Skip Navigation

Immunopathogenesis of Hepatic Flare in HIV/Hepatitis B Virus (HBV)–Coinfected Individuals after the Initiation of HBV-Active Antiretroviral Therapy

  1. Megan Crane1,
  2. Ben Oliver3,
  3. Gail Matthews5,
  4. Anchalee Avihingsanon6,
  5. Sasiwimol Ubolyam,
  6. Vesna Markovska1,
  7. J. Judy Chang1,a,
  8. Gregory J. Dore5,
  9. Patricia Price3,4,
  10. Kumar Visvanathan1,
  11. Martyn French3,4,
  12. Kiat Ruxrungtham6,b and
  13. Sharon R. Lewin1,2,b
  1. 1Department of Medicine, Monash University, and
  2. 2Infectious Diseases Unit, Alfred Hospital, Melbourne,
  3. 3School of Pathology and Laboratory Medicine, University of Western Australia, and
  4. 4Department of Clinical Immunology and Immunogenetics, Royal Perth Hospital and PathWest Laboratory Medicine, Perth, and
  5. 5National Centre in HIV Epidemiology and Clinical Research, University of New South Wales, Sydney, Australia;
  6. 6The HIV Netherlands Australia Thailand Research Collaboration, Thai Red Cross AIDS Research Centre, and Vaccine and Cellular Immunology Laboratory, Faculty of Medicine, Chulongkorn University, Bangkok, Thailand
  1. Reprints or correspondence: Prof. Sharon R. Lewin, Infectious Diseases Unit, Alfred Hospital, Burnet Institute, Level 2, Commercial Rd., Melbourne, Victoria 3004, Australia (s.lewin{at}alfred.org.au)
 
Next Section

Abstract

BackgroundThe pathogenesis of and risk factors for hepatic flare (HF) after the initiation of hepatitis B virus (HBV)–active antiretroviral therapy (ART) in HIV/HBV-coinfected individuals is not well understood

MethodsWe studied HF in ART-naive HIV/HBV-coinfected individuals in Thailand (n=36) who were beginning HBV-active ART as part of a prospective clinical trial. HF was defined as an alanine aminotransferase (ALT) level >5 times the upper limit of normal or >200 IU/L higher than that at baseline. Immune mediators (interleukin [IL]–18, IL-2, IL-6, IL-8, IL-10, soluble CD26 [sCD26], sCD30, sCD8, CXCL-10, CCL-2, tumor necrosis factor–α, interferon [IFN]–γ, and IFN-α) and activated NK cells were quantified

ResultsHBV DNA and ALT levels at baseline were higher in patients with HF (n=8) than in patients without HF (n=28) (P=.01). After the initiation of ART, CXCL-10 levels remained elevated in patients with HF but decreased in patients without HF (P<.01). sCD30 levels increased and were significantly higher at week 8 in patients with HF (P<.05). There was a positive correlation between levels of ALT and levels of CXCL-10, sCD30, CCL-2, and IL-18 at week 8 (the time of peak ALT level) but not at other time points

ConclusionElevated HBV DNA and ALT levels before the initiation of HBV-active ART are risk factors for HF. The pathogenesis of HF after the initiation of HBV-active ART is probably consistent with immune restoration disease

Coinfection with HIV and hepatitis B virus (HBV) is common, largely because of shared routes of transmission. Hepatotoxicity (grade 3 or 4 transaminitis) after commencement of antiretroviral therapy (ART) occurs more frequently in HIV/HBV-coinfected individuals than in those with HIV monoinfection [1–3]. After exclusion of new hepatotoxic medications, opportunistic infections, and/or alcohol excess, an increase in alanine aminotransferase (ALT) levels or hepatic flare (HF) is usually attributed to immune restoration disease (IRD). However, the pathogenesis of HF after ART in HIV/HBV-coinfected individuals is unclear

In HBV monoinfection, spontaneous HF is attributed to immune-mediated damage of HBV-infected hepatocytes and is often followed by seroconversion to hepatitis B e antigen [4]. HF in this setting may be mediated by proinflammatory and antiviral cytokines, such as tumor necrosis factor (TNF)–α and interferon (IFN)–γ [5]. IFN-γ can induce the production of several chemokines, including CXCL-10 [6]. These can mediate recruitment of monocytes, NK cells, and T cells [7, 8]. A recent study of spontaneous HF in HBV monoinfection demonstrated an association between HF and production of IFN-α and interleukin (IL)–8, leading to the recruitment of activated TNF-related apoptosis-inducing ligand (TRAIL)–expressing NK cells to the liver and subsequent hepatocyte damage [9]. In other studies, both HBV-specific and nonspecific CD8+ T cells have been shown to be present in the liver during spontaneous HF in HBV monoinfection [8, 10]. However, hepatocyte damage is thought to be mainly mediated by non–HBV-specific cytolytic CD8+ T cells [7, 8, 10]. A previous study of HIV/HBV-coinfected individuals beginning ART with or without anti-HBV activity demonstrated some recovery of circulating HBV-specific CD4+ and CD8+ T cells after ART, but HF was not observed in this small (n=5) longitudinal cohort [11]. It is currently unclear whether the immunopathogenesis of HF after the initiation of ART in HIV/HBV-coinfected individuals is analogous to spontaneous HF in HBV monoinfection

