Improved Immune Status Corresponds with Long-Term Decline of Quantitative Serum Hepatitis B Surface Antigen in HBV/HIV Co-infected Patients

Abstract

In HBV/HIV–co-infected individuals, the course of hepatitis B is aggravated, leading to higher morbidity and mortality rates. Increasing evidence suggests an important role for hepatitis B surface antigen (HBsAg) quantification in monitoring treatment efficacy in HBV monoinfection. However, data concerning any HBsAg decline during treatment of HBV/HIV coinfection are limited. Fifty-one HBV/HIV–co-infected patients were retrospectively followed for a mean of 43 months (median 2 years, interquartile range 2 years). Baseline and on-treatment parameters were associated with longitudinal HBsAg levels. At baseline, serum HBsAg levels were comparable between patients on antiretroviral therapy (ART; n=43) and patients without (n=8). Longitudinally, HBsAg decreased in ART patients (−0.20±0.09 log₁₀ IU/mL/y), but slightly increased in subjects without therapy (0.22±0.26 log₁₀ IU/mL/y; p<0.001). In 58% of the ART subjects an HBsAg decline >10% was seen during the initial 24 mo. They showed higher baseline CD4 counts (401±42 versus 265±50 cells/μL, p=0.03), and had significantly higher CD4 counts at the last follow-up compared to patients without a decline (506±39 versus 310±51 cells/μL, p=0.01). A significant correlation was found between HBsAg decline from baseline to the last follow-up and the absolute increase of CD4 cells (r=0.44, p=0.003), as well the last CD4 count (r=0.41, p=0.006). This association was strongest in patients with complete suppression of HBV-DNA and HIV-RNA at the last follow-up visit. The highest HBsAg decline (−1.63±0.32 versus −0.43±0.24 log₁₀ IU/mL, p=0.001), and yearly HBsAg decline (−0.47±0.13 versus −0.19±0.12 log₁₀ IU/mL, p=0.03), were found in patients with CD4 increases >100 cells/μL at the last follow-up (n=21, 49%). Four cases of HBsAg loss were observed (8%). HBsAg declines steadily in >50% of HBV/HIV patients on ART. Long-term follow-up of HBV/HIV–co-infected patients is needed to identify distinct HBsAg patterns. Increasing CD4 counts indicating the restoration of immune competence in HBV/HIV–co-infected patients is associated with a stronger HBsAg decline.

Introduction

T he World Health Organization (WHO) estimates that one-third of the world population has serological evidence of prior or persistent hepatitis B virus (HBV) infection, and 350 million suffer from chronic hepatitis B virus infection (24). Due to the shared route of transmission, co-infection of HBV with human immunodeficiency virus (HIV) is common and can be found in an estimated 10% of the 40 million HIV-infected individuals (2,11). HIV co-infection in patients positive for hepatitis B surface antigen (HBsAg) negatively influences the course of hepatitis B. While HBV becomes chronic in <5% of adult HBV-monoinfected patients, HBV persists after infection in about 25% of HIV-infected adults, and is one of the principal causes of liver-related morbidity and mortality, leading to faster fibrosis progression, and the development of liver cirrhosis and hepatocellular carcinoma (10,13,22).

Current guidelines underline the beneficial effect of early immune reconstitution on the reduction of liver disease progression. According to the guidelines of the European AIDS Clinical Society (EACS), the initiation of highly-active antiretroviral therapy (HAART) to suppress HBV replication in HBV/HIV co-infection is given regardless of CD4 counts in patients who need anti-HBV therapy, and in HBsAg-positive patients with CD4 cell counts <500/mm³ (1,16,26). Quantitative serum HBsAg assessment in HBV monoinfection is emerging as a useful tool for the analysis of the natural course, as well as to evaluate treatment efficacy and patient response to antiviral treatment (4,8,17,19,23,25). The ultimate treatment goal for hepatitis B is HBsAg loss and the development of anti-HBs antibodies, indicating immunological control of the disease, and is associated with persistent HBsAg declines. The goal of the present study was to analyze factors influencing HBsAg levels in an HBV/HIV–co-infected group over time.

