If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Faculty of Medicine, Hasanuddin University, Makassar, South Sulawesi, IndonesiaEijkman Institute for Molecular Biology, Jl. Diponegoro 69, Jakarta Pusat 10430, DKI Jakarta, IndonesiaSydney Medical School, University of Sydney, Sydney, New South Wales, Australia
Perinatal transmission of hepatitis B virus (HBV) still occurs despite infant hepatitis B vaccination.
•
Maternal HBV viraemia is the main risk factor for perinatal transmission.
•
The hepatitis B surface antigen (HBsAg) level is a good viral load predictor in hepatitis B e antigen (HBeAg)-positive pregnant women.
•
HBeAg-negative status could be due to HBV variants defective for HBeAg production.
•
HBV DNA quantification is not required for pregnant women with low HBsAg levels.
Summary
Background
New cases of hepatitis B virus (HBV) infection continue to occur worldwide. Most of these are due to mother-to-child transmission (MTCT), with maternal viraemia as the most important contributing factor. The hepatitis B surface antigen (HBsAg) level, which correlates positively with viral load, has been used for treatment monitoring in chronic hepatitis B. This study evaluated the usefulness of quantitative HBsAg for viral load prediction in HBsAg-positive pregnant women.
Methods
A total of 943 pregnant women in Makassar, Indonesia, were screened for HBsAg. Sixty-four women were HBsAg-positive and investigated. HBsAg level and hepatitis B e antigen (HBeAg)/hepatitis B e antibody (anti-HBe) status were determined serologically. Viral load was measured by real-time PCR. HBV DNA was sequenced and analysed for identification of genotype and basal core promoter (BCP)/precore (PC) mutations.
Results
Of 64 subjects, 12 (18.8%) were HBeAg-positive and 52 (81.3%) were HBeAg-negative. HBsAg and HBV DNA levels were significantly higher in the HBeAg-positive group (p < 0.001). HBsAg and HBV DNA levels were positively correlated in the HBeAg-positive group (r = 0.659; p = 0.02), but not in the HBeAg-negative group (r = 0.194; p = 0.168). Low HBsAg levels (<3.0 log10 IU/ml) corresponded with HBV DNA levels < 6.0 log10 IU/ml (r = 0.404; p = 0.001), a recognized threshold for MTCT. Genotype C was more prevalent than genotype B, but not associated with HBsAg level, viral load, or HBeAg status. Two-thirds of HBeAg-negative subjects with high HBV DNA levels harboured BCP (A1762T/G1764A) and/or PC (G1896A) variants.
Conclusions
HBsAg levels provide a good viral load predictor in HBeAg-positive but not HBeAg-negative pregnant women. The HBeAg-negative group had a frequent occurrence of BCP/PC variants, which may have contributed to the lack of correlation observed. Samples with a low HBsAg level, which is associated with a low risk of MTCT, do not require HBV DNA measurement.
More than 240 million people worldwide are chronically infected with the hepatitis B virus (HBV), and about 780 000 die from hepatitis B annually. In highly endemic areas of Asia, the Pacific, and Sub-Saharan Africa, most HBV infection occurs perinatally or during early childhood. This is associated with a high rate of persistent infection and increased risk of morbidity and mortality from cirrhosis and hepatocellular carcinoma later in life.
Efforts in the prevention of hepatitis B have focused on the immunization of infants implemented in 183 World Health Organization (WHO) member states. As of 2012, 94 member states including Indonesia had introduced the hepatitis B birth dose immunization.
However, 50 million new cases of HBV infection continue to be diagnosed annually, with the highest incidence due to mother-to-child transmission (MTCT). Despite the administration of hepatitis B immunization (active) or active plus hepatitis B immune globulin (HBIG) at birth, at least 10% of infants born to HBV-carrying mothers still suffer HBV infection.
Several factors such as maternal serum HBV DNA level, hepatitis B e antigen (HBeAg) status, HBV S gene variation, mode of delivery, and neonatal immune deficiency have been related to MTCT. Of these factors, maternal HBV DNA level has been identified as the most relevant.
