, Inc., Branchburg, NJ, USA). T lymphocytes were measured using flow cytometry analysis with a BD Accuri C6 flow cytometer according to the manufacturer’s instructions.
According to guideline of prevention and treatment of chronic hepatitis B from APASL [
Ethical approval was provided by the Ethics Committee of Medical Clinical Trials, the Third Affiliated Hospital of Sun Yat-sen University (March 20, 2015).
Continuous data were indicated with mean ± SD while categorical data were reported as number and percentage (%). Spearman correlation coefficient was used to investigate the correlation among HBsAg, HBV DNA, and T lymphocytes. Nonparametric tests including Mann–Whitney
The outcome variables included the occurrences of virologic relapse or retreatment at baseline (month 0) and 1, 2, 3, 6, and 12 months after cessation. Associations between independent variables and outcome variables were analyzed using univariate/multivariate generalized estimating equation (GEE) and binary logistic regression models. An independent working correlation matrix was adopted for the repeated measure data. ROC analysis was further used to assess the diagnostic effectiveness of independent variables which were found associated with outcome. The statistical significance level for all the tests was set at a
A total of 92 patients treated with NA were recruited in this study. Nine of the patients were lost to follow-up, and 62 finished the 48 weeks of follow-up. Thirty-nine of the 62 patients were origin HBeAg positive before NA treatment; they gained HBeAg seroconversion before NA cessation. Twenty-three patients were origin HBeAg negative before NA treatment. Nineteen patients were treated with ETV for 4.55 ± 1.93 years before cessation, 23 patients were treated with LDT for 3.35 ± 1.34 years, 5 patients were treated with LAM for 5.02 ± 2.46 years, 7 patients were treated with ADV for 7.51 ± 2.65 years, and 8 patients were treated with combination of LAM and ADV for 4.94 ± 2.95 years. The flow of patient recruiting and clinical development was indicated in Figure
Baseline characteristics of 62 patients before NA treatment cessation.
origin HBeAg positive | origin HBeAg negative | All | |
---|---|---|---|
Age, year | 33.18 ± 8.33 | 44.39 ± 8.99 | 37.34 ± 10.11 |
Gender, male (%) | 26 (66.7) | 19 (82.6) | 45 (72.6) |
BMI, kg/cm2 | 21.31 ± 2.91 | 23.01 ± 3.12 | 21.95 ± 3.08 |
NA treatment, ratio of ETV : LDT : LAM : ADV : LAM + | 10 : 17 : 3 : 5 : 4 | 9 : 6 : 2 : 2 : 4 | 19 : 23 : 5 : 7 : 8 |
Duration of NA treatment, year | 4.64 ± 2.63 | 4.33 ± 1.74 | 4.52 ± 2.33 |
Duration of negative HBV DNA maintenance, year | 3.83 ± 2.15 | 3.68 ± 1.31 | 3.78 ± 1.87 |
Duration of HBeAg seroconversion maintenance, year | 3.19 ± 2.47 | - | - |
EOT HBsAg, | 2.65 ± 1.22 | 2.36 ± 1.38 | 2.54 ± 1.28 |
Flow chart of patient recruitment and clinical development.
In the course of the 48 weeks of follow-up, none of the 62 patients died or developed liver failure, cirrhosis, or hepatocellular carcinoma. Twenty-one (33.9%) patients were retreated with NA (origin NA or ETV or TDF) and regained normal ALT and undetectable HBV DNA within 24 weeks.
According to the development of patient’s levels of HBV DNA and ALT across time (from 0 to 48 weeks) and features of relapse in the follow-up, the 62 patients could be divided into 4 categories (Figures
Change of levels of ALT and HBV DNA in four categories from 0 to 48 weeks.