After the initiation of ART in individuals with low CD4+ T cell counts (<100 cells/μL), 10%–30% present with an apparently new opportunistic infection or worsening of an established infection. These infections constitute IRD, reflecting poorly regulated immune responses in the presence of high pathogen load [12, 13]. Pathogenic mechanisms of IRD differ with the causal pathogen. Mycobacterium tuberculosis–related IRD is driven by antigen-specific IFN-γ production and is also associated with increased production of IL-2, IL-6, IL-10, IL-12, IFN-γ, CXCL-10, TNF-α, CCL5, and CCL-2 [14]. In contrast, increases in levels of soluble CD30 (sCD30), cytomegalovirus (CMV)–specific antibodies, and IL-6 were observed during CMV-related IRD [15, 16], whereas baseline CD4+ T cell numbers were low [17]. CD8+ T cells have been implicated in the immunopathogenesis of JC virus–related IRD [18, 19], and both CD8+ T cells and NK cells have been implicated in the immunopathogenesis of varicella-zoster virus–related IRD [12, 20]. In hepatitis C virus (HCV)–related IRD, levels of anti-HCV antibody and sCD26 (dipeptidyl peptidase IV) activity increased, but levels of sCD30 remained low [16]

HF after the commencement of ART in patients with HIV/HBV coinfection may be a consequence of IRD in the liver [21]. In the present study of HF in HIV/HBV-coinfected individuals who began HBV-active ART as part of a randomized prospective study, we show that HF was common and was associated with increased plasma levels of the immune mediators CXCL-10, sCD30, IL-18, and CCL-2 and that this finding is probably consistent with HBV-related IRD

Previous SectionNext Section

Methods

SubjectsSubjects were recruited from the Tenofovir in Coinfection (TICO) trial [22], a prospective randomized (1:1:1) clinical trial of tenofovir versus lamivudine versus tenofovir-lamivudine within an efavirenz-based HBV-active ART regimen in ART-naive HIV/HBV-coinfected individuals in Thailand (n=36). There was no significant difference between the 3 treatment arms in the median decline in HBV DNA levels during 48 weeks of treatment by time-weighted area-under-the-curve analysis [22]. The number of patients with undetectable HBV DNA levels (<170 copies/mL) at week 12 was 1 (8%) of 13 for the lamivudine, 3 (25%) of 12 for the tenofovir, and 1 (10%) of 10 for the lamivudine-tenofovir treatment arm (P=.49). All subjects provided written informed consent before enrollment

Study designIn the present study, HF was defined as an ALT level >5 times the upper limit of normal or >200 IU/L higher than that at baseline occurring within the first 12 weeks of treatment (n=8). Patients with HF were compared with patients who were enrolled in the TICO trial but did not experience HF (n=28). Plasma and peripheral blood mononuclear cells (PBMCs) were collected at weeks 0, 4, 8, and 12 and stored at −80°C and −140°C, respectively. Details on sex, age, HIV RNA levels, CD4 cell counts, concomitant medications, new opportunistic infections, alcohol use, and medication adherence were also collected and have been reported elsewhere [22, 23]

HBV DNA levelsHBV DNA levels were measured at weeks 0, 4, 8, and 12 at a central laboratory, the Victorian Infectious Diseases Laboratory (Melbourne). HBV DNA levels were measured in all samples by means of the Versant HBV DNA bDNA Assay (version 3.0; Bayer HealthCare). The linear dynamic range of the assay was from 2×103 to 1×108 copies/mL or from 3.6×102 to 1.8×107 IU/mL. For samples below the lower limit of detection (LLOD) for the bDNA assay (2×103 copies/mL), testing was repeated using the COBAS TaqMan HBV Test (Roche Diagnostics). The LLOD for this assay was ∼1.7×102 copies/mL or 3×101 IU/mL. HBV DNA levels are reported as copies per milliliter

Quantification of immune mediators in plasmaIFN-γ, IL-10, IL-2, IL-6, IL-8, CCL-2, TNF-α, and CXCL-10 levels in plasma were quantified using BD Cytometric Bead Array Flex Sets (BD Biosciences). More than 300 events were acquired for each sample, using a FACSCanto II cytometer (BD Immunocytometry Systems). Analysis was performed using FCAP Array software (version 1.0; BD Biosciences). All samples were tested undiluted. The LLOD for these assays was 20 pg/mL