Patients and Methods

Patients and study design

The HBV/HIV–co-infected group consisted of 51 (17 HBeAg-positive, mean age 49 y) individuals enrolled over a period of 13 y (1998–2011) in two university hospitals. Forty-three patients received antiretroviral therapy (ART) with anti-HBV active components (tenofovir [TDF], emtricitabine [FTC], and/or lamivudine [3TC]). ART included TDF and FTC or 3TC in 36 patients; 7 individuals had 3TC alone. Besides nucleoside reverse transcriptase inhibitors (NRTIs), ART included non-nucleoside reverse transcriptase inhibitors (NNRTIs) and protease inhibitors (PIs). Information about the time since the initiation of antiviral therapy was not available. Eight patients did not receive antiviral treatment. The group was longitudinally studied for a mean of 43 mo (range 12–120 mo, median 2 years, interquartile range [IQR] 2 y; Table 1). HBsAg was quantified from stored samples if missing. The mean number of HBsAg determinations per patient was four, and the mean interval between follow-up visits was 10 mo. None had hepatocellular carcinoma, decompensated liver cirrhosis, or serological evidence of hepatitis delta virus (HDV) or hepatitis C virus (HCV) coinfection. The study was approved by the local ethics committee.

Table 1.

Characteristics of the HBsAg-positive HBV/HIV–Co-infected Cohort

	All (n=51)	HBsAg decrease ^a (n=25)	No HBsAg decrease (n=16)	p Value ^b
Age, years	49.0±1.40	49.4±1.88	48.3±3.09	0.74
HBsAg, log₁₀ IU/mL	3.57±0.17	3.49±0.20	3.87±0.17	0.34
Follow-up, months	43.3±3.84	44.1±5.70	43.5±6.30	0.76
HIV-RNA, copies/mL	2.55±0.18	2.67±0.27	2.25±0.27	0.58
Baseline CD4 count (cells/μL)	326±31	401±42	265±50	0.03
Baseline CD8 count (cells/μL)	1097±84	1130±106	1046±187	0.44
Last follow-up CD4 count (cells/μL)	411±32	506±39	310±51	0.01
Last follow-up CD8 count (cells/μL)	972±77	920±89	992±170	0.66
ART, n (%)	43 (84)	25 (100)	16 (100)	-
TDF, n (%)	36	22 (88)	12 (75)	0.28
AIDS, n (%)	19 (37)	9 (36)	6 (37)	0.92
HBeAg-positive, n (%)	17 (33)	8 (32)	6 (37)	0.90
HBV-DNA log₁₀ IU/mL	3.64±0.60	4.08±0.92	4.44±1.13	0.86
ALT U/mL	58±10	63±13	46±11	0.26

HBsAg decrease is defined as a >10% decline between the first and second year of follow-up in patients on ART (n=41, data are missing for early HBsAg decrease in 2 patients).

p Value for difference between patients with and without HBsAg decrease (the second versus the third column).

Baseline characteristics and comparisons between all patients and patients on ART with a >10% yearly decline versus no decline of serum HBsAg during the first 24 months of follow-up. All values mean±standard error of the mean unless otherwise stated.

HBsAg, hepatitis B surface antigen; HBV, hepatitis B virus; ALT, alanine aminotransferase; ART, antiretroviral therapy; AIDS, acquired immunodeficiency syndrome; TDF, tenofovir.

Serum HBsAg quantification

Serum HBsAg levels were quantified using the Abbott ARCHITECT^® assay (Abbott Diagnostics, Abbott Park, IL). The test has a range of 0.05–250 IU/mL. The samples were diluted 1:100 in horse serum, and if the results were >250 IU/mL, the sample was retested at a higher dilution. Samples with HBsAg levels <0.05 IU/mL were retested undiluted. The results are given in IU/mL. The inter- and intra-assay variability are approximately 10% (6.7–8.8%), according to the manufacturer's data sheet.