For all 62 patients who finished 48-week follow-up, the 48-week cumulative virologic relapse, clinical relapse, and retreatment rate were 62.9%, 38.7%, and 33.9%, respectively. For the 39 origin HBeAg positive patients, the 48-week cumulative virologic relapse, clinical relapse, and retreatment rate were 56.4%, 35.9%, and 30.8%, respectively. Sixteen of 22 (72.7%) virologic relapses and 11 of 14 (78.6%) clinical relapses occurred in the first 24 weeks in origin HBeAg positive patients. For the 23 origin HBeAg negative patients, the 48-week cumulative virologic relapse, clinical relapse, and retreatment rate were 73.9%, 43.5%, and 39.1%, respectively. Fourteen of 17 (82.4%) virologic relapses and 6 of 10 (60%) clinical relapses occurred in the first 12 weeks in origin HBeAg negative patients. The results were shown in Figure
Forty-eight-week cumulative rates of virologic relapse, clinical relapse, and retreatment.
To further investigate the association among HBsAg, HBV DNA, and flow cytometry results, correlation analyses were used. A negative correlation was found between HBsAg and CD4+T (
Changes in the ratios of CD4+T, CD8+T, Th1, and Th2 in nonrelapse and virologic relapse patients.
Univariate and multivariate logistic regression under GEE models were used to investigate the possibly predictive factor for virologic relapse or retreatment. Independent variables which were not significant in univariate results would still be modeled in multivariate model as adjustment of covariates. Only the variables which reached significance in both univariate and multivariate models would be recognized as possibly predictive factor to virologic relapse or retreatment.
As shown in Table
Independent variables associated with virologic relapse in GEE models.
Parameters | Univariate | Multivariate | ||
---|---|---|---|---|
OR (95% CI) | | OR (95% CI) | | |
Sex | ||||
Male | Ref | - | Ref | - |
Female | 1.12 (0.51–2.49) | 0.774 | 1.14 (0.59–2.19) | 0.704 |
Age, year | 1.05 (1.01–1.08) | 0.009 | 1.06 (1.02–1.10) | 0.003 |
BMI, kg/m2 | 0.97 (0.87–1.09) | 0.617 | 0.85 (0.75–0.95) | 0.006 |
Duration of NA treatment, year | 0.89 (0.74–1.08) | 0.255 | 1.40 (1.14–1.73) | 0.001 |
Duration of negative HBV DNA maintenance, year | 0.81 (0.65–1.01) | 0.066 | 0.52 (0.33–0.83) | 0.005 |
HBsAg, | 1.79 (1.32–2.45) | <0.001 | 2.21 (1.47–3.32) | <0.001 |
Origin HBeAg | ||||
Negative | Ref | - | Ref | - |
Positive | 0.38 (0.19–0.79) | 0.010 | 0.32 (0.14–0.74) | 0.008 |
The results of retreatment were indicated in Table
Independent variables associated with retreatment in GEE models.
Parameters | Univariate | Multivariate | ||
---|---|---|---|---|
OR (95% CI) | | OR (95% CI) | | |
Sex | ||||
Male | Ref | - | Ref | - |
Female | 1.75 (0.42–7.35) | 0.445 | 2.17 (0.45–10.43) | 0.332 |
Age, year | 1.06 (0.98–1.14) | 0.177 | 1.04 (0.92–1.18) | 0.490 |
BMI, kg/m2 | 1.00 (0.80–1.26) | 0.975 | 1.02 (0.78–1.34) | 0.878 |
Duration of NA treatment, year | 1.25 (0.81–1.95) | 0.314 | 1.08 (0.53–2.20) | 0.830 |
Duration of negative HBV DNA maintenance, year | 1.45 (0.90–2.33) | 0.130 | 1.11 (0.42–2.91) | 0.830 |
HBV DNA, | 1.28 (1.08–1.50) | 0.003 | 1.34 (1.13–1.58) | <0.001 |
HBsAg, | 0.69 (0.26–1.83) | 0.460 | 0.64 (0.24–1.72) | 0.382 |
Origin HBeAg | ||||
Negative | Ref | - | Ref | - |
Positive | 1.07 (0.30–3.80) | 0.921 | 1.23 (0.17–8.93) | 0.838 |
ROC analysis was further used to investigate the diagnostic effectiveness of these associated factors. As indicated in Figure
ROC curve and parameters of associated factors, including age to virologic relapse (a), the level of HBsAg to virologic relapse (b), and the level of HBsAg to clinical relapse (c).