Plasma was also tested for IL-18 and IFN-α by ELISA (BD Biosciences), in accordance with the manufacturer’s instructions. Absorbance was read at 450 nm, using a microplate reader (model 550; Bio-Rad). Microplate Manager software (version 5.2; Bio-Rad) was used for analysis. The LLODs for the IL-18 and IFN-α assays were 6.0 and 7.8 pg/mL, respectively

sCD30 and sCD8 were assayed by ELISA. Half-volume (50-mL) 96-well plates were coated with anti-sCD30 or anti-sCD8 antibody (Bender MedSystems) in PBS, blocked with PBS/1% bovine serum albumin (BSA; 1 h), and washed with PBS/0.05% Tween 20. Samples, standards, and anti-sCD30 or anti-sCD8 peroxidase conjugate were prediluted in PBS/BSA/Tween 20 and added to the plate together (3 h, on a plate shaker). The plates were washed, bound sCD30 and sCD8 were detected with tetramethylbenzidine substrate, the reaction was stopped with 0.5 mol/L H2SO4, and absorbance was read at 450 nm. The LLODs for the sCD30 and sCD8 assays were 15.6 and 9.6 U/mL, respectively

sCD26 (dipeptidyl peptidase IV) enzyme activity was quantified using an antigen capture enzyme assay [22]. Plates were coated with anti-sCD26 antibody (PharMingen) in bicarbonate buffer and blocked with PBS/1% BSA (1 h). Prediluted samples and a standard control pool were added (4 h). After a wash in PBS/0.05% Tween 20, bound enzyme activity was detected with chromogenic substrate Gly-Pro-pNA (Sigma) for 5 h at 37°C. Absorbance was read at 405 nm. The LLOD for this assay was 7.8 U/mL

Flow cytometry analysis of NK cellsCryopreserved PBMCs were thawed and stained with 0.6 μg/mL CD3–peridinin–chlorophyll protein, 0.12 μg/mL CD56 (neural cell adhesion molecule 16.2)–fluorescein isothiocyanate, and 10 μg/mL biotinylated TRAIL or biotinylated TRAIL isotype control antibodies (BD Biosciences) for 30 min at 4°C followed by 4 μg/mL strepavidin-allophycocyanin conjugate and 5 μg/mL propidium iodide (Invitrogen) for another 20 min at 4°C and then resuspended in 100 μL of FacsFix solution (PBS and 10% formaldehyde; Polysciences) for analysis within 24 h. Positive TRAIL-expressing NK cell controls were prepared by isolating PBMCs from a single uninfected blood bank donor over a Ficoll-Paque centrifugation gradient and activated in culture for 3 days (37°C in 5% CO2) at 5×106 cells/mL in RPMI 1640 (Invitrogen)/10% cosmic calf serum with 100 U/mL human recombinant IL-2. Cells were then harvested and cryopreserved for use in each assay. Cryopreserved PBMCs from HIV/HBV-uninfected control subjects (n=6) were also analyzed with every assay. Cell data were acquired on a 4-color FACSCalibur flow cytometer (BD Immunocytometry Systems) and analyzed using WEASEL (Walter and Eliza Hall Eclectic and Lucid) software (version 2.3.1)

Statistical analysisData were log-transformed and assessed for normality. Levels of immune mediators were compared between patients with and those without HF over time using a nonparametric test (Kruskal-Wallis test with Dunn’s posttest) for nonnormally distributed data or a parametric test (repeated-measures analysis of variance [ANOVA] with Bonferoni adjustments) for normally distributed data. Significant differences were confirmed using the Mann-Whitney U test (nonparametric) or Student’s t test (parametric) with Bonferroni adjustments. Differences between patients with and those without HF were assessed using the Mann-Whitney U test. Nominal values were compared using Fisher’s exact test if the sample contained a subpopulation <4 or using the χ2 test if all subpopulations were ⩾4. The linear relationship of a given parameter to ALT level at each time point was assessed for significance (Spearman’s correlation) in a univariate analysis. Spearman’s&amp;rank correlation coefficients were also calculated for each pairwise comparison of changes over time in HIV RNA level, HBV DNA level, or CD4+ T cell count with changes over time in selected cytokine levels between weeks 0 and 12. Data were analyzed using STATA (version 9; StataCorp) and GraphPad Prism software (version 4; GraphPad Software). Statistical significance was set at P⩽.05

Previous SectionNext Section

Results

Clinical characteristics of case patients with HF and control patients without HFThe clinical details at enrollment of patients with and those without HF are summarized in table 1. Before the initiation of ART, patients with HF (n=8) had significantly higher levels of HBV DNA and ALT than did the patients without HF (n=28) (P=.01 for both comparisons; Mann-Whitney U test) (table 1). There were no significant differences in CD4+ T cell count, HIV RNA level, treatment arm, sex, or age at enrollment between the patients with and those without HF (table 1). The median decrease in HBV DNA level within 12 weeks was significantly higher in the patients with HF than in those without HF (P=.02), but there were no significant differences in the increase in CD4+ T cell count or decline in HIV RNA level

Figure 1
View larger version:
Figure 1

Changes in levels of immune mediators in plasma. A Representative examples of the changes in hepatitis B virus (HBV) DNA (white circles) and alanine aminotransferase (ALT) (black circles) levels (top) and in levels of key immune mediators (bottom) in a patient (1314) without hepatic flare (HF) (left) and in a patient (1341) with HF (right). B Concentrations of CXCL-10, soluble CD30 (sCD30), sCD26, interleukin (IL)–18, sCD8, and CCL-2 before and after the initiation of HBV-active antiretroviral therapy (ART) in patients without HF (left) and in those with HF (right). **P=.04 for the comparison between patients with and those without HF. Individual data points are shown in gray, the bottom and top edges of the boxes represent the 25th and 75th percentiles, the horizontal lines within the boxes represent medians, and the whiskers represent outliers

