Lamivudine plus interleukin-12 combination therapy in chronic hepatitis B: Antiviral and immunological activity

Authors


  • Potential conflict of interest: Dr. Naoumov received an unrestricted grant from GSK.

Abstract

Interleukin-12 (IL-12) is an immunomodulatory cytokine that promotes cellular immunity. Pre-clinical data suggest that IL-12 inhibits hepatitis B virus (HBV) replication by stimulating interferon-gamma (IFN-γ) production. We investigated whether a combination treatment with lamivudine plus recombinant human interleukin-12 (rhIL-12) will result in a greater and prolonged suppression of HBV replication in comparison with lamivudine monotherapy. Fifteen patients with HBeAg-positive chronic hepatitis B were randomized to receive either lamivudine alone for 24 weeks (group 1); combination of lamivudine for 16 weeks and rhIL-12 (200 ng/kg twice weekly), starting 4 weeks after initiation of lamivudine, for 20 weeks (group 2), or the same schedule as for group 2, with lamivudine and a higher dose of rhIL-12 (500 ng/kg, group 3). Serum HBV DNA levels, T-cell proliferation, frequency of virus-specific T-cells, and IFN-γ production were evaluated serially during and 24 weeks posttreatment. Lamivudine plus rhIL-12/500 showed greater antiviral activity than lamivudine monotherapy. However, after stopping lamivudine in groups 2 and 3, serum HBV DNA increased significantly despite continuing rhIL-12 administration. Lamivudine plus rhIL-12 treatment was associated with a greater increase in virus-specific T-cell reactivity, IFN-γ production, and an inverse correlation between the frequency of IFN-γ–producing CD4+ T-cells and viremia. The T-cell proliferative response to HBcAg did not differ between the three groups. In conclusion, the addition of IL-12 to lamivudine enhances T-cell reactivity to HBV and IFN-γ production. However, IL-12 does not abolish HBV replication in HBeAg-positive patients and does not maintain inhibition of HBV replication after lamivudine withdrawal. (HEPATOLOGY 2005.)

Hepatitis B virus (HBV) is a noncytopathic DNA virus, which can cause acute, self-limited hepatitis or chronic hepatitis, cirrhosis, and/or hepatocellular carcinoma.1 The diversity of clinical outcomes after exposure to HBV is determined primarily by the host immune response.2, 3 The resolution of HBV infection after acute self-limited hepatitis is associated with strong CD4+ and CD8+ T-cell responses, with a type 1 cytokine profile.4–7 In contrast to this strong antiviral T-cell reactivity, which persists in the convalescent phase after hepatitis B surface antigen (HBsAg) clearance, patients with chronic HBV infection have weak or undetectable T-cell reactivity to HBV, which is the dominant factor that permits an ongoing, high level of viral replication. A series of investigations of the mechanisms by which T-cells control HBV replication showed the major role of cytokine-mediated, intracellular inactivation of HBV that does not require cell death.8 This noncytolytic antiviral effect, first demonstrated in an HBV transgenic mouse model and subsequently in other animals, is primarily mediated by interferon-gamma (IFN-γ).9–11 A detailed analysis of HBV clearance during acute infection in chimpanzees showed that the disappearance of viral DNA from the liver coincides with IFN-γ induction before the increase of serum alanine aminotransferase (ALT), in other words, without destruction of hepatocytes.11

Interleukin-12 (IL-12) is a proinflammatory cytokine, produced by activated phagocytic cells (monocytes/macrophages) and dendritic cells, that plays a central role in stimulating cell-mediated immunity.12, 13 IL-12 stimulates natural killer cells and T-lymphocytes to produce IFN-γ, promotes T-helper 1 responses, and enhances CD8+ cytotoxic T-cell activity. These unique properties of IL-12 indicate that it might have an important role in achieving sustained control of HBV replication.14 This notion is supported by preclinical data, because the administration of recombinant IL-12 to HBV transgenic mice resulted in complete inhibition of HBV replication in the liver and undetectable viremia, mediated through IFN-γ induction.15 In addition, IL-12 restored in vitro the hypo-responsiveness to viral antigens of T-cells obtained from patients with chronic hepatitis B.16, 17

We conducted a pilot study to investigate in patients with chronic hepatitis B whether a combination treatment with lamivudine plus recombinant human interleukin-12 (rhIL-12) will result in a greater antiviral effect and prolonged suppression of HBV replication in comparison with lamivudine monotherapy. In addition, by monitoring longitudinally the HBV-specific T-cell reactivity, the study aimed to determine whether IL-12–stimulated T-cell responsiveness to HBV can maintain the control of viral replication after stopping the antiviral agent.