Practice guidelines from major professional associations address the decision for antiviral treatment in pregnant women based on the HBV DNA threshold.
Assays for HBV DNA quantification with high sensitivity and specificity are currently available. However, the routine application of these methods for screening pregnant women is hampered by cost and limited resources. In recent years, the hepatitis B surface antigen (HBsAg) level, which correlates positively with viral load, has been used as a biomarker to predict disease status and to monitor the treatment response in chronic hepatitis B (CHB).
One attraction of the use of the HBsAg level is that the assay is less costly, is suitable for high-throughput screening, and its platforms are commonly used in many laboratories. However, some studies have reported that HBsAg quantification correlates poorly with the HBV DNA level. HBsAg and HBV DNA levels may vary during different phases of CHB and the correlation is associated with the HBeAg status of the patient.
Serum hepatitis B surface antigen and hepatitis B e antigen titers: disease phase influences correlation with viral load and intrahepatic hepatitis B virus markers.
The presence of these variants needs to be taken into account since they often occur in endemic areas and may be associated with certain HBV genotypes that differ among geographical regions.
This study aimed to evaluate the usefulness of quantitative HBsAg as a viral load predictor in pregnant women with CHB in Makassar, Indonesia, and to analyse the association of the HBsAg level and viral load with the HBeAg status, as well as the molecular characteristics of HBV variants defective for HBeAg production.
2. Materials and methods
2.1 Study population
This cross-sectional study was carried out from January to July 2014 in the antenatal care units of Wahidin Sudirohusodo Hospital, Hasanuddin University Hospital, Fatimah Mother and Child Hospital, Pertiwi Mother and Child Hospital, Labuang Baji Hospital, and Ibnu Sina Hospital, as well as several maternity clinics, in Makassar, South Sulawesi. A total of 943 pregnant women were screened for HBV infection; 64 of them were HBsAg-positive and considered eligible for enrolment. For inclusion it was required that the woman had been HBsAg-positive for >6 months without prior antiviral therapy.
Subjects co-infected with hepatitis A virus, hepatitis C virus, or human immunodeficiency viruses, as well as those with evidence of liver diseases, were excluded. This study was approved by the Ethics Committee of the Faculty of Medicine, Hasanuddin University, Makassar, Indonesia. Written informed consent was obtained from each patient.
2.2 Serological examination
HBsAg status was determined by VIDAS HBsAg immunoassay (bioMérieux SA, Marcy l’Etoile, France). HBeAg and hepatitis B e antibody (anti-HBe) were tested using Monolisa HBeAg-Ab PLUS immunoassay (Bio-Rad, Marnes-la-Coquette, France). HBsAg quantification was done using Elecsys HBsAg Quant II (Roche Diagnostics, Indianapolis, USA) on a Roche Cobas e411 Immunoanalyzer following the manufacturer's protocol.
2.3 HBV DNA detection and analysis
The HBV DNA level was determined from 500 μl of serum by quantitative real-time PCR (CobasTaqman HBV Test; Roche Diagnostics, Indianapolis, USA) with a range of linearity between 6 and 1.1 × 108 IU/ml. HBV DNA for molecular analysis was obtained by extracting the DNA from 140 μl of serum using the QIAamp DNA Mini Kit (Qiagen, Valencia, CA, USA) and amplification by nested PCR using primers S2-1/S1-2 for the first-round and S88/S2-2 for the second-round (Supplementary Material, Table S1).
Amplicons were purified using a PCR purification column (Qiagen, Valencia, CA, USA) and subjected to direct sequencing on a DNA sequence analyzer ABI 3130xl (Applied Biosystems, Carlsbad, CA, USA).
The HBV genotype was determined by phylogenetic analysis based on the 226-nucleotide sequences of the S gene compared with 70 reference sequences of known genotypes (A–H) retrieved from GenBank, using Phylip 3.68 software with the Kimura 2-parameter model, neighbour-joining algorithm, and 1000 bootstrapping.
Amplification of BCP and PC regions was done by nested PCR using primers PC1/PC2 for the first-round and S012/S013 for the second-round (Supplementary Material, Table S1). Amplicons were purified and sequenced as described previously.
