Treatment of chronic hepatitis B virus infection via oral immune regulation toward hepatitis B virus proteins

THE AMERICAN JOURNAL OF GASTROENTEROLOGY © 2003 by Am. Coll. of Gastroenterology Published by Elsevier Inc.

Vol. 98, No. 11, 2003 ISSN 0002-9270/03/$30.00 doi:10.1016/S0002-9270(03)00706-8

Treatment of Chronic Hepatitis B Virus Infection via Oral Immune Regulation Toward Hepatitis B Virus Proteins Rifaat Safadi, M.D., Eran Israeli, M.D., Orit Papo, M.D., Oren Shibolet, M.D., Alaa Melhem, M.D., Aharon Bloch, M.D., Mina Rowe, R.N., Ruslana Alper, B.Sc., Athalia Klein, M.A., Nilla Hemed, R.N., Ori Segol, M.D., Barbara Thalenfeld, Ph.D., Dean Engelhardt, Ph.D., Elazar Rabbani, Ph.D., and Yaron Ilan, M.D. Liver Unit, Department of Medicine, and Department of Pathology, Hadassah-Hebrew University Medical Center, Jerusalem, Israel; Emek Medical Center, Afula, Israel; and ENZO Biochem, New York, New York

OBJECTIVES: Hepatitis B virus (HBV) is a noncytopathic virus, and hepatocellular injury is mediated by a defective host antiviral immune response. We have previously shown that antiviral immunity can be modulated through oral feeding of viral proteins. The aims of this study were to determine the safety and efficacy of treatment of patients with chronic HBV by means of p.o. administration of HBV envelope proteins. METHODS: A total of 42 chronic HBV patients were treated p.o. with HBV envelope proteins (HBsAg⫹preS1⫹preS2), three times/wk for 20 –30 wk, and followed for an additional 20 wk. Patients were monitored for HBV-DNA levels, liver enzymes, and liver histology. HBV-directed T cell immune modulation was assessed in vitro by HBV specific T cellproliferation, cytotoxicity, IFN␥, and IL10 ELISPOT assays, and reverse transcription–polymerase chain reaction cytokines assay. RESULTS: Favorable response in one of the primary endpoints was achieved in 28/42 patients (66.6%) by means of p.o. immune regulation. A significant decrease in viral load was observed in 15 patients (35.7%). HBsAg/HBcAg biopsy scores improved in 41% and 57.1% of patients, respectively. Histological improvement in liver necroinflammatory score was noted in 12/40 patients (30%). In all, 80% showed biochemical response. Five of 19 HBeAg positive patients (26.3%) became negative for HBeAg. A favorable augmentation in anti-HBV specific T cell response, with increased HbsAg specific T cell proliferation (78%), cytotoxicity (75%), and IFN␥ positive T cell clones (62.9%) was noted. In addition, a decrease in the IL10 ␥ positive T cell clones was achieved (48.1%). Natural killer T (NKT) lymphocytes increased significantly in all treated patients. CONCLUSIONS: Immune regulation of the anti-HBV immune response via p.o. administration of HBV envelope proteins alleviated the immune-mediated liver injury while augmenting the effective antiviral immunity. (Am J Gastro-

enterol 2003;98:2505–2515. © 2003 by Am. Coll. of Gastroenterology)