Abbreviations

HBV, hepatitis B virus; HBeAg, hepatitis B e antigen; HBcAg, hepatitis B core antigen, IL-12, interleukin-12; IFN-γ, interferon-gamma; HIV, human immunodeficiency virus; ALT, alanine aminotransferase; ULN, upper limit of normal; PBMC, peripheral blood mononuclear cells; SI, stimulation index.

Patients and Methods

Patients.

Fifteen patients with chronic hepatitis B, followed at University College Hospital in London, were included in this study (Table 1). All patients were seropositive for HBsAg, hepatitis B e antigen (HBeAg), and HBV DNA for more than 12 months. No patient had received antiviral treatment for at least 12 months before entry into this study. Seven patients were treatment naïve, and 8 had previous treatment attempts that failed: with interferon monotherapy for 6 months (n = 7); famciclovir (n = 1), or adefovir (n = 1) for 6 months, or interferon plus lamivudine for 4 months (n = 4). Five of the 8 patients had failed two previous treatments. All patients were negative for anti-hepatitis D virus, anti-hepatitis C virus, anti-human immunodeficiency virus (HIV)-1/2, and autoantibodies. Pretreatment serum alanine aminotransferase (ALT) levels were normal in 6 and raised—1.5 to 5× upper limit of normal (ULN) in 9 patients. A liver biopsy was performed in each patient as part of the routine diagnostic evaluation, and the inflammation grade and fibrosis stage were scored according to established criteria.18

Table 1. Baseline Characteristics of Patients
CharacteristicsLamivudine Monotherapy (n = 5)Lamivudine + IL-12/200 (n = 5)Lamivudine + IL-12/500 (n = 5)Total (n = 15)
  • *

    The data shown are mean ± SEM.

Sex (M/F)2/35/04/111/4
Age (year)*33.6 ± 3.139.0 ± 3.042.2 ± 3.638.3 ± 1.9
Race    
 White2338
 Asian3227
Previous treatment    
 None4127
 Failed1438
HBV DNA (mEq/mL)*2581 ± 10182854 ± 12893336 ± 881 
ALT (IU/L)*76 ± 23106 ± 2883 ± 23 
HLA-A2 positive2114

Study Design and Monitoring.

This was a pilot study comparing lamivudine monotherapy with a combination of lamivudine plus rhIL-12 at two doses (200 ng/kg or 500 ng/kg). Before randomization into the treatment groups, patients were divided into two strata according to the histological activity index and ALT levels. Stratum 1 included patients with mild liver disease, defined as hepatic necroinflammation grade (<2) and fibrosis stage (<1), and normal ALT levels. Stratum 2 included patients with moderate hepatitis—inflammation grade between 3 and 6, fibrosis stage 2 to 5, and raised ALT. After randomization, 2 patients from stratum 1 and 3 patients from stratum 2 were included in each group. Patients were assigned to one of the three regimens: (i) lamivudine monotherapy (GlaxoSmithKline, Greenford, UK) 100 mg daily for 24 weeks; (ii) lamivudine for 16 weeks plus rhIL-12 (Genetics Institute, Cambridge, MA) at 200 ng/kg given twice weekly subcutaneously for 20 weeks, starting 4 weeks after initiation of lamivudine; (iii) lamivudine for 16 weeks with rhIL-12 at 500 ng/kg given twice weekly subcutaneously for 20 weeks starting 4 weeks after initiation of lamivudine. Both the dosage and duration of rhIL-12 administration in the current study were greater than in previous pilot studies of rhIL-12 monotherapy in chronic hepatitis B or C.19, 20 The study drug was supplied and quality certified by Wyeth-Ayerst Research Pharmaceutical (Gosport, UK). The treatment protocol and associated investigations were approved by the UCL Hospital Ethics Committee, and all patients gave a written informed consent. Patients were monitored every 2 weeks for the first 8 weeks of treatment and then monthly during treatment. Posttreatment, patients were reviewed every 8 weeks for a period of 24 weeks. At each visit, the tolerability of treatment and adverse events were assessed together with physical examination, routine biochemistry, and hematology testing.