The sequences were aligned with reference sequence retrieved from GenBank.
2.5 Statistical analysis
The baseline data were summarized descriptively. Continuous and categorical variables were compared between groups using the Mann–Whitney test and Chi-square/Fisher's exact test, respectively. Pearson's correlation coefficient was used to describe the correlation between two continuous, normally distributed variables. Spearman's correlation was used for categorical variables or continuous variables that were not normally distributed.
Serum hepatitis B surface antigen and hepatitis B e antigen titers: disease phase influences correlation with viral load and intrahepatic hepatitis B virus markers.
Statistical analyses were performed using IBM-SPSS v. 20 software (IBM Corp., Armonk, NY, USA). All statistical significance values were assessed at p < 0.05.
3. Results
3.1 Characteristics of study subjects
Among 943 pregnant women attending several antenatal clinics in Makassar, 64 (6.8%) were HBsAg-positive. Of these women, 12 (18.8%) were HBeAg-positive and 52 (81.3%) were HBeAg-negative. HBV DNA levels were significantly higher in the HBeAg-positive group (median 7.43 log10 IU/ml) than in the HBeAg-negative group (median 1.55 log10 IU/ml) (p < 0.001). Similarly, HBsAg levels were significantly higher in the HBeAg-positive group (median 4.21 log10 IU/ml) than in the HBeAg-negative group (median 2.91 log10 IU/ml) (p < 0.001). Age, alanine aminotransferase (ALT) levels, and HBV genotype distribution were comparable in the two groups (Table 1 and Supplementary Material Table S2).
Table 1Baseline characteristics of HBsAg-positive pregnant women
3.2 Distribution of serum HBV DNA levels among HBsAg-positive pregnant women according to HBeAg status
HBV DNA levels were categorized into <3.0 log10 IU/ml (close to 3.3 log10 IU/ml or 2000 IU/ml, which is the threshold to define inactive carrier state of CHB),
The proportions of pregnant women with HBV DNA levels <3.0 log10 IU/ml, 3.0–6.0 log10 IU/ml, and ≥6.0 log IU/ml in the HBeAg-positive group were 9% (1/12), 25% (3/12), and 67% (8/12), respectively; in the HBeAg-negative group, the proportions were 83% (43/52), 13% (7/52), and 4% (2/52), respectively (Figure 1A) .
Figure 1Distribution of (A) HBV DNA and (B) HBsAg levels among pregnant women according to HBeAg status: (i) among the HBeAg-positive group (n = 12) and (ii) among the HBeAg-negative group (n = 52).
The proportions of subjects with HBsAg levels <3.0 log10 IU/ml, 3.0–4.0 log10 IU/ml, and ≥4.0 log10 IU/ml were 0% (0/12), 42% (5/12), and 58% (7/12), respectively, in the HBeAg-positive group, and 58% (30/52), 35% (18/52), and 8% (4/52), respectively, in the HBeAg-negative group (Figure 1B).
3.4 Correlation between HBsAg and HBV DNA levels
In all 64 HBsAg-positive subjects, serum HBsAg and HBV DNA levels showed a significantly moderate correlation (r = 0.513; p < 0.001) (Figure 2A) . When analysed separately according to HBeAg status, there was a strong correlation between HBV DNA and HBsAg levels in the HBeAg-positive group (r = 0.659; p = 0.02) (Figure 2B), but no correlation was observed in the HBeAg-negative group (r = 0.194; p = 0.168) (Figure 2C).
Figure 2Correlation of HBsAg and HBV DNA levels in pregnant women according to HBeAg status: (A) overall correlation in all pregnant women (n = 64); (B) correlation of HBsAg and HBV DNA levels in HBeAg-positive women (n = 12); (C) correlation of HBsAg and HBV DNA levels in HBeAg-negative women (n = 52). Both the HBV DNA and HBsAg levels are calculated in log10 IU/ml. The correlation is regarded very weak for r = 0–0.19, weak for r = 0.20–0.39, moderate for r = 0.40–0.59, strong for r = 0.60–0.79, and very strong for r = 0.80–1.