HBV-DNA, HIV-RNA, and T-cell measurements

HBV-DNA was quantified with the COBAS AmpliPrep/COBAS TaqMan test (Roche Diagnostics, Mannheim, Germany), with a lower limit of detection (LLoD) of 12 IU/mL. HIV-RNA was measured with the COBAS TaqMan HIV-1 test (Roche Diagnostics), with an LLoD of 50 copies/mL. CD4⁺, CD8⁺, and CD3⁺ T-lymphocyte counts (CD4 counts) were evaluated by flow cytometry (Navios; Beckman Coulter, Krefeld, Germany).

Statistical analysis

Serum HBsAg and HBV-DNA levels were logarithmically transformed. Data are presented as means (±SEM, standard error of the mean), unless otherwise indicated. Non-parametric (distribution-free) tests were applied. The Mann-Whitney U and Kruskal-Wallis analysis of variance (ANOVA) tests were used for univariate and multivariate comparisons of independent continuous variables, Fisher's exact test was used for discrete variable comparisons, and Wilcoxon's signed-rank test was used for paired variables. Multiple regressions with overall HBsAg decline as the dependent value, and CD4, HBV-DNA, and HIV-RNA levels were performed. Moreover, multivariate analysis was conducted by use of generalized linear model (GLM) analysis, including overall HBsAg decline as a dependent value, and the following predictors: qualitative - HBeAg-status, HBV-DNA, and HIV-RNA suppression during follow-up; quantitative - overall CD4 change, baseline HBV-DNA, and HIV-RNA levels. Statistical analyses were performed with Statistica 9.1 (Statsoft, Tulsa, OK) and GraphPad Prism 5.0 (GraphPad Software, La Jolla, CA).

Results

Baseline serum HBsAg levels

At baseline, serum HBsAg levels were comparable between patients with and without ART (3.60±0.18 versus 3.38±0.51 log₁₀ IU/mL; p=0.77; Fig. 1a). While HBeAg-positive subjects had higher baseline HBsAg levels compared to HBeAg-negative patients (4.00±0.22 versus 3.15±0.28 log₁₀ IU/mL; p=0.02), there was no association with Centers for Disease Control and Prevention (CDC) stage (CDC A: 3.64±0.38 log₁₀ IU/mL; CDC B: 3.50±0.26 log₁₀ IU/mL; CDC C: 3.70±0.30 log₁₀ IU/mL; p=0.69). Baseline serum HBsAg shows a positive correlation with serum HBV-DNA (r=0.64, p<0.001), but not with alanine aminotransferase (ALT; r=0.27, p=0.07), HIV-RNA (r=−0.01, p=0.97), or CD4 count (r=−0.17, p=0.24).

FIG. 1.

Serum HBsAg levels in HBV/HIV–co-infected patients. (a) Baseline serum HBsAg levels in various subgroups of the studied cohort (bars denote mean values; the differences were not statistically significant except for HBeAg⁺ versus HBeAg⁻; p=0.02). (b) Mean (±standard error of the mean) decline in HBsAg levels during the study in patients receiving ART with an overall increase in CD4 T-cell counts >100 cells/μL (●), versus patients with CD4 cell increases ≤100 cells/μL (□). Asterisks denote statistical significance between groups (*p<0.05 by the Mann-Whitney U test). Numbers refer to the patient sample size. (c) Comparison of HBsAg declines per year in patients with a CD4 count increase during ART >100 cells/μL, CD4 counts <100 cells/μL, and those with CD4 counts >350 cells/μL at the last follow-up. The difference in HBsAg decline between the groups was significant.

Longitudinal kinetics of serum HBsAg and immunological status

HBsAg increased slightly in patients without ART (median change 0.22±0.26 log₁₀ IU/mL/y), but declined in subjects receiving ART (median −0.20±0.09 log₁₀ IU/mL/y; p<0.001). A delayed HBsAg decline of >10%/y (i.e., between years 1 and 2 of follow-up), occurred in 25 patients receiving ART (58%), and in 16 untreated individuals it did not (37%), with data missing for two individuals (5%). Treated patients with delayed HBsAg decreases had higher baseline CD4 counts (401±42 versus 265±50 cells/μL; p=0.03), and had significantly higher CD4 counts at the last follow-up compared to treated patients without declines (506±38 versus 310±63 cells/μL; p=0.01; Table 1).