Figure 2
View larger version:
Figure 2

Relationship between levels of alanine aminotransferase (ALT) and immune mediators at the time of hepatic flare (HF). At week 8 after the initiation of hepatitis B virus–active antiretroviral therapy, significant positive correlations were seen between levels of ALT and CXCL-10, soluble CD30 (sCD30), CCL-2, and interleukin (IL)–18, both in patients without HF (black circles) and in patients with HF (white circles). The significance of the correlation was determined by Spearman’s test

Figure 3
View larger version:
Figure 3

Expression of activated tumor necrosis factor–related apoptosis-inducing ligand (TRAIL) on circulating NK cells. CD3− lymphocytes (gated on forward scatter [FSC] and side scatter [SSC] profiles) were analyzed for CD56 and TRAIL expression by flow cytometry to determine the percentage of activated NK cells (A). TRAIL-expressing CD3−CD56bright cells were measured as a percentage of total CD3−CD56bright cells in cryopreserved peripheral blood mononuclear cells (PBMCs) (B). In panels B–D, representative flow cytometry plots of staining with allophycocyanin (APC)–conjugated isotype control antibody (left) or APC-conjugated TRAIL antibody (right) are shown for unstimulated PBMCs (B) interleukin (IL)–2 stimulated PBMCs from an HIV-negative donor (C) and PBMCs from a control patient (patient 1308) (D). Panel E shows the percentage of TRAIL-expressing CD3−CD56bright cells (i.e., the percentage of activated NK cells) among PBMCs from HIV-negative donors (left) that were unstimulated or stimulated with IL-2 and used as negative or positive controls, respectively. The percentage of TRAIL-expressing CD3−CD56bright cells before and after the initiation of hepatitis B virus–active antiretroviral therapy is shown for patients without hepatic flare (HF) (middle) and for patients with HF (right). Individual data points are shown in gray, the bottom and top edges of the boxes represent the 25th and 75th percentiles, the horizontal lines within the boxes represent medians, and the whiskers represent outliers

Table 1
View larger version:
Table 1

Comparison of clinical parameters between patients with and those without hepatic flare (HF) before the initiation of hepatitis B virus (HBV)–active antiretroviral therapy (ART)

The median ALT levels at weeks 0, 4, 8, and 12 were 79 (interquartile range, 59–96), 73 (50–150), 328 (165–676), and 178 (60–398) IU/mL, respectively, in the patients with HF and 36 (22–59), 35 (25–57), 44 (26–70), and 36 (27–67) IU/mL in the patients without HF. ALT levels peaked at week 8 and were significantly different between patients with and those without HF during the entire follow-up period (week 0, P=.014; weeks 4, 8, and 12, P<.01; ANOVA). HBV DNA levels were significantly higher in the patients with HF than in the patients without HF at weeks 0 and 4 (P=.01 and P<.01, respectively) but not at weeks 8 or 12 (P=.07 and P=.77, respectively)

Immunological differences between patients with and those without HFTo determine the pathogenesis of HF in this setting, plasma levels of immune mediators were quantified in patients with or without HF at weeks 0, 4, 8, and 12 after the initiation of ART. A significant decrease in CXCL-10 level over time was observed in patients without HF (for week 0 vs. 8 and week 0 vs. 12, P<.01; Kruskal-Wallis test), but this was not observed in patients with HF (P>.05) (figure 1)

Levels of sCD30 decreased significantly over time in patients without HF (for week 0 vs. 12, P<.01; ANOVA), but they increased in patients with HF (for week 0 vs. 8, P<.01; ANOVA). At week 8, concentrations of sCD30 were higher in patients with HF than in patients without HF (P=.04; ANOVA and Student’s t test with Bonferroni adjustment) (figure 1). The peak in sCD30 levels in patients with HF at week 8 mirrored the change in ALT levels (data not shown)

Levels of sCD26 significantly increased in both patients with and those without HF over time (for week 0 vs. 8 and week 0 vs. 12 in patients with HF, P<.01; for week 0 vs. 8 and week 0 vs. 12 in patients without HF, P=.033 and P<.01, respectively), but no differences were observed between patients with and those without HF at any time point (P>.05; Student’s t test with Bonferroni adjustment). No differences were observed over time or between patients with and those without HF for IL-18, CCL-2, or sCD8 (figure 1). IL-2, IL-6, IL-10, IL-8, IFN-α, IFN-γ, and TNF-α were not detected in the majority of patients (data not shown)

Correlations between levels of ALT, HBV DNA, and immune mediatorsTo investigate the relationship between virological and biochemical parameters and plasma immune mediators at each time point, we assessed the correlation between values for ALT, HBV DNA, HIV RNA, and CD4+ T cells and the concentrations of immune mediators in univariate analyses. Significant positive correlations were found between levels of ALT and CXCL-10 (r=0.46; P=.005), sCD30 (r=0.46; P=.006), CCL-2 (r=0.36; P=.032), and IL-18 (r=0.44; P=.009) at week 8, the time of peak ALT levels (Spearman’s test) (figure 2). No other significant correlations were found between levels of immune mediators and ALT, HBV DNA, HIV RNA, or CD4+ T cells at any other time point (data not shown). In addition, there was no correlation between changes in CD4+ T cell count, HIV RNA levels, or HBV DNA levels and changes in levels of any of the immune mediators (data not shown)