There were four subjects (M150, M414, M415, and M818) with high HBsAg levels (>4.0 log10 IU/ml) but low HBV DNA levels (<3.0 log10 IU/ml). In contrast, there were four subjects (M167, M173, M258, and M810) with low HBsAg levels (<3.0 log10 IU/ml) but who had moderate viraemia (3.0–6.0 log10 IU/ml). However, in most cases, low levels of HBsAg were associated with low levels of HBV DNA; HBsAg levels <3.0 log10 IU/ml were significantly correlated to HBV DNA levels <3.0 log10 IU/ml (r = 0.363; p = 0.003) and to HBV DNA levels <6.0 log10 IU/ml (r = 0.404; p = 0.001). No subjects with HBsAg levels <3.0 log10 IU/ml had HBV DNA ≥6.0 log10 IU/ml (Figure 3).
Figure 3Distribution of HBV DNA levels among pregnant women according to HBsAg levels. In most cases, low levels of HBsAg were associated with low levels of HBV DNA; HBsAg levels <3.0 log10 IU/ml were significantly correlated to HBV DNA levels <3.0 log10 IU/ml (r = 0.363; p = 0.003) and to HBV DNA levels <6.0 log10 IU/ml (r = 0.404; p = 0.001). No subjects with HBsAg levels <3.0 log10 IU/ml had HBV DNA ≥6.0 log10 IU/ml.
HBV genotype was successfully determined in 47 pregnant women based on the S gene sequences (sequencing was not possible in the other subjects because of insufficient HBV DNA content). The sequences generated have been deposited in the GenBank database (accession numbers KP241791–KP241837). Median HBsAg and HBV DNA levels were not significantly different in each genotype. Genotype C was more prevalent than genotype B (83% vs. 17%), but the proportions were comparable between the HBeAg-positive and HBeAg-negative groups (Figure 4).
Figure 4Distribution of serum HBV DNA (A) and HBsAg (B) levels in subjects with HBV genotype B (n = 8) and HBV genotype C (n = 39). HBV DNA and HBsAg levels were comparable in the two genotypes. Median values (max, min; log10 IU/ml) with 95% confidence intervals are shown.
3.6 Subgroup analysis for BCP and PC mutations in HBeAg-negative pregnant women
Sequencing of the BCP/PC region of the HBV genome was performed successfully in 12 HBeAg-negative subjects (sequencing was not possible in the other subjects because of low HBV DNA content). The sequences generated have been deposited in GenBank (accession numbers KP241838–KP241844 and KP241846–KP241850). Four subjects had the wild-type HBV DNA sequence at both the BCP and PC sites, four had the BCP mutation alone, and two had the PC mutation alone. Concurrent BCP and PC mutations were detected in two subjects. All subjects with the BCP mutation had HBV genotype C, while the PC mutation was detected in two subjects with genotype B and one subject with genotype C (Table 2).
Table 2BCP/PC mutations, HBV DNA and HBsAg levels, and HBV genotype in 12 HBeAg-negative pregnant women
Subject code
BCP mutation A1762T/G1764A
PC mutation G1896A
HBV DNA (log10 IU/ml)
HBsAg (log10 IU/ml)
Genotype
M177
+
+
6.48
3.46
C
M384
+
−
6.08
3.04
C
M167
−
+
4.55
2.69
B
M810
−
−
3.50
2.70
C
M218
+
−
3.29
3.56
C
M336
+
+
2.70
3.74
C
M150
−
−
2.62
4.00
C
M898
−
+
2.08
3.04
B
M253
−
−
1.36
3.05
C
M818
+
−
1.34
4.11
C
M19A
−
−
0.78
2.51
C
M212
+
−
0.78
3.55
C
BCP, basal core promoter; PC, precore; HBV, hepatitis B virus; HBsAg, hepatitis B surface antigen; HBeAg, hepatitis B e antigen.