Consequently, the association between HBsAg and CD4 cell kinetics was studied in detail in 43 patients receiving ART. There was a significant correlation between HBsAg decline from baseline to last follow-up and the absolute increase in CD4 cell numbers (r=0.44, p=0.003), as well as with the last CD4 cell count (r=0.41, p=0.03). The relative HBsAg decline seen in ART-treated patients (mean −36.6%) also correlated with a relative CD4 cell increase (mean+66.7%, r=0.42, p=0.004). A significant association persisted with a yearly HBsAg decline and a CD4 cell increase (r=0.31, p=0.03). Higher overall HBsAg declines (−1.63±0.32 versus −0.43±0.24 log₁₀ IU/mL, p=0.001; Fig. 1b), and higher yearly HBsAg declines (−0.47±0.13 versus −0.19±0.12 log₁₀ IU/mL, p=0.03), were found in patients with a CD4 increase >100 cells/μL from baseline to last follow-up (n=21, 49%). Twenty-seven patients with a final CD4 count >350 cells/μL had more pronounced HBsAg declines (−1.41±0.26 versus −0.84±0.37 log₁₀ IU/mL, p=0.02; Fig. 1c).

Association between serum HBsAg, HIV-RNA, and HBV-DNA

Next the association between HIV-RNA levels, HBV-DNA concentrations, and HBsAg declines was analyzed. HIV-RNA was still detectable in 11/43 (26%) patients on ART at the last follow-up. There was a trend toward lower overall HBsAg declines in individuals with incomplete suppression of HIV (−0.63±0.32 versus −1.39±0.26 log₁₀ IU/mL; p=0.07). HBV-DNA at the last follow-up could be analyzed in 22 patients. It was detectable in 7/22 individuals on ART who had a significantly lower HBsAg decline during the study compared to patients with negative HBV-DNA (−0.73±0.45 versus −1.71±0.34 log₁₀ IU/mL; p=0.03). On the other hand, the total HBsAg decline was higher in patients with complete suppression of both HBV-DNA and HIV-RNA and >100 cells/μL increase in CD4 counts (−1.32±0.27 versus −0.88±0.31 log₁₀ IU/mL; p=0.01). Upon multiple regression analysis, the HBsAg decline showed an independent association with overall CD4 count changes during the study (β=0.50, p=0.03). In the constructed GLM, an HBsAg decline during the study was independently associated with an overall increase in the CD4 cell count (p=0.004), but not with baseline HBV-DNA (p=0.37), and HIV-RNA (p=0.10). After creating a similar model, but including only the subgroup of patients with complete suppression of HBV and HIV viral loads, the association between HBsAg and CD4 dynamics was less pronounced, but was near the level of significance (p=0.06).

HBsAg decline and HBsAg loss

In this long-term observational study, four (8%) cases of HBsAg-loss were observed, all in patients receiving ART including TDF and FTC. All four patients had a relatively strong HBsAg decline (>0.5 log₁₀ IU/mL) during the initial 24 mo. Three of them showed a solid CD4 increase (256, 212, and 148 cells/μL) during the study, while in one patient the baseline and last CD4 cell counts were high (798 cells/μL). A complete suppression of HBV and HIV viral load has been achieved in all.

Discussion

Quantitative HBsAg is increasingly recognized as an important biomarker for the classification of hepatitis B during the natural course (8,23), and for the analysis of treatment response (4,6,17 –19,25). In short, lower HBsAg titers seem to reflect later phases of chronic hepatitis B and less active disease. Also, patients showing rapid on-treatment HBsAg declines are more likely to achieve serological end-points like HBeAg seroconversion or HBsAg loss, and virological end-points including sustained viral suppression. While it could be shown that low CD4 counts are a risk factor for the persistence of HBV-DNA in HBV/HIV–co-infected subjects (3,12), data on quantitative HBsAg in special cohorts are scarce, and could help identify the parameters associated with HBsAg declines (14,20,21). Here the longitudinal course of HBsAg in 51 HBV/HIV co-infected patients followed for a mean of approximately 3.5 y is reported.