T-Cell Proliferation Assay.

T-cell proliferative response to hepatitis B core antigen (HBcAg) was analyzed with fresh peripheral blood mononuclear cells (PBMC), as previously described.21 Briefly, PBMC were isolated from heparinized blood by density gradient centrifugation and were cultured for 6 days in 96-well plates (2 × 105 cells/ well) in RPMI-1640 with 10% (vol/vol) heat-inactivated human AB serum (Gemini, Calabasas, CA) with and without HBcAg (1 μg/mL American Research Products, Belmont, MA). Tetanus toxoid (0.5 μg/mL, Connaught Laboratories, Ontario, Canada) and phytohemagglutinin (Sigma, Dorset, UK) were included as positive controls. The proliferative response was evaluated by [3H]-thymidine uptake and a β-counter (Wallac, Turku, Finland). The stimulation index (SI) was calculated from the mean counts per minute in the test wells divided by that found in the control wells with PBMC with medium alone. HBcAg response was tested also in 20 controls (HBsAg and anti-HBc negative). A significant proliferative response was defined as SI >2.5, which was greater then the mean counts per minute plus 3 standard deviations in the negative controls. By depletion of CD4+ T-cells from PBMC samples with a significant SI, we confirmed that the proliferative response to HBcAg and tetanus toxoid were due to CD4+ T-cells.

In Vitro IFN-γ Production.

In parallel with the T-cell proliferation assay, fresh PBMC (3×105 cells/well) were cultured with and without HBcAg for 6 days. After this period, the plates were centrifuged and the supernatants were stored at −20°C. The IFN-γ level in the supernatants was measured by a commercial enzyme immunoassay (ELISA) (R&D Systems, Abingdon, UK).

Enumeration of HBV-Specific, IFN-γ–Producing T-Cells.

PBMC were cryopreserved during the study and used to determine the frequency of HBcAg-specific, IFN-γ–producing T-cells using an Elispot assay.21 The PBMC samples from all time-points per patient were tested simultaneously. Briefly, nitrocellulose-bottom 96-well plates (Millipore, Bedford, MA) were coated overnight at 4°C with monoclonal antibody to IFN-γ (anti–IFN-γ, Mabtech, Nacka, Sweden) at 10 μg/mL in phosphate-buffered saline. In parallel, PBMC (2 × 105/well) were incubated in triplicate in RPMI/AB serum with HBcAg (final concentration 2 μg/mL), tetanus toxoid (0.5 μg/mL), PHA (2 μg/mL) or medium alone in round-bottom plates (Becton-Dickinson, Oxford, UK) at 37°C for 28 hours. The unbound antibody was removed from the nitrocellulose-bottom plates by washings with phosphate-buffered saline/Tween 0.05%, and the membrane was blocked with RPMI/10% AB serum for 2 hours. Subsequently, PBMC were transferred from the round-bottom plates to the coated plates and incubated for 20 hours. After multiple washings, biotin-conjugated anti–IFN-γ (Mabtech) at 1 μg/mL was added for 2 hours at room temperature. The plates were washed, and streptavidin-alkaline phosphatase (Mabtech) was added for 2 hours in the dark. The unbound conjugate was removed and the substrate, nitro-blue tetrazolium chloride/bromo-chloro-indolyl-phosphatetoluidune-salt (Roche Diagnostics, Lewes, UK), was added to the wells. The reaction was stopped, and after drying, the spots were counted with an Elispot reader (AID, Strassberg, Germany). The number of specific spot-forming cells was determined by subtracting the mean number of spots in the wells with medium alone from the mean number in wells with antigens, and expressed per 1 × 106 PBMC. In repeat testing of certain time points, the depletion of CD4+T-cells using immunomagnetic beads (Miltenyi Biotech, Bisley, UK) abrogated the response to HBcAg and tetanus, thus confirming that the spot-forming cells to the recombinant proteins represent CD4+T-cell reactivity.

To determine the frequency of HBcAg-specific IFN-γ–producing CD8+ T-cells, PBMC (2 × 105/well) from HLA-A2–positive patients were incubated in round-bottom plates with a synthetic peptide (Mimotope, Clayton, Australia) representing a known HLA-A2–restricted epitope in HBcAg amino acids 18-27,2 at a concentration of 10 μg/mL. Anti-CD3 (10 μg/mL) was used as a positive control. The detection of IFN-γ–producing cells was as described above.