This study represents one of few reports of HBV infection in pregnant women from Indonesia. Of 943 pregnant women attending several antenatal clinics in Makassar, 64 (6.8%) were HBsAg-positive. This figure is higher than that reported recently from Jakarta (2.2%),
The wide variation in HBV infection rates may be associated with the general HBsAg prevalence in Indonesia (3.4–19.5%), geographical variation, and differences in cultural practices, as well as the methods used to detect HBV infection.
This fact is of concern, because it occurs in pregnant women who tend to be in the immune-tolerant phase of CHB with normal physical/laboratory examinations and high-level viraemia, but unaware of their HBsAg-positive status.
Varying thresholds of maternal HBV DNA have been discussed in association with MTCT and immunoprophylaxis failure. Wiseman reported that immunoprophylaxis failure occurred in infants when the maternal viral load was ≥8 log10 copies/ml (>1.7 × 107 IU/ml).
When the mothers’ HBV DNA levels were stratified to <6, 6–6.99, 7–7.99, and ≥8 log10 copies/ml, the corresponding rates of immunoprophylaxis failure were 0%, 3.2%, 6.7%, and 7.6%, respectively, and it was concluded that an antenatal HBV DNA level >6 log10 copies/ml (>200 000 IU/ml) was the most important predictor of MTCT.
The European Association for the Study of the Liver (EASL) and the Asian Pacific Association for the Study of the Liver (APASL) guidelines recommend treating pregnant women when HBV DNA levels are >2 × 106 IU/ml in the third trimester for the prevention of MTCT.
In the present study, 15.6% (10/64) of all subjects had HBV DNA levels >6.0 log10 IU/ml, distributed in 67% (8/12) of the HBeAg-positive group and 4% (2/52) of the HBeAg-negative group. The fact that the subjects had a skewed distribution toward the higher levels of HBV DNA should be regarded as important, as this shows a higher possibility of MTCT.
There are few studies on the potential applications of quantitative HBsAg in the management of hepatitis B during pregnancy.
To the best of the authors’ knowledge, this is the first study to evaluate the relationship between HBsAg and HBV DNA levels with regard to HBeAg status in pregnant women with CHB in Indonesia. This study revealed that among all 64 HBsAg-positive pregnant women, the HBsAg level was correlated with the HBV DNA level regardless of age and viral genotype. When stratified based on HBeAg status, the correlation was strong in HBeAg-positive pregnant women but missing in the HBeAg-negative women. A possible explanation for this finding is that HBsAg synthesis has a pathway distinct from HBV DNA synthesis and under the influence of different immune-control mechanisms.
Pregnant women with an HBeAg-positive status could be in the immune-tolerant phase where HBV virions and their antigens are minimally subjected to the host immune reaction, while those with an HBeAg-negative status could be in the low-replicative phase of CHB where the number of virions has decreased as a result of successful immune control.
Therefore, the reduction in HBsAg levels was not proportional to that of HBV DNA levels. The ratio of HBsAg/HBV DNA reflects the association between HBsAg production and HBV replication; this was significantly higher in HBeAg-negative subjects than in the HBeAg-positive group (median 1.43 vs. 0.61 log10 IU/ml, respectively).
Four pregnant women had high HBsAg levels with low HBV DNA. As explained, this could be due to the larger excess of HBsAg over the number of virions.
This implies that a high level of HBsAg cannot be used to predict the HBV DNA level. The serum HBsAg level should thus be used together with, but not as a substitute for, HBV DNA.
Four other subjects had low HBsAg levels but moderate viraemia (>4 log10 IU/ml). Possible explanations for the decreased detection of HBsAg include (1) differences in analytical sensitivity and specificity in HBsAg detection of viruses of different genotypes; (2) mutations in the pre-S/S gene that cause HBsAg detection failure; (3) treatment-associated mutations that cause derangement of the P gene with subsequent alteration of the overlapping S gene; or (4) the concomitant presence of hepatitis B surface antibodies (anti-HBs) leading to the formation of immune complexes poorly displaced by HBsAg-capture antibodies.