Decline in the level of circulating activated NK cells in both patients with and those without HP receiving ARTBecause activated NK cells are implicated in HF in monoinfected patients [9], we next examined the surface expression of TRAIL on circulating CD3−CD56+ NK cells [9]. Cryopreserved PBMCs were available from a subset of patients with and those without HF (for patients with HF, n=5 at week 0, n=3 at week 4, and n=5 at week 12; for patients without HF, n=14 at each time point). There were no differences between patients with and those without HF in the number of TRAIL-expressing CD3−CD56bright cells as a percentage of total CD3−CD56bright cells, but levels in both groups significantly decreased after the initiation of ART (for week 0 vs. 12 in patients with HF, P<.01; for week 0 vs. 4 and week 0 vs. 12 in patients without HF, P<.001; Kruskal-Wallis test and Mann-Whitney U test with Bonferroni adjustment) (figure 3). There were also no differences between patients with and those without HF in the total number of CD3−CD56+ NK cells or the total number of cells in either the CD56bright or CD56dim NK cell populations (data not shown). We found no significant correlation between the percentage of TRAIL-expressing CD3−CD56bright cells and ALT, HBV DNA, or HIV RNA levels or CD4+ T cell counts (data not shown)

Previous SectionNext Section

Discussion

This is the first prospective study of HF after the initiation of HBV-active ART in HIV/HBV-coinfected patients. The frequency of HF was high and was associated with high levels of HBV DNA, ALT, CXCL-10, and sCD30 in plasma. ALT levels were correlated with CXCL-10, sCD30, CCL-2, and IL-18 levels at week 8, the time of peak ALT levels. Therefore, HF after the initiation of HBV-active ART in HIV/HBV-coinfected individuals was associated with an increased antigen burden (elevated HBV DNA level) before HF and with elevated levels of IFN-γ–inducible cytokines (CXCL-10 and IL-18), as described for IRD caused by other pathogens. These data suggest that HF in this setting is probably secondary to HBV-related IRD in the liver

CXCL-10 levels remained elevated in patients with HF but decreased over time during ART in patients without HF. We were unable to address why CXCL-10 levels decreased in some patients and not others after ART, but the persistent elevation of CXCL-10 levels was clearly associated with HF in this study. CXCL-10 is induced by IFN-γ [6] and can skew the immune response toward the production of Th1 cytokines [24, 25]. In addition, CXCL-10 is a chemoattractant for T cells, NK cells, and monocytes to the liver, and persistent elevation of CXCL-10 levels in the setting of a rising CD4+ T cell count may therefore drive enhanced T cell recruitment to the liver, leading to hepatocyte damage [26]. In HIV/HCV coinfection, CXCL-10 has been positively correlated with liver fibrosis scores and liver enzyme levels [27]

IL-18, a potent antiviral cytokine induced in response to IFN-γ, was also correlated with levels of ALT in the present study. IL-18 may cause nonspecific damage to infected hepatocytes in HBV monoinfection [28]. The association of both CXCL-10 and IL-18 with HF suggests that IFN-γ may mediate HBV-related IRD, as has been demonstrated for M. tuberculosis– and cryptococcus-related IRD [14, 29]

After the initiation of ART, there is limited recovery of HBV-specific T cells in HIV/HBV-coinfected individuals [11, 30]. We were unable to measure HBV-specific T cells in the present study directly, but we examined 2 plasma markers of T cell activation, sCD30 and sCD26. Levels of sCD30 increased over time in patients with HF but decreased in patients without HF. CD30 belongs to the family of TNF receptors, and sCD30 is shed from activated T and B cells [31]. Circulating levels of sCD30 are elevated in autoimmune diseases and HIV infection [15, 31] and in CMV-related IRD [15, 32]. We found no difference in levels of sCD26 between patients with and those without HF. sCD26 is a T cell activation marker, and levels are transiently elevated during HCV-related IRD [16]. Because of the limited availability of PBMCs, we were unable to directly measure cellular markers of T cell activation, including expression of CD69, HLA-DR, or CD38, which have previously been shown to be elevated in M. tuberculosis– and cryptococcus-related IRD [14, 29]. Therefore, we were unable to determine whether changes in sCD30 and sCD26 levels in this setting were a consequence of T cell activation, B cell activation, or both

In contrast to spontaneous HF in HBV-monoinfected individuals, we found no association between circulating activated NK cells and HF [9]. We did, however, demonstrate an increase in CXCL-10 levels in patients with HF compared with those in patients without HF and a positive correlation between CCL-2 and ALT levels at the time of HF. Given that both of these chemokines are monocyte and T cell chemoattractants [26, 33], it is possible that liver damage is mediated by cells other than activated NK cells in HBV-related IRD. Recruitment of cells to tissue usually requires a gradient of chemokine concentration, as demonstrated for CXCR3 ligands and HCV infection [34–36]. We were unable to measure the concentrations of chemokines or cytokines in liver, but these will be important to assess in future studies