Hepatitis Serology and HBV-DNA Quantitation.

Serum HBsAg, HBeAg, and anti-HBe were tested by commercial immunoassays (Abbott Diagnostics, Maidenhead, UK). Serum HBV-DNA was quantitated at all time points with a signal amplification assay (Quantiplex bDNA, Chiron Corporation, Emerville, CA). The threshold for HBV-DNA detection is 0.7 × 106 copies/mL (0.7 mEq/mL). To further investigate the antiviral effect of treatment, serum samples between treatment week 2 and treatment week 16, were also tested with a quantitative polymerase chain reaction assay (NGI, National Genetics Institute, Cambridge, MA) with a lower limit of detection 100 copies/mL.

Statistics.

The results were analyzed by ANOVA with Fisher's protected least significant difference, Wilcoxon matched pairs test, or Mann-Whitney test. The Spearman correlation coefficient was used where appropriate. A two-sided P value less than .05 was considered significant. Statistical analyses were performed using the SPSS v.11.0 (SPSS Inc., Chicago, IL) package.

Results

Virological Response to Treatment

The pretreatment HBV DNA levels (mean ± SEM) were 2,581 ± 1,018, 2,854 ± 1,289, and 3,336 ± 881 mEq/mL in the lamivudine monotherapy, lamivudine plus IL-12/200, and lamivudine plus IL-12/500 groups, respectively (Fig. 1). After 4 weeks of lamivudine monotherapy, serum HBV DNA levels were markedly reduced in all patients, being undetectable with the signal amplification assay in seven of 15 patients—three in lamivudine, two in lamivudine plus IL-12/200, and two in the lamivudine plus IL-12/500 group.

Figure 1.

Serum HBV-DNA (bDNA assay) and ALT levels during the course of treatment and posttreatment follow up: group 1, lamivudine monotherapy; group 2, lamivudine plus IL-12/200 ng/kg; group 3, Lamivudine plus IL-12/500 ng/kg); *P = .02, for the difference in the HBV-DNA levels between baseline and week 20; **P = .03, for the difference in the HBV-DNA levels between baseline and week 24; BL, baseline; HBV, hepatitis B virus; ALT, alanine aminotransferase.

The viremia levels after 4 weeks' treatment with lamivudine alone (treatment week 4) were not different between the three groups when quantified with the polymerase chain reaction–based assay (NGI): HBV DNA log10copies/mL (mean ± SEM) was 6.43 ± 0.36, 6.95 ± 0.48, 6.61 ± 0.41 in the lamivudine, lamivudine plus IL-12/200, and lamivudine plus IL-12/500 groups, respectively (P < .05 for all comparisons, Fig. 2A). Between treatment week 4 and treatment week 16, all treatment regimens resulted in a further reduction of viremia. In particular, at treatment week 16 the HBV DNA levels were 5.7 ± 0.44, 5.01 ± 0.94 and 4.39 ± 0.6 in lamivudine, lamivudine plus IL-12/200, and lamivudine plus IL-12/500 group, respectively (Fig. 2A). The mean HBV DNA reduction (log10copies/mL) from treatment week 4 to treatment week 16 with lamivudine monotherapy was 0.6 (range, 0.02-1.14; P = .07). The combination regimens showed a greater, dose-dependent antiviral effect. Between treatment week 4 and treatment week 16, lamivudine plus IL-12/200 reduced the HBV DNA levels by 1.64 log10copies/mL (range, 0.97-2.85, P = .04); lamivudine plus IL-12/500 by 2.2 log10copies/mL (range, 1.02-2.88; P = .04; Fig. 2A). The viral load decline from treatment week 4 to treatment week 16 was significantly higher with the lamivudine plus IL-12/500 combination therapy compared with lamivudine alone (P = .01), but not between lamivudine plus IL-12/200 versus lamivudine (P = .9, ANOVA). The greater, dose-dependent antiviral effect of lamivudine plus IL-12 was also apparent when assessing the number of patients with greater than 2 log10 reduction of viremia between treatment week 4 and treatment week 16—0/5 in lamivudine, 2/5 in lamivudine plus IL-12/200; 4/5 in lamivudine plus IL-12/500 group (Fig. 2B).