INTRODUCTION Chronic hepatitis B virus (HBV) infection is a serious health problem worldwide, as some 300 million persons comprising 5% of the world population are carriers (1). Results of experimental infection in cultured hepatocytes and the presence of asymptomatic carriers of the virus indicate that the virus itself is noncytopathic, and suggest that the antiviral immune response may be accountable for both viral clearance and severity of hepatitis (2). The host antiviral response is a critical determinant of the pathological consequences of the infection (3– 6). After HBV infection, the majority of adult patients develop an effective antiviral response that leads to viral clearance and long lasting antiviral immunity (4). Upon recognition of viral peptides on major histocompatibility (MHC) class I molecules of HBV-infected cells, CD8⫹ T cells either can cure HBV-infected cells through a noncytopathic cytokine-mediated inhibition of HBV replication or can destroy them through perforin-Fas ligand and tumor necrosis factor-7alpha; (TNF␣)–mediated death pathways (7–9). Individuals with acute self-limited HBV infection mount a vigorous polyclonal and multispecific Th and cytotoxic T lymphocytes (CTL) response to epitopes within the HBV envelope (HBe), nucleocapsid, and polymerase proteins (9, 10). This type of response coincides with the maximal elevation of serum alanineaminotransferase levels and precedes clearance of HBe and HBs antigens as well as the development of neutralizing antibodies. In contrast, HBV specific immune response is weak or undetectable in the blood of chronically infected patients (4, 8). In these patients, the immune response fails to clear the virus and continues to be directed at the infected hepatocytes, leading to chronic liver disease (5, 6). HBV is a noncytopathic virus, and the

2506

Safadi et al.

degree of intrahepatic inflammatory leukocytic infiltrate is regarded as a histological hallmark of severity. It has been postulated that the HBV specific immune response is too weak to eliminate the virus from all infected hepatocytes. At the same time, however, the response is strong enough to continuously destroy HBV-infected hepatocytes and to induce a chronic inflammatory liver in persistently infected patients (4, 8). Some HBV-infected patients who do not clear the virus may become carriers. These patients are immunologically tolerant to the virus and may have normal liver histology despite evidence of marked intrahepatic viral load (11–13). Oral immune regulation is the induction of immunological hypo- or hyperresponsiveness toward specific antigens or, through a bystander effect, toward other antigens present at the target site (14). Enhanced immunity toward p.o. administered antigens has been achieved in several systems (15, 16). Recently we and others have shown that it is possible to modulate antiviral immunity (17–21). Of the methods used for immune modulation, p.o. administration of viral structural proteins has emerged as the most efficient and least toxic and therefore has had the greatest potential for clinical application (21). Moreover, p.o. immune regulation effectively regulated the pre-existing antiviral immunity (22). It was also demonstrated that p.o. administration of low doses of HBsAg⫹PreS1⫹PreS2 induced peripheral immune tolerance and downregulated a pre-existing anti-HBV immune response in a murine model (23). Because the net balance between an effective and ineffective subtype of immune responses toward the virus determines viral growth and liver damage, it would seem to be appropriate to treat these patients by inducing antigen specific immune regulation to HBV. The aim of the present study was to determine the safety and efficacy of p.o. administration of HBV envelope proteins to patients chronically infected with hepatitis B virus. The results of the study suggest that induction of p.o. immune regulation toward HBV surface proteins alleviate immune-mediated liver injury and augment effective antiviral immunity.

MATERIALS AND METHODS Patient Population A group of 42 patients was followed in an open label, nonrandomized, single-center, prospective trial. All experiments were carried out according to the guidelines of the Hebrew-University–Hadassah Institutional Committee for Human Clinical Trials and with the Committee’s approval. All experiments were approved by the Israel Ministry of Health Committee for Human Trials. Inclusion Criteria Eligible participants were patients chronically infected with HBV, both male and female, and aged 18 –70 yr. All pa-