In the majority of subjects, however, low levels of HBsAg (<3.0 log10 IU/ml) correlated significantly with low levels of HBV DNA (<6.0 log10 IU/ml) and an attendant lower risk of MTCT.
This result suggests that HBV DNA quantification may not be necessary for pregnant mothers with low HBsAg levels.
Another important finding from this study was the presence of HBeAg-negative pregnant women with high viraemia. These women, with no evidence of liver disease, were apparently in the inactive carriers of CHB. However, some inactive carriers may have high HBV DNA levels accompanied by persistently normal ALT levels.
Studies have documented that certain HBV variants with nucleotide substitutions in the PC and/or BCP regions could abolish or down-regulate HBeAg production. These variants may replicate rapidly in HBeAg-negative CHB, where HBsAg and HBV DNA levels are preserved.
Serum hepatitis B surface antigen and hepatitis B e antigen titers: disease phase influences correlation with viral load and intrahepatic hepatitis B virus markers.
Analysis for the presence of BCP (A1762T/G1764A) and PC (G1896A) mutations was performed on HBeAg-negative subjects who had HBV DNA levels >2000 IU/ml. A substantial portion of isolates analysed, particularly from subjects with HBV DNA levels >6.0 log10 IU/ml, had BCP and PC mutations either alone or in combination. This finding is important because the emergence of these variants frequently occurs in regions with HBV endemicity.
Notably, all BCP mutants identified had genotype C, which was prevalent among the subjects studied. It has been recognized that HBV genotype C and BCP mutations are independent risk factors for progression to severe liver disease.
Important limitations of the present study include the relatively small number of pregnant women with chronic HBV infection. Wider clinical and community-based studies in different areas of the Indonesian archipelago will be necessary to estimate the true national burden of HBV infection in pregnant women. Also, the cross-sectional design of the study could not represent the fluctuating profile of CHB. Serial measurements of ALT levels are necessary to distinguish the true inactive carriers from active HBeAg-negative individuals. However, most studies have reported that ALT levels are lower during pregnancy and viral load is more likely to increase due to the natural immune suppression processes linked to pregnancy.
In conclusion, this study confirms that serum HBsAg level may be used as a predictor of the serum HBV DNA level in HBeAg-positive pregnant women, but not in HBeAg-negative pregnant women. The measurement of HBV DNA is not necessary if the level of HBsAg is low, since the probability of detecting a high viral load is low. An important role of BCP/PC variants influencing HBeAg status independent of viral load was also identified, providing a cautionary note to the interpretation of negative results of HBeAg testing when classifying HBV-infected individuals. These results offer the promise of practical guidance in using quantitative HBsAg as a tool to better manage HBV-infected pregnant women. Follow-up studies are needed to assess the impacts of the maternal and virological factors discussed in this study on HBV carriage and immunoprophylaxis failure in the infants born to these HBsAg-positive mothers.
Acknowledgements
The authors would like to convey special thanks to the pregnant women who participated in this study. The authors would also like to express their gratitude to the Eijkman Institute for Molecular Biology in Jakarta for the molecular work support, the Hasanuddin University Medical Research Center (HUM-RC) for laboratory facilities, several maternity clinics in Makassar, and the Wahidin Sudirohusodo Teaching Hospital, as well as the Academic Health Center of Hasanuddin University with their staff for the recruitment of study subjects. Special thanks are given to Dr Kevin Baird from the Eijkman–Oxford Clinical Research Unit in Jakarta for proofreading this manuscript.
Funding: This study was supported by an Operational Support Grant for State University (BOPTN) from the Directorate General of Higher Education, Indonesian Ministry of Education (grant number 2058/UN4.20/PL.09/2013). PT Roche Indonesia donated the reagents used for HBsAg quantification.
Conflict of interest: No conflict of interest to declare.
Serum hepatitis B surface antigen and hepatitis B e antigen titers: disease phase influences correlation with viral load and intrahepatic hepatitis B virus markers.
00262-3&id=gr1.jpg){kind=link}
00262-3&id=gr2.jpg){kind=link}
00262-3&id=gr3.gif){kind=link}
00262-3&id=gr4.jpg){kind=link}