Figure 2.

The antiviral effect of lamivudine plus IL-12 combination is greater than lamivudine monotherapy. (A) Serum HBV-DNA (log10copies/mL) determined by polymerase chain reaction assay from treatment week (treatment week) 2 to treatment week 16.*P = .04, for the difference in the HBV-DNA (log10copies/mL) between treatment week 4 and treatment week 16 in the lamivudine plus IL-12/200 group; *P = .04, for the difference in the HBV-DNA (log10 copies/mL) between treatment week 4 and treatment week 16 in the lamivudine plus IL-12/500 group; **P = .01 for the reduction of HBV-DNA (log10copies/mL) from treatment week 4 to treatment week 16 between the lamivudine and the lamivudine plus IL-12/500 group. (B) Number of patients in each treatment group who had viral load reduction more than 1 and more than 2 log10HBV DNA copies/mL between treatment week 4 and treatment week 16. IL-12, interleukin-12; HBV, hepatitis B virus.

After stopping lamivudine at treatment week 16 in the combination arms, rhIL-12 was continued up to treatment week 24. Serum HBV DNA levels increased at weeks 20 and 24 during IL-12 monotherapy in both groups, but were lower compared with baseline. In group 3 (IL-12/500), HBV DNA levels were significantly lower at treatment week 20 and treatment week 24 compared with baseline (912 ± 643 vs. 3,336 ± 881 mEq/mL, P = .02 and 1,503 ± 432 vs. 3,336 ± 881 mEq/mL, P = .03, respectively; Fig. 1). In group 2 (IL-12/200), HBV DNA levels were lower at treatment week 20 and treatment week 24, but not significantly different compared with baseline (1,367 ± 670 vs. 2,854 ± 1,289 mEq/mL, P = .2 and 1,782 ± 731 vs. 2,854 ± 1,289 mEq/mL, P = .4, Fig. 1). In group 1, lamivudine monotherapy was stopped at treatment week 24, which was followed by a rapid rebound of viremia with HBV DNA at week 32 exceeding the baseline levels: 3,972 ± 337 vs. 2,581 ± 1,018 mEq/mL (P = .2, Fig. 1).

A transient HBeAg loss was seen in 2 of 15 patients, both on combination therapy—lamivudine plus IL-12/200 (n = 1), or lamivudine plus IL-12/500 (n = 1). No HBeAg loss, or anti-HBe seroconversion, were observed during the posttreatment follow-up.

Biochemical Response to Treatment

In patients receiving lamivudine monotherapy, serum ALT levels decreased at treatment week 16 when compared with baseline (mean ± SEM): 32 ± 7 versus 76 ± 23 IU/L (P = .07,Wilcoxon test). An ALT flare (defined as ALT elevation >3×ULN) was observed in three of five patients posttreatment (group 1, Fig. 1). Serum ALT in these three patients increased to 250 ± 64 and 511 ± 224 IU/L, 8 and 16 weeks after stopping lamivudine, respectively. The flares were clinically asymptomatic and were not associated with decompensated liver functions. Serum ALT in the two patients who did not develop a flare were 30 ± 7 and 31 ± 2, at weeks 8 and 16 after stopping lamivudine.

The pattern of serum ALT levels in the ten patients receiving lamivudine plus IL-12 was different. There was no decrease in ALT levels in patients receiving combination treatment between baseline and treatment week 16 (95 ± 17 vs. 85 ± 74, P = .2,Wilcoxon test). One of ten patients (on lamivudine plus IL-12/500) developed a transient flare (ALT was 4.5 × ULN at treatment week 16), with no decompensation of liver function. Importantly, none of the ten patients developed an ALT flare when lamivudine was stopped at treatment week 16 and treatment continued with IL-12 alone, and there were no flares after cessation of IL-12 monotherapy.

T-Cell Responses to HBV During Treatment

T-Cell Proliferation.