There were limitations in the present study. First, although this was the first prospective study of the initiation of HBV-active ART in treatment-naive HIV/HBV-coinfected individuals with a high rate of HF, the overall sample size was small, and PBMCs for evaluating activated NK cells were available in only a subset of patients. Second, the TICO trial randomized patients to 3 different HBV-active ART regimens. However, significant and similar declines in HBV DNA and HIV RNA levels were demonstrated in all treatment arms within the first 12 weeks [22]. Although the number of patients with HF was higher in the lamivudine-only arm, there was no statistically significant difference in the numbers of patients with HF between treatment arms. However, the overall number of patients with HF in this study was small, and the study was not sufficiently powered to identify a difference in the incidence of HF among the different treatment arms. Finally, we were unable to monitor HBV-specific T cells in this study. Studies of HBV monoinfection suggest that both HBV-specific and nonspecific T cells infiltrate the liver during HF, with liver damage being largely mediated by non–HBV-specific CD8+ T cells [7, 8, 10] and proinflammatory cytokines [9]. Future studies of HF after the initiation of HBV-active ART should examine circulating and intrahepatic HBV-specific T cell responses by means of an overlapping peptide library and multiparameter cytokine staining [37]

In summary, HF after the initiation of ART in HIV/HBV-coinfected individuals was associated with elevated plasma levels of HBV DNA, ALT, and immune mediators (consistent with an elevated Th1 immune response); enhanced lymphocyte activation; and elevated levels of chemokines associated with the recruitment of both monocytes and T cells to the liver. We found no evidence for an NK cell–mediated mechanism, as implicated in HF in HBV monoinfection. These data suggest that HF in this setting is probably different from spontaneous flare in HBV monoinfection and is a result of HBV-related IRD

Previous SectionNext Section

Acknowledgments

We thank Tim Spelman for his expert statistical advice, Pip Marks for provision of clinical data, Drs. Laveena Sharma and Miranda Xhilaga for assistance with the figures, and the patients for their participation in this study

Previous SectionNext Section

Footnotes

  • ↵Potential conflicts of interest: K.R. has recently received research grants/funding, lecture sponsorships, or honoraria from or has been a consultant or advisor to Abbott Laboratories, Boehringer Ingelheim, Bristol-Myers Squibb, Gilead Sciences, GlaxoSmithKline, Hoffmann–La Roche, Janssen-Cilag, Merck Sharp & Dohme, Tibotec, and Virco. S.R.L. has received research grants/funding from Roche, Bristol-Myers Squibb, Gilead Sciences, and Pfizer. All other authors report no potential conflicts

    Presented in part: 15th Conference on Retroviruses and Opportunistic Infections, Boston, 3–6 February 2008 (abstract 1033)

    Financial support: National Institutes of Health (grant R21 AI055379-01 A1 to J.J.C., G.J.D., and S.R.L.); Australian Postgraduate Awards (to J.J.C.); Gilead Sciences; Alfred Foundation (support to S.R.L.); National Health and Medicine Research Council (Practitioner Fellowship 251651 to S.R.L.); National Health and Medical Research Council of Australia (support to B.O., P.P., and M.F.)

  • ↵Present affiliation: Partners AIDS Research Center, Massachusetts General Hospital, Charleston