AJG – Vol. 98, No. 11, 2003

tients tested positive for hepatitis B surface antigen in serum for at least 6 months and were positive for HBV-DNA for at least 3 months as determined by hybridization or HBVDNA–polymerase chain reaction (PCR). Diagnosis of chronic active HBV infection was based on liver biopsy (active inflammatory response). Both HBeAg or anti-HBe positive patients were enrolled in the study. Patients in whom other antiviral treatments failed, and those who had undergone liver transplantation with evidence of reinfection of the graft and active inflammatory reaction in the liver, were also eligible. Written informed consent was obtained from all patients. Exclusion Criteria Fulminant liver failure, or deteriorating liver functions (bilirubin levels ⬎2.5 mg/100 ml, prothrombin time prolonged by 3 s, albumin ⬎3 g); creatinine levels ⬎2 mg/100 ml Hb levels ⬍10 mg/100 ml, white blood cell count ⬍3,000/mm3 or platelet count ⬍100,000/mm3, or history of varices, ascites or encephalopathy. Also ineligible were patients who exhibited the following: irreversible neurological deficit or active coinfection with hepatitis C, A, or D virus, or HIV; other clinical conditions not related to the primary disease; a history of major psychiatric disturbance; history of serum sickness; acute infectious disease; fever from any cause; or immunosuppressive treatment. Patients were also excluded for showing evidence, whether clinical or laboratory, of any autoimmune disease not related to HBV; allergies to bovine proteins; or pregnancy or lactation. Patients were tested to determine that they were negative for anti–nuclear factor, anti-liver kidney microsomal (anti-LKM), antimitochondrial, and anti–smooth muscle antibodies and that they had normal IgM levels. Recombinant Antigen Preparation and Safety Guidelines BioHepB, recombinant hepatitis B vaccine (BioTechnology General, Rehovot, Israel), which contains three surface antigens of the hepatitis B virus (HBsAg, PreS1, and preS2) was used for induction of anti-HBV immune regulation in the study (24). This vaccine was chosen because it has been shown to improve immunogenicity as compared with other vaccines, and to induce a high level of seroconversion and high antibody titer (24). BioHepB was prepared according to U.S. Food and Drug Administration regulations, and criteria for recombinant HBV antigens used for vaccine therapy. The solution was administered in a liquid form diluted in normal saline solution. Oral Antigen Administration All patients were administered recombinant HBsAg⫹pre S1⫹pre S2 p.o. (BioHep B, BioTechnology General), every other day at the appropriate dose for 20 (n ⫽ 20) or 30 (n ⫽ 22) wk. Patients were followed for 20 wk after completion of the oral administration as described below.

AJG – November, 2003

Dosages The administered doses were calculated on the basis of previous antiviral p.o. immune regulation studies performed in animals as well as human oral tolerance studies with other antigens (21, 25, 26). Three different doses were used: 105 ␮g (n ⫽ 37), 700 ␮g (n ⫽ 3), and 1400 ␮g (n ⫽ 2). Based on preliminary data from animal studies and previous clinical trials of oral tolerance, all three dosages were within the “low dose” p.o. tolerance range. Follow-Up Parameters All patients were followed for clinical, biochemical, virological, and histological parameters, as described below. CLINICAL. Patients were followed biweekly during the oral administration period, and every 4 wk during the follow-up period, by physical examination. BIOCHEMICAL. Serum bilirubin and liver enzymes, kidney function, prothrombin time, and complete blood counts were followed biweekly during the feeding period and every 4 wk during follow-up. HBV VIRAL MARKERS. HBsAg, HBeAg, anti-HBe, anti-HBcIgG, and anti-HBcIgM were assessed every 4 wk throughout the study using standard assays (Abbott Laboratories, North Chicago, IL). Serum HBV DNA levels were quantified using a solution hybridization assay (Abbott Laboratories), with a lower limit detection of 6 ⫻ 106 (2 pg/ml). Serum samples that were negative by this method at screening were tested for the presence of low level HBV DNA by quantitative PCR using the HBV-DNA PCR– quantitative assay (Roche Diagnostics, Systems, Branchburg, NJ). The lower limit of detection of this assay is 100 copies/ml. HBV DNA levels were followed biweekly during the feeding period and every 4 wk during the follow-up period. Results were given as copies/ml. MORPHOLOGICAL. All patients underwent a liver biopsy within 3 months before treatment. A repeat biopsy was performed by two blinded experienced pathologists 2– 6 wk after completion of the feeding during the follow-up period. The intrahepatic inflammatory score was evaluated using the standard Histological Activity Index (HAI) as described by Knodell et al. (27). An HBV score was determined for HBsAg and for HBcAg. Degrees of immunoperoxidase staining for HBsAg and HBcAg were evaluated as previously described (28). Staging for both HBV antigens was scored as follows: 0 ⫽ no staging; ⫹1 ⫽ ⬍5% of cells stained; ⫹2 ⫽ 5–25% of cells stained; and ⫹3 ⫽ ⬎25% of cells stained. Assessment of Immune Modulation of the Anti-HBV Immune Response Immune modulation of the anti-HBV cellular and humoral immune responses was evaluated by IFN␥ and IL10 ELISPOT assays, HBV specific cytotoxicity and proliferation T cell assays, and serum cytokines by a