HBcAg-specific T-cell proliferation increased during treatment and peaked at treatment week 16 (Fig. 3). There were no significant changes in HBcAg-specific T-cell reactivity in all groups during the initial 4 weeks of treatment with lamivudine alone (P = .8,Wilcoxon test). The SI (mean ± SEM) increased from treatment week 4 to treatment week 16—from 2.8 ± 1.0 to 4.7 ± 0.7 in the lamivudine group (P = .043); 4.0 ± 1.0 to 6.0 ± 1.4 (P = .043) in the lamivudine plus IL-12/200 group, and 3.9 ± 1.8 to 9.6 ± 4.9 (P = .043) in the lamivudine plus IL-12/500 group. The biggest SI increase was seen in patients receiving lamivudine plus IL-12/500; however, the differences between the three groups at treatment week 4, treatment week 8, treatment week 16, or treatment week 24 were not significant (P >.05 for all comparisons, ANOVA). The proliferative response to tetanus antigen did not increase in patients receiving IL-12, and there was no significant difference in the SI at any time between the three treatment groups.

Figure 3.

HBcAg-specific T-cell responses during the course of treatment and posttreatment follow up. *P = .043; SI, stimulation index; BL, baseline; HBcAg, hepatitis B core antigen.

In patients who received lamivudine monotherapy, a marked increase occurred in the proliferative response to HBcAg (SI 9.2 ± 5.2) at week 48, 24 weeks after stopping lamivudine. This increase in T-cell proliferation to HBcAg followed the ALT flare in three of five patients after stopping lamivudine. In detail, the SI (mean ± SEM) was 21 ± 6 in three patients with post-lamivudine flare and 1.4 ± 0.5 in two patients without a flare.

Enumeration of HBcAg-Specific, IFN-γ–Producing T-Cells.

The frequency of HBcAg-specific IFN-γ–producing CD4+T-cells increased after 4 weeks of lamivudine treatment, compared with baseline, in 15 patients studied (P = .023,Wilcoxon test). Throughout the treatment period, no significant difference was found in HBV-specific IFN-γ–producing CD4+T-cells between patients receiving lamivudine alone and those receiving lamivudine plus IL-12/200 (Fig. 4A). In contrast, a marked increase occurred in the frequency of IFN-γ–producing cells in patients receiving lamivudine plus IL-12/500 at treatment week 8, which was significantly higher than the frequency of virus-specific CD4+ T-cells in patients receiving lamivudine monotherapy (349 ± 60 vs. 142.5 ± 36.2, P = .043, ANOVA) or those receiving lamivudine plus IL-12/200 (349 ± 60 vs. 162.4 ± 71, P = .05, Fig. 4A). The peak of HBcAg-specific, IFN-γ–producing CD4+T-cells was observed earlier (treatment week 8) in the lamivudine plus IL-12/500 group, compared with the lamivudine monotherapy and the lamivudine plus IL-12/200 groups (treatment week 16, Fig. 4A). An inverse relationship occurred between the frequency of HBV-specific CD4+T-cells and serum HBV DNA levels between treatment week 4 and treatment week 16. This was significant only for the two combination arms (r = −1.0, P = .001 for both arms, Spearman correlation) (Fig. 5). The frequency of IFN-γ–producing CD4+T-cells in response to tetanus antigen showed no significant changes during 24-week treatment period in any of the three groups.

Figure 4.

(A) Enumeration of HBcAg-specific, IFN-γ–producing CD4+T-cells in the 3 groups of patients. *P = .043; **P = .05. (B) Individual data for HBcAg-specific, IFN-γ–producing CD8+T-cells in 4 HLA-A2–positive patients during treatment and posttreatment follow up. SFC, specific spot forming cells; BL, baseline; HBcAg, hepatitis B core antigen; IFN-γ, interferon gamma.

Figure 5.

Correlation between the number of HBcAg-specific IFN-γ–producing CD4+T cells and serum HBV DNA (log10copies/mL) levels in all treatment groups (group 1, lamivudine; group 2, lamivudine plus IL-12/200; group 3, lamivudine plus IL-12/500) between weeks 4 and 16 of treatment (r = −1.0, P = .001 for both combination arms, Spearman correlation). HBcAg, hepatitis B core antigen; IFN-γ, interferon gamma.

Only 4 of 15 patients studied were HLA-A2 positive (Table 1). The frequency of HBcAg-specific CD8+T cells increased in all 4 patients during treatment but was greater in the two patients receiving IL-12 (Fig. 4B).

HBcAg-Specific IFN-γ Production.