At baseline, higher HBsAg levels were found in HBeAg-positive patients than in HBeAg-negative patients, a finding in accord with previous observations of higher HBsAg levels during the early stages of HBV infection (23). The only other significant correlation was with HBV viral load, though there was no association with HIV/AIDS-related parameters, including the disease and treatment status or ALT in this small group. The missing correlation might argue against the influence of CD4 cells on HBsAg acutely. However, long-term observations show a correlation between improving CD4 counts and HBsAg declines in patients undergoing antiviral treatment, a dynamic process that might eventually lead to HBsAg loss. In an earlier study researchers also found that patients experiencing combined serological and virological responses, defined as HBe/sAg loss and HBV-DNA undetectability, had a more than fourfold median CD4 count increase during ART over patients that did not achieve an HBV response. In addition, elevated baseline ALT levels increased the likelihood of an HBV response, thereby reflecting an active immune system (15).

There was a trend toward lower overall HBsAg declines in individuals with incomplete suppression of HIV, but this was not statistically significant. ART-treated patients who remained HBV-DNA-positive at the last follow-up had a significantly lower HBsAg decline compared to patients with negative HBV-DNA (p=0.03). The HBsAg decline was most pronounced in patients with complete suppression of HBV-DNA and HIV-RNA in addition to >100 cells/μL increase in CD4 counts (−1.32±0.27 versus −0.88±0.31 log₁₀ IU/mL; p=0.01). In a recent article by Thibault et al., they noted a constant HBsAg decline in only 13 of 33 patients despite complete HBV suppression (21). When Iser et al. looked for intrahepatic changes in the immune cells in patients receiving ART, they found a reduction in intrahepatic HBsAg staining, but no significant changes in the hepatic number of CD4 and CD8 T cells, or T-cell activation markers (7). The lack of significant immunological changes in the livers of treated co-infected patients might support the observation of infrequent and small HBsAg decreases despite effective viral suppression, since pronounced HBsAg reductions require the eradication of infected liver cells, which is mediated by CD4 and CD8 T cells. Only patients with a strong virological and immunological response to ART may display such an HBsAg decline.

Still, a large group of patients experience only a minor or no HBsAg decrease at all despite effective viral suppression. This was seen in the present HBV/HIV–co-infected as well as in HBV-monoinfected patients receiving antiviral therapy (25). In the latter the failure to achieve any substantial HBsAg decrease when receiving antiviral therapy with pegylated interferon or telbivudine predicts a poor course regarding quantitative HBsAg, and with it the chance of consecutive HBsAg loss (19,25), potentially requiring prolonged antiviral treatment to continuously suppress viral replication.

A delayed HBsAg decrease >10% between the first and the second year of antiretroviral treatment occurred in 58% of the treated patients. This late slope was also found in HBV-monoinfected patients on antiviral treatment. In these patients, the late HBsAg decrease seen following effective HBV-DNA suppression on nucleoside/nucleotide analogues was correlated with a higher chance of HBsAg loss (9). In the group with ART, a correlation of the HBsAg decline and the CD4 count at baseline and at the last follow-up could be found. In addition, there was a significant correlation between the HBsAg decline, the absolute increase in CD4 cells, and the last CD4 count. If the difference between the baseline and the last CD4 count exceeded 100 cells/μL, higher overall and yearly HBsAg decreases became statistically apparent. Although this increase was not great, it does help to distinguish between patients with an improving immune system including factors like treatment adherence and those without. All four patients with HBsAg loss belonged to this group. A reduction and even elimination of HBsAg is thought to occur mainly by the immune-mediated clearance of HBV-infected cells, rather than by nucleoside-analogue-induced inhibition of the viral polymerase, supporting the importance of a competent immune system for viral control (5). The negative effect of low CD4 counts on HBV-related parameters could be shown previously in HBV/HIV co-infected patients, in whom it was associated with detectable HBV viremia (20). Patients with a final CD4 count >350 cells/μL also had a more pronounced HBsAg decline. This confirms the results of a recently published study by Maylin et al., who showed a significantly faster HBsAg (and HBeAg) decline in HBeAg-positive HBV/HIV–co-infected patients with a CD4 count of >350 cells/μL during treatment with tenofovir (14). However, the baseline and on-treatment changes in CD4 counts were not associated with improved serological outcomes of HBeAg and HBsAg loss in their analysis, but the CD4 count change was a significant confounder of HBsAg quantification, at least in the HBeAg-negative population. Therefore an improvement of the immunological status triggered by antiretroviral therapy might positively influence the decline of HBsAg, and thereby potentially improve the course of hepatitis B, at least in selected cohorts.