Oral Tolerance for Hepatitis B Virus

2507

Th1/Th2 reverse transcription–polymerase chain reaction (RT-PCR) test. IFN␥ and IL10 Elispot Assays HBV envelope proteins and IFN␥ spot forming cells (SFC) were determined using an HBsAg specific ELISPOT assay (Mabtech, Nacka, Sweden) as described with the following modifications (29). In brief, 96-well filtration plates coated with high protein binding hydrophobic PVDF membrane (polyvinylidene disulfide) were used (Millipore, Bedford, MA). Plates were coated using 1-D1K anti-IFN␥ coating antibody (15 mg/ml; Mabtech) for 24 h at 4°C. Peripheral blood mononuclear cells (PBMC) were isolated by Ficoll gradient separation from 20-ml blood samples collected in acid citrate dextrose tubes and processed within 1 h. PBMC were washed twice in RPMI 1640 with 10% fetal calf serum (FCS). Cells were cultured in 96-well plates (1 ⫻ 105 cells/well) with RPMI 1640 and 10% FCS. Three triplicates were prepared with HBV envelope proteins (recombinant HBsAg⫹pre S1⫹pre S2, BioHep B, BioTechnology General, 50 ␮g/ml), phytohemagglutinin (2.5 ␮g/ml), or RPMI without antigen. Plates were incubated for 48 h at 37°C and 5% CO2. After washing, dilute biotinylated antibodies (7B6 –1-biotin, Mabtech) were added in filtered PBS with 0.5% FCS to 1 ␮g/ml, in a total volume of 100 ␮l/well. Plates were incubated for 3 h at room temperature. After washing, 100 ml of streptavidin-alkaline phosphatase was added, and plates were incubated for 90 min at room temperature. After washing, a substrate was added (BCIP/NBT, BioRad, Richmond, VA) for 30-min until dark red-purple spots emerged. Two independent investigators counted dark spots, which reflected IFN␥-secreting clones, using a dissection microscope. Results are expressed as means of triplicate IFN␥-secreting cells per 105 PBMC, after subtracting the mean spots from wells without viral antigens. Determination of IL10 spot forming cells was performed using a similar method with the following modifications; only 104 PBMC/well were plated. For coating the plates, an anti-IL10 coating antibody was used (9D7, 15 ␮g/ml). A diluted biotinylated antibody was used to develop IL10 secreting spots (12G8-biotin, Mabtech). T Cell Proliferation Assays Sample collection, preparation and testing was conducted, as previously reported, to evaluate CTL activity in the peripheral blood of subjects at various points in the study (29). Peripheral blood mononuclear cells (PBMC) were isolated by Ficoll gradient separation, and blood samples (60 –90 ml) collected in acid citrate dextrose tubes and processed within 24 h. PBMC were grown in triplicates of 105 cells in RPMI with 10% FCS and were stimulated in vitro using 0.001 ␮g/ml, 0.1 ␮g/ml, or 1 ␮g/ml of BioHepB (BioTechnology General). Then, 5–7 days later, methyl-H3 thymidine was added to HBsAgpulsed T cells (1 ␮Ci/ml, Amersham Pharmacia, Biotech, London, UK). T cell cultures were harvested after 18 h.

2508

Safadi et al.