The IFN-γ production from PBMC in response to HBcAg in vitro increased during treatment with lamivudine alone and during lamivudine plus IL-12/200 (Fig. 6). Because of insufficient number of cells, testing the IFN-γ production from patients receiving lamivudine plus IL-12/500 was not possible. Treatment with lamivudine plus rhIL-12/200 increased IFN-γ production at treatment weeks 8, 16, and 24, compared with treatment week 4 (P = .07 for all, Wilcoxon test). The IFN-γ production in response to HBcAg at treatment week 24 was significantly higher in the lamivudine plus IL-12/200 group, compared with lamivudine monotherapy (P = .034, Fig. 6). Patients receiving the combination maintained high IFN-γ production for a longer period (treatment week 6 to treatment week 24) compared with patients receiving lamivudine monotherapy (treatment week 6 to treatment week 16). The IFN-γ production decreased rapidly after stopping IL-12 at treatment week 24 (Fig. 6).

Figure 6.

HBcAg-specific IFN-γ production in patients treated with lamivudine alone and lamivudine plus IL-12/200. *P = .034; BL, baseline; HBcAg, hepatitis B core antigen; IFN-γ, interferon gamma.

Safety and Tolerability

No serious adverse events occurred during or after subcutaneous applications of rhIL-12 at 200 or 500 ng/kg twice weekly. A range of expected side effects were observed after the first two-three injections, which involved pyrexia, headache, myalgia, arthralgia, asthenia, flu-like symptoms, and insomnia. These were transient and usually grade 1 to 2 in severity. Very few symptoms were seen thereafter apart from tiredness. One patient (receiving lamivudine plus IL-12/500) developed productive cough. The chest x-ray was normal, sputum cultures were sterile and the cough subsided after completing IL-12 treatment. The laboratory abnormalities included absolute lymphopenia in 2 patients receiving rhIL-12/500 ng/kg, which resolved on treatment cessation. Three of 5 patients on lamivudine monotherapy developed posttreatment ALT flares (weeks 32-48), which were not observed in any of the 10 patients after combination treatment.

Discussion

This study demonstrates that administration of rhIL-12 in patients with chronic hepatitis B stimulates Th1 cell reactivity to HBV, increases the frequency of virus-specific CD4+ T-cells, and enhances IFN-γ production. As a result, the addition of IL-12 to lamivudine treatment led to a dose-dependent increase in antiviral activity with an inverse relationship between the frequency of HBV-specific, IFN-γ–producing CD4+T-cells and viremia levels. The antiviral effect of IL-12 was not associated with serum ALT increase compared with baseline, thus providing evidence in patients with chronic hepatitis B for IFN-γ–mediated, non-cytolytic inhibition of HBV replication, as previously demonstrated in animal models.9–11 However, unlike in HBV transgenic mice, where hepatic HBV DNA and viremia were abolished after only three injections of rIL-12,15 in patients with chronic hepatitis B, continued rhIL-12 administration, after stopping lamivudine, was not able to maintain suppression of HBV replication. In a previous study using 3 doses of rhIL-12 alone (30, 250 or 500 ng/kg, once per week) for 12 weeks, the reduction in serum HBV DNA was modest, with no significant difference between the three groups.19 In our study, rhIL-12 was added to lamivudine, which increased significantly the on-treatment antiviral effect. Similarly, pegylated interferon-alfa plus lamivudine showed a greater antiviral effect than monotherapy, but this was not sustained after stopping treatment.22, 23 The antiviral effect of rhIL-12 was also apparent after stopping lamivudine (treatment week 16 to treatment week 24) in groups 2 and 3. Although the suppression of HBV replication was not sustained, viremia increased slowly, in contrast to the rapid viremia rebound in group 1 associated with post-lamivudine ALT flares. Both the dosage and duration of rhIL-12 therapy that we used were greater than in pilot monotherapy trials,19, 20 and a further dose increase is unlikely to be tolerated. A trial with rhIL-12 (500 ng/kg twice weekly) in chronic hepatitis C was terminated prematurely because of poor tolerance and treatment-related severe adverse events.24

The current study extends the in vitro data that rhIL-12 can stimulate PBMC from chronic hepatitis B patients to produce IFN-γ and increases their reactivity to HBV antigens.16, 17, 25 The study design included a lead period with lamivudine treatment, based on our earlier in vitro data that the antiviral effect of IFN-γ was more pronounced in human hepatocytes with low numbers of HBV DNA copies/cell, whereas it had minimal effect in cases with a high viral load.16 This rationale is supported by a recent, longitudinal analysis of HBV-specific CD8+T-cells in chronic HBV patients, which found an inverse relationship with viremia, and these cells were present/detectable in circulation only in patients with HBV DNA <107copies/mL.26