  • ↵K.R. and S.R.L. are joint senior authors

  • Received July 16, 2008.
  • Accepted October 23, 2008.
Previous Section
 

References

  1. ↵
    1. Sulkowski MS,
    2. Thomas DL,
    3. Chaisson RE,
    4. Moore RD
    . Hepatotoxicity associated with antiretroviral therapy in adults infected with human immunodeficiency virus and the role of hepatitis C or B virus infection. JAMA 2000;283:74-80.
    Abstract/FREE Full Text
  2. ↵
    1. Martinez E,
    2. Blanco JL,
    3. Arnaiz JA,
    4. et al
    . Hepatotoxicity in HIV-1-infected patients receiving nevirapine-containing antiretroviral therapy. AIDS 2001;15:1261-8.
    CrossRefMedlineWeb of Science
  3. ↵
    1. den Brinker M,
    2. Wit FW,
    3. Wertheim-van Dillen PM,
    4. et al
    . Hepatitis B and C virus co-infection and the risk for hepatotoxicity of highly active antiretroviral therapy in HIV-1 infection. AIDS 2000;14:2895-902.
    CrossRefMedlineWeb of Science
  4. ↵
    1. Milich D,
    2. Liang TJ
    . Exploring the biological basis of hepatitis B e antigen in hepatitis B virus infection. Hepatology 2003;38:1075-86.
    CrossRefMedlineWeb of Science
  5. ↵
    1. Guidotti LG,
    2. Ishikawa T,
    3. Hobbs MV,
    4. Matzke B,
    5. Schreiber R
    . Intracellular inactivation of the hepatitis B virus by cytotoxic T lymphocytes. Immunity 1996;4:25-36.
    CrossRefMedlineWeb of Science
  6. ↵
    1. Bonecchi R,
    2. Bianchi G,
    3. Bordignon PP,
    4. et al
    . Differential expression of chemokine receptors and chemotactic responsiveness of type 1 T helper cells (Th1s) and Th2s. J Exp Med 1998;187:129-34.
    Abstract/FREE Full Text
  7. ↵
    1. Sprengers D,
    2. van der Molen RG,
    3. Kusters JG,
    4. et al
    . Different composition of intrahepatic lymphocytes in the immune-tolerance and immune-clearance phase of chronic hepatitis B. J Med Virol 2006;78:561-8.
    CrossRefMedlineWeb of Science
  8. ↵
    1. Maini MK,
    2. Boni C,
    3. Lee CK,
    4. et al
    . The role of virus-specific CD8+ cells in liver damage and viral control during persistent hepatitis B virus infection. J Exp Med 2000;191:1269-80.
    Abstract/FREE Full Text
  9. ↵
    1. Dunn C,
    2. Brunetto M,
    3. Reynolds G,
    4. et al
    . Cytokines induced during chronic hepatitis B virus infection promote a pathway for NK cell-mediated liver damage. J Exp Med 2007;204:667-80.
    Abstract/FREE Full Text
  10. ↵
    1. Tang TJ,
    2. Janssen HL,
    3. Kusters JG,
    4. de Man RA,
    5. Schalm SW
    . The intrahepatic immune response during chronic hepatitis B infection can be monitored by the fine-needle aspiration biopsy technique. FEMS Immunol Med Microbiol 2003;39:69-72.
    CrossRefMedlineWeb of Science
  11. ↵
    1. Lascar RM,
    2. Gilson RJ,
    3. Lopes AR,
    4. Bertoletti A,
    5. Maini MK
    . Reconstitution of hepatitis B virus (HBV)-specific T cell responses with treatment of human immunodeficiency virus/HBV coinfection. J Infect Dis 2003;188:1815-9.
    Abstract/FREE Full Text
  12. ↵
    1. Clark BM,
    2. Krueger RG,
    3. Price P,
    4. French MA
    . Compartmentalization of the immune response in varicella zoster virus immune restoration disease causing transverse myelitis. AIDS 2004;18:1218-21.
    CrossRefMedlineWeb of Science
  13. ↵
    1. French MA
    . Disorders of immune reconstitution in patients with HIV infection responding to antiretroviral therapy. Curr HIV/AIDS Rep 2007;4:16-21.
    CrossRefMedline
  14. ↵
    1. Bourgarit A,
    2. Carcelain G,
    3. Martinez V,
    4. et al
    . Explosion of tuberculin-specific Th1-responses induces immune restoration syndrome in tuberculosis and HIV co-infected patients. AIDS 2006;20:F1-7.
    MedlineWeb of Science
  15. ↵
    1. Keane NM,
    2. Price P,
    3. Lee S,
    4. Stone SF,
    5. French MA
    . An evaluation of serum soluble CD30 levels and serum CD26 (DPPIV) enzyme activity as markers of type 2 and type 1 cytokines in HIV patients receiving highly active antiretroviral therapy. Clin Exp Immunol 2001;126:111-6.
    CrossRefMedlineWeb of Science
  16. ↵
    1. Stone SF,
    2. Lee S,
    3. Keane NM,
    4. Price P,
    5. French MA
    . Association of increased hepatitis C virus (HCV)–specific IgG and soluble CD26 dipeptidyl peptidase IV enzyme activity with hepatotoxicity after highly active antiretroviral therapy in human immunodeficiency virus–HCV-coinfected patients. J Infect Dis 2002;186:1498-502.
    Abstract/FREE Full Text
  17. ↵
    1. Hsieh SM,
    2. Pan SC,
    3. Hung CC,
    4. Tsai HC,
    5. Chen MY
    . Association between cytomegalovirus-specific reactivity of T cell subsets and development of cytomegalovirus retinitis in patients with acquired immunodeficiency syndrome. J Infect Dis 2001;184:1386-91.
    Abstract/FREE Full Text
  18. ↵
    1. Tantisiriwat W,
    2. Tebas P,
    3. Clifford DB,
    4. Powderly WG,
    5. Fichtenbaum CJ
    . Progressive multifocal leukoencephalopathy in patients with AIDS receiving highly active antiretroviral therapy. Clin Infect Dis 1999;28:1152-4.
    Abstract/FREE Full Text
  19. ↵
    1. Kotecha N,
    2. George MJ,
    3. Smith TW,
    4. Corvi F,
    5. Litofsky NS