Regardless of our inherent limitations, including a limited number of subjects, and a lack of HBV-genotypes or information on ART compliance, the present study provides evidence of stronger HBsAg declines in patients who respond to ART with an increase in CD4 counts. These data may indicate an influence of the immune status on HBsAg levels, adding to recent observations by others.

In summary, the HBsAg decline seen in HBV/HIV–co-infected patients seems to be associated with an improving immunological status. Successful ART resulting in full suppression of HBV and HIV, but also an increase in CD4 cells, may be necessary to induce stronger HBsAg declines, and may eventually lead to complete HBsAg loss.

Footnotes

Acknowledgments

This work was supported by a grant from the Deutsche Zentrum für Infektionsforschung.

Author Disclosure Statement

Jerzy Jaroszewicz gave lectures for Abbott, Roche and Gilead. Jürgen Rockstroh received grants, consultancy, pay for lectures, and/or board membership from Merck, Abbott, Bionor, BMS, Boehringer Ingelheim, Gilead, GSK, MSD, Novartis, Pfizer, Roche, Tibotec, and ViiV. Ingmar Mederacke received grants and/or pay for lectures from Abbott Diagnostics and Roche Diagnostics. Michael P. Manns received grants, consultancy, pay for lectures, and/or board membership from Roche, BMS, Gilead, Novartis, GSK, and Merck. Heiner Wedemeyer received grants, consultancy, pay for lectures, and/or board membership from Gilead, BMS, Novartis, Roche, and Merck. Markus Cornberg received grants, consultancy, pay for lectures, and/or board membership from Gilead, BMS, Novartis, and Roche. Karsten Wursthorn received grants, consultancy, pay for lectures, and/or board membership for Roche Pharma, Roche Diagnostics, Novartis Pharma, Gilead, and BMS. The other authors have nothing to disclose.

References

European AIDS Clinical Society: T. E. A. C. S. EACS Guidelines Version 6. October 2011; 61.

Alter

. Epidemiology of viral hepatitis and HIV co-infection. J Hepatol, 2006; 44:S6–S9.

Bodsworth

, Cooper

, Donovan

. The influence of human immunodeficiency virus type 1 infection on the development of the hepatitis B virus carrier state. J Infect Dis, 1991; 163:1138–1140.

Brunetto

, Moriconi

, Bonino

et al. Hepatitis B virus surface antigen levels: a guide to sustained response to peginterferon alfa-2a in HBeAg-negative chronic hepatitis B. Hepatology, 2009; 49:1141–1150.

Grandvaux

, tenOever

, Servant

, Hiscott

. The interferon antiviral response: from viral invasion to evasion. Curr Opin Infect Dis, 2002; 15:259–267.

Heathcote

, Marcellin

, Buti

et al. Three-year efficacy and safety of tenofovir disoproxil fumarate treatment for chronic hepatitis B. Gastroenterology, 2010; 140:132–143.

Iser

, Avihingsanon

, Wisedopas

et al. Increased intrahepatic apoptosis but reduced immune activation in HIV-HBV co-infected patients with advanced immunosuppression. AIDS, 2011; 25:197–205.

Jaroszewicz

, Calle Serrano

, Wursthorn

et al. Hepatitis B surface antigen (HBsAg) levels in the natural history of hepatitis B virus (HBV)-infection: a European perspective. J Hepatol, 2010; 52:514–522.