In this study, under supervision, none of the patients in our study died or developed cirrhosis, liver failure, or hepatocellular carcinoma. The 48-week cumulative virologic relapse rates in origin HBeAg positive and negative CHB patients were 56.4% and 73.9%, respectively, which were similar results to a study [
Increased attention should be paid during the first 24 weeks of the follow-up after NA cessation in origin HBeAg positive patients, since 16 of 22 (72.7%) virologic relapses and 11 of 14 (78.6%) clinical relapses occurred in these first 24 weeks. Similarly, the first 12 weeks of follow-up after NA cessation also require increased vigilance in origin HBeAg negative patients, since 14 of 17 (82.4%) virologic relapses and 6 of 10 (60%) clinical relapses occurred in these first 12 weeks.
The 62 patients could be divided into four categories in this study. We designated Category A as “immune control,” since no relapse occurred in the category. We designated Category B as “immune retolerance,” since we assumed that the immune status of these patients is similar to “immune tolerance” in childhood during chronic HBV infection with a high load of HBV replication and normal ALT levels. We designated Category C as “immune recontrol,” since we assumed that host immunity blocked HBV replication without help from the NA therapy. We designated Category D as “immune reactivation,” since we assumed these patients were similar to naïve patients receiving NA therapy with elevated ALT and HBV DNA levels. The four different immune statuses may relate to the function of HBV-specific T cells, because the antiviral drug-induced attenuation of viremia provides a window for the reconstitution of the HBV-specific immune response [
In this study, we found a negative correlation between HBsAg and CD4+T, a positive correlation between HBsAg and CD8+T, and a negative correlation between HBV DNA and CD4+T. In patients with chronic HBV infection, cytotoxic T cell response is weak [
HBsAg clearance at the end of treatment indicated successful NA cessation in this study. Higher levels of HBsAg indicated increased risk of virologic relapse. Similar results were reported in some studies [
In this study, age was a predictive factor for virologic relapse. Some studies have shown that age > 40 years predicted relapse in origin HBeAg positive patients [
As mentioned above, there are difficulties to stop NA therapy without relapse. Discontinuation of NA therapy in CHB continues to be a hot topic with contrasting views in the recent liver meeting: patients may benefit from NA therapy cessation [
This study has some limitations. First, the case number was small, since we recruited patients from only one center. Second, pretreatment HBV genotype results were not carried out, since diagnostic reagents were scarce and HBV genotype was not a routine test in the past ten years in China which is still a developing country. Third, further study on HBV-specific cytotoxic T lymphocytes immune response was not carried out.
NA cessation is safe under supervision. Increased vigilance was required in the first 24 weeks in origin HBeAg positive patients and the first 12 weeks in negative patients. Age, HBsAg level before NA cessation, and origin HBeAg status before NA treatment can be predictive factors for virologic relapse. HBV DNA can be predictive factor for retreatment.
Nucleos(t)ide analogues
Hepatitis B virus
Chronic hepatitis B
Hepatocellular carcinoma
Entecavir
Telbivudine
Lamivudine
Adefovir dipivoxil
Tenofovir disoproxil fumarate
Hepatitis B surface antigen
Hepatitis B e antigen
Aspartate transaminase
Alanine transaminase
CD4 positive T lymphocytes
CD8 positive T lymphocytes
Type 1 helper T lymphocytes
Type 2 helper T lymphocytes
Upper limit of normal
End of treatment
Virological remission.
The authors declare that there are no conflicts of interest regarding the publication of this article.
Thanks are due to Pan Shun-wen and Li Wan-tao of the laboratory of Third Affiliated Hospital of Sun Yat-sen University. This study was supported by Plan of Science and Technology of Guangdong (nos. 411308023039 and 2016A020215221), Guangzhou Science and Technology Project (nos. 201508020118 and 2014Y2-00544), the National Natural Science Foundation of China (nos. 81570539 and 81370535), and the Fundamental Research Funds for the Central Universities (Precision Medicine and Biological Big Data) (no. 15ykjc21d).