IL-12 is an important cytokine for immune control of intracellular pathogens. Application of IL-12 was protective against infections controlled by Th1 immune responses.27, 28 However, its efficacy for treatment of chronic infections seems limited.29 In experimental models, IL-12 application has shown benefit in combination with antibiotics in Mycobacterium-infected SCID mice30 or with chloroquine for malaria.31In vitro, IL-12 stimulation of PBMC from HIV-infected patients showed that these cells are responsive to IL-1232; however, in phase I clinical trial, 4-week treatment with rhIL-12 (doses from 30 to 300 ng/kg, twice weekly) showed no improvement of T-cell reactivity to HIV or recall antigens and no reduction in viremia.33

Although the current study demonstrates in vivo that T-cells from chronic HBV patients can respond to IL-12 stimulation with increased frequency of HBV-specific cells and enhanced IFN-γ production, contrary to the expectation, the combination treatment with lamivudine plus IL-12 did not result in true restoration of antiviral immunity with sustained control of HBV replication. Longitudinal analysis of serum IL-12 levels in patients undergoing interferon-alfa treatment showed increased endogenous IL-12 production, along with IFN-γ and interleukin-2, as markers of immune activation in treatment responders, but not in non-responders.34 Prolonged administration of high doses of exogenous rhIL-12, however, may have limited efficacy in chronic hepatitis B, for several reasons. First, the current study showed that its effect on HBV-specific T-cell proliferation is modest, with no significant difference between the treatment groups. Experimental data indicate that IL-12 does not induce proliferation of resting peripheral T-cells, whereas it stimulates directly the proliferation of pre-activated T-cells.13 Therefore, IL-12 may be more useful as an adjuvant to therapeutic vaccination. Second, the frequency of HBV-specific CD4+T-cells plateau after treatment week 16 and even decreased after stopping lamivudine, which could be attributable to lymphopenia, as observed in patients receiving rhIL-12/500 ng/kg, or the rebound in viral replication. Third, IL-12–induced IFN-γ secretion continued until treatment week 24, but dropped thereafter. Two different pathways for the induction of IFN-γ production from T-cells have been identified: stimulation with IL-12 alone, or stimulation through T-cell receptor.35 The latter may be preferable if associated with expansion of antiviral T-cells. Fourth, IFN-γ responsiveness may be altered in hepatocytes by persistent HBV replication. In controlled clinical trials, treatment with recombinant IFN-γ or interleukin-2 in chronic hepatitis B had no antiviral effect, and there were no ALT flares.36, 37In vitro treatment of chronically infected woodchuck hepatocytes with IFN-γ, or in vivo IFN-γ expression in liver, did not suppress woodchuck hepatitis virus replication in chronically infected animals.38, 39

The treatment strategy in chronic hepatitis B aims to suppress HBV replication and to restore HBV-specific T-cell responses to achieve sustained remission. Suppression of HBV replication with an antiviral agent can enhance T-cell reactivity to HBV in some patients,40 but this is transient41, 42 and does not occur in all patients.43 Therefore, a combination of antiviral and immunomodulatory drugs is likely to be more successful. The failure of lamivudine plus rhIL-12 to achieve sustained improvement suggests that combination therapy with an antiviral drug and immunomodulatory cytokine is unlikely to establish sustained resolution of HBV replication. The use of therapeutic vaccines or cell-based immunotherapies, together with antiviral drugs, could be more successful, because these will increase both the frequency and functionality of virus-specific effector cells. We have shown that adoptive transfer of immunity to HBV, with T-cells reactive to HBcAg in bone marrow recipients, results in resolution of chronic HBV infection.21 Immunotherapy with dendritic cells can induce efficient antiviral T-cells responses,44 and a clinical application of virus-pulsed dendritic cells in HIV-infected patients was effective in inducing lasting Th1 responses with prolonged viral suppression.45 Another approach, which deserves further investigations, is ex vivo expansion of autologous virus-specific T-cells. Infusions of such virus-specific T-cells were effective for controlling cytomegalovirus replication in bone-marrow transplant recipients.46

Acknowledgements

The authors thank Miss Tania Machel for her assistance in the clinical monitoring of patients.

Ancillary