Jaroszewicz

, Ho

, Markova

et al. Hepatitis B surface antigen (HBsAg) decrease and serum interferon-inducible protein-10 levels as predictive markers for HBsAg loss during treatment with nucleoside/nucleotide analogues. Antiviral Ther, 2011; 16:915–924.

10.

Kourtis

, Bulterys

, Hu

, Jamieson

. HIV-HBV coinfection—a global challenge. N Engl J Med, 2012; 366:1749–1752.

11.

Lacombe

, Bottero

, Lemoine

, Boyd

, Girard

. HIV/hepatitis B virus co-infection: current challenges and new strategies. J Antimicrob Chemother, 2009; 65:10–17.

12.

Landrum

, Fieberg

, Chun

et al. The effect of human immunodeficiency virus on hepatitis B virus serologic status in co-infected adults. PLoS One, 2010; 5:e8687.

13.

Mallet

, Dhalluin-Venier

, Verkarre

et al. Reversibility of cirrhosis in HIV/HBV coinfection. Antiviral Ther, 2007; 12:279–283.

14.

Maylin

, Boyd

, Lavocat

et al. Kinetics of hepatitis B surface and envelope antigen and prediction of treatment response to tenofovir in antiretroviral-experienced HIV-hepatitis B virus-infected patients. AIDS, 2012; 26:939–949.

15.

Miailhes

, Trabaud

, Pradat

et al. Impact of highly active antiretroviral therapy (HAART) on the natural history of hepatitis B virus (HBV) and HIV coinfection: relationship between prolonged efficacy of HAART and HBV surface and early antigen seroconversion. Clin Infect Dis, 2007; 45:624–632.

16.

Panel

o. A. G. f. A. a. A.

Guidelines for the use of antiretroviral agents in HIV-1-infected adults and adolescents. D. o. H. a. H. Services, 2011; 1–166.

17.

Reijnders

, Rijckborst

, Sonneveld

et al. Kinetics of hepatitis B surface antigen differ between treatment with peginterferon and entecavir. J Hepatol, 2010; 54:449–454.

18.

Rijckborst

, Hansen

, Cakaloglu

et al. Early on-treatment prediction of response to peginterferon alfa-2a for HBeAg-negative chronic hepatitis B using HBsAg and HBV DNA levels. Hepatology, 2010; 52:454–461.

19.

Sonneveld

, Rijckborst

, Boucher

, Hansen

, Janssen

. Prediction of sustained response to peginterferon alfa-2b for hepatitis B e antigen-positive chronic hepatitis B using on-treatment hepatitis B surface antigen decline. Hepatology, 2010; 52:1251–1257.

20.

Soriano

, Mocroft

, Peters

et al. Predictors of hepatitis B virus genotype and viraemia in HIV-infected patients with chronic hepatitis B in Europe. J Antimicrob Chemother, 2010; 65:548–555.

21.

Thibault

, Stitou

, Desire

, Valantin

, Tubiana

, Katlama

. Six-year follow-up of hepatitis B surface antigen concentrations in tenofovir disoproxil fumarate treated HIV-HBV-coinfected patients. Antiviral Ther, 2011; 16:199–205.

22.

Thio

. Hepatitis B and human immunodeficiency virus coinfection. Hepatology, 2009; 49:S138–S145.

23.

Thompson

, Nguyen

, Iser

et al. Serum hepatitis B surface antigen and hepatitis B e antigen titers: disease phase influences correlation with viral load and intrahepatic hepatitis B virus markers. Hepatology, 2010; 51:1933–1944.

24.

World Health Organization: Hepatitis B. 2011.

25.

Wursthorn

, Jung

, Riva

et al. Kinetics of hepatitis B surface antigen decline during 3 years of telbivudine treatment in hepatitis B e antigen-positive patients. Hepatology, 2010; 52:1611–1620.

26.

Wursthorn

, Wedemeyer

, Manns

. Managing HBV in patients with impaired immunity. Gut, 2010; 59:1430–1445.