Controls were incubated in the presence of 2.5 ␮g/ml phytohemagglutinin or in medium alone. Data were given as mean stimulation indices of triplicates and SEM, calculated from the ratios of incorporated radioactivities of T cell cultures, expressed as counts per minute (cpm), in the presence or absence of antigen. Cytotoxic T Cell 51CR Release Assay Cytolytic activity of restimulated T cell cultures was measured using a 51Cr release assay, as previously described, with the following modifications (8, 30). Target cells were prepared for each patient by infecting fresh peripheral blood lymphocytes. Target cells were available for eight patients in the trial. EBV-infected target cells were incubated overnight in the presence of 1 mg BioHepB and 200 ␮Ci 51Cr (Amersham Pharmacia). Targeted cells were washed and plated at 3,000 cells/well. Effector cells, peripheral blood lymphocytes in serial dilutions were cocultured for 5 h at 37°C with 51Crlabeled target cells that had been sensitized by loading of HBsAg (15,000/well). Release of 51Cr into the supernatant was quantified as a measurement of cytolytic activity. After washing, and centrifugation (54 ⫻ g), supernatants were adsorbed by cellulose acetate cartridge and counted with a gamma counter. The percent specific lysis was determined using the standard formula: % cytotoxicity ⫽ 100 ⫻ (experimental 51Cr release ⫺ spontaneous release)/(maximal 51Cr release ⫺ spontaneous release). Spontaneous release of all targets never exceeded 20%. Total release was measured in the presence of triton (1%). Cytokine Tests Lymphocyte cytokine expression was assessed by RT-PCR performed on peripheral lymphocytes as previously described (21), with the following modification. RNA extraction was performed from PBMC, using the EZ-RNA kit (Biological Industries, Beit Haemek, Israel). Cells were homogenized in 0.5 ml of denaturating solution for 5 min. Extraction solution was added to the homogenate and centrifuged. The aqueous phase was separated and 0.5 ml of isopropanol was added for 10 min and centrifuged. The RNA pellet was washed in 75% ethanol and dried and the pellet was dissolved in water. The cDNA preparation was made using a reaction mixture containing RNA 0.2 ␮g oligo dT and 10 u reverse transcriptase (Promega, Madison, WI, USA). PCR amplification was performed using cDNA for IL2, IFN␥, and IL4, IL10, and amplified using Cytoxpress Quantitative PCR Detection Kit for human Th1/Th2 cytokines (BioSource International, Camarillo, CA). PCR products were analyzed using 2% agarose gel. IFN␥ and IL4 Serum Levels IL4 and IFN␥ levels were measured by a “sandwich” ELISA, using Genzyme Diagnostics kits (Genzyme Diagnostics, Cambridge, MA) according to the manufacturer’s instructions.

AJG – Vol. 98, No. 11, 2003

Flow Cytometry Analysis for Determination of the Effect of Oral Immune Regulation on Natural Killer T Lymphocytes in Peripheral Blood Blood samples were collected every 2 wk throughout the study period. Immediately after lymphocyte isolation, duplicates of 2–5⫻104 cells/500 ␮l PBS were put into Falcon 2052 tubes incubated with 4 ml of 1% bovine serum albumin (BSA) for 10 min, and centrifuged at 1400 rpm for 5 min. Cells were resuspended in 10 ␮l of FCS with 1:20 FITC-antihuman CD3, CD56, and CD16, antibodies (Pharmingen R&D, San Diego, CA), and mixed every 10 min for 30 min. Cells were washed twice in 1% BSA with 0.5 ml of paraformaldehyde 1% added and kept in 4°C until reading. For the control group, only 5 ␮l of BSA 1% was added. Analytical cell sorting was performed on 1⫻104 cells from each group with a fluorescence-activated cell sorter (FACSTAR plus, Becton Dickinson, Franklin Lanes, NJ, USA). Only live cells were counted, and background fluorescence from non– antibody-treated lymphocytes was deducted from levels obtained. Primary Endpoints Response was determined if any of following occurred. A decrease in viral load from screening level by ⱖ2 logs for two of the following measurements: a decrease in biopsy HAI score by ⱖ2 points; a decrease in biopsy viral score by ⱖ1 point; a decrease in liver enzymes to normal or by 30% from baseline; or seroconversion of eAg⫹ to eAb⫺ or loss of eAg. Secondary Endpoints Secondary endpoints were as follows: an increase in T cell proliferation by at least 0.5 point in the proliferation index; an increase in cytotoxicity by at least 10%; an increase in ELISPOT IFN assay by at least 1 colony forming unit; a decrease in ELISPOT IL10 by at least 1 colony forming units; positive RT PCR for IL2, IFN␥, IL4, or IL10 in serum; and an increase in serum IFN␥ and IL4 levels by 5 pg/ml. Statistical Analysis Statistical differences were calculated by the Student’s t test or the Mann-Whitney test, as appropriate. All p values were two-tailed.