    . Enhancing progressive multifocal leukoencephalopathy: an indicator of improved immune status. Am J Med 1998;105:541-3.
    CrossRefMedlineWeb of Science
  20. ↵
    1. Domingo P,
    2. Torres OH,
    3. Ris J,
    4. Vazquez G
    . Herpes zoster as an immune reconstitution disease after initiation of combination antiretroviral therapy in patients with human immunodeficiency virus type-1 infection. Am J Med 2001;110:605-9.
    CrossRefMedlineWeb of Science
  21. ↵
    1. Crane M,
    2. Matthews G,
    3. Lewin SR
    . Hepatitis virus immune restoration disease of the liver. Curr Opin HIV AIDS 2008;3:446-52.
    CrossRefMedline
  22. ↵
    1. Matthews GV,
    2. Avihingsanon A,
    3. Lewin SR,
    4. et al
    . A randomised trial of combination hepatitis B therapy in HIV/HBV coinfected antiretroviral naïve individuals in Thailand. Hepatology 2008;48:1062-9.
    CrossRefMedlineWeb of Science
  23. ↵
    1. Avihingsanon A,
    2. Matthews G,
    3. Lewin SR,
    4. et al
    . XVI International AIDS Conference. 2006. Characteristics and predictors of hepatic flares within a randomized study of hepatitis B (HBV) therapy in antiretroviral naïve HIV/HBV coinfected patients in Thailand.
  24. ↵
    1. Romagnani P,
    2. Maggi L,
    3. Mazzinghi B,
    4. et al
    . CXCR3-mediated opposite effects of CXCL10 and CXCL4 on TH1 or TH2 cytokine production. J Allergy Clin Immunol 2005;116:1372-9.
    CrossRefMedlineWeb of Science
  25. ↵
    1. Gangur V,
    2. Simons FE,
    3. Hayglass KT
    . Human IP-10 selectively promotes dominance of polyclonally activated and environmental antigen-driven IFN-gamma over IL-4 responses. FASEB J 1998;12:705-13.
    Abstract/FREE Full Text
  26. ↵
    1. Taub DD,
    2. Lloyd AR,
    3. Conlon K,
    4. et al
    . Recombinant human interferon-inducible protein 10 is a chemoattractant for human monocytes and T lymphocytes and promotes T cell adhesion to endothelial cells. J Exp Med 1993;177:1809-14.
    Abstract/FREE Full Text
  27. ↵
    1. Roe B,
    2. Coughlan S,
    3. Hassan J,
    4. et al
    . Elevated serum levels of interferon-γ–inducible protein–10 in patients coinfected with hepatitis C virus and HIV. J Infect Dis 2007;196:1053-7.
    Abstract/FREE Full Text
  28. ↵
    1. Falasca K,
    2. Ucciferri C,
    3. Dalessandro M,
    4. et al
    . Cytokine patterns correlate with liver damage in patients with chronic hepatitis B and C. Ann Clin Lab Sci 2006;36:144-50.
    Abstract/FREE Full Text
  29. ↵
    1. Tan DB,
    2. Yong YK,
    3. Tan HY,
    4. et al
    . Immunological profiles of immune restoration disease presenting as mycobacterial lymphadenitis and cryptococcal meningitis. HIV Med 2008;9:307-16.
    CrossRefMedlineWeb of Science
  30. ↵
    1. Chang J,
    2. Wightman F,
    3. Bartholomeusz A,
    4. et al
    . Reduced HBV-specific CD4+ T-cell responses in HIV-1-HBV co-infected individuals receiving HBV-active antiretroviral therapy. J Virol 2005;79:3038-51.
    Abstract/FREE Full Text
  31. ↵
    1. Kennedy MK,
    2. Willis CR,
    3. Armitage RJ
    . Deciphering CD30 ligand biology and its role in humoral immunity. Immunology 2006;118:143-52.
    CrossRefMedlineWeb of Science
  32. ↵
    1. Stone SF,
    2. Price P,
    3. Tay-Kearney ML,
    4. French MA
    . Cytomegalovirus (CMV) retinitis immune restoration disease occurs during highly active antiretroviral therapy–induced restoration of CMV-specific immune responses within a predominant Th2 cytokine environment. J Infect Dis 2002;185:1813-7.
    Abstract/FREE Full Text
  33. ↵
    1. Leonard EJ,
    2. Yoshimura T
    . Human monocyte chemoattractant protein-1 (MCP-1). Immunol Today 1990;11:97-101.
    CrossRefMedlineWeb of Science
  34. ↵
    1. Zeremski M,
    2. Petrovic LM,
    3. Talal AH
    . The role of chemokines as inflammatory mediators in chronic hepatitis C virus infection. J Viral Hepat 2007;14:675-87.
    MedlineWeb of Science
  35. ↵
    1. Shields PL,
    2. Morland CM,
    3. Salmon M,
    4. Qin S,
    5. Hubscher SG
    . Chemokine and chemokine receptor interactions provide a mechanism for selective T cell recruitment to specific liver compartments within hepatitis C-infected liver. J Immunol 1999;163:6236-43.
    Abstract/FREE Full Text
  36. ↵
    1. Helbig KJ,
    2. Ruszkiewicz A,
    3. Semendric L,
    4. Harley HA,
    5. McColl SR
    . Expression of the CXCR3 ligand I-TAC by hepatocytes in chronic hepatitis C and its correlation with hepatic inflammation. Hepatology 2004;39:1220-9.
    CrossRefMedlineWeb of Science
  37. ↵
    1. Chang JJ,
    2. Thompson AJ,
    3. Visvanathan K,
    4. et al
    . The phenotype of hepatitis B virus-specific T cells differ in the liver and blood in chronic hepatitis B virus infection. Hepatology 2007;46:1332-40.
    CrossRefMedlineWeb of Science
| Table of Contents

This Article

  1. J Infect Dis. (2009) 199 (7): 974-981. doi: 10.1086/597276
  1. AbstractFree
  2. » Full Text (HTML)Free
  3. Full Text (PDF)Free

Classifications

Share

  1. Email this article

Navigate This Article

Current Issue

  1. March 1, 2012 205 (5)
  1. Journal of Infectious Diseases
  1. Alert me to new issues

Most