RESULTS Patient Characteristics A total of 42 patients, 27 male and 15 female, with a mean age of 56.5 yr (range 22–72 yr), were enrolled in the trial. All patients tested positive for HBsAg and HBV DNA. All had evidence of active inflammation on liver biopsy and a positive HBV histological score. Of the patients, 38 had chronic active hepatitis and four patients showed HBV reinfection of a liver allograft after liver transplantation. In all, 20 patients had elevated liver enzymes on screening and

AJG – November, 2003

Oral Tolerance for Hepatitis B Virus

2509

Figure 2. Oral immune regulation toward HBV proteins suppressed HBV DNA serum levels. A decrease in HBV DNA levels by ⱖ2 logs occurred in 15/42 patients during the feeding period (36%). Results are shown for 15 responders before p.o. feeding (filled bars) and at maximal point of decrease (open bars). Week of maximal decrease is shown at the head of each bar.

Figure 1. Effect of induction of p.o. immune regulation toward HBV proteins on ALT and HBV-DNA serum levels in two patients with chronic active hepatitis. Normalization of liver enzymes and a decrease in viral load were noted during the feeding period (A). This patient seroconverted from HBeAg to anti-HBe during wk 18. Normalization of liver enzymes and a significant decrease in viral load followed a transient increase in enzymes and HBV DNA levels (B).

22 had normal liver enzymes despite viremia and active inflammation on biopsy. All patients were followed for clinical, biochemical, and viral measures biweekly during the feeding period (20 –30 wk) and every 4 wk during follow-up (additional 20 wk). Oral Immune Regulation Toward HBV Proteins Alleviated Liver Inflammation Normalization of liver enzymes occurred in 16 of 20 patients with elevated liver enzymes (80%). Two patterns of response were noted in responders. In 11/16 patients, a gradual decrease in enzymes beginning 2– 4 wk after the feeding was noted (Fig. 1A). In five of 16 additional patients, normalization of liver enzymes followed a transient increase in liver enzymes (Fig. 1B). In these patients, enzymes increased up to 8-fold in parallel with an increase in serum HBV DNA levels, followed by normalization of

enzymes and a marked decrease in viral load. In 13 of the responders (81.2%), enzymes remained normal throughout the follow-up period. All four patients with liver transplants were nonresponders. Oral Immune Regulation Toward HBV Proteins Suppressed HBV DNA Serum Levels Decrease in HBV DNA levels by two or more logs for two consecutive measurements occurred in 15/42 patients during the feeding period (35.7%, Fig. 2). In nine of the responders, HBV DNA levels were negative by hybridization (HBV DNA levels ⬍ 105 copies/ml). In nine patients, a gradual decrease in HBV DNA levels was observed beginning from 2– 8 wk after feeding (Fig. 1A). In six patients, a marked decrease in HBV DNA levels followed a transient increase in viral load (Fig. 1B). This pattern paralleled the paradigm observed in liver enzymes in these patients. The median time to response was 6 wk (range 2–21). In eight responders, HBV DNA levels remained low throughout the follow-up period (53.3%). Effect of Oral Immune Regulation Toward HBV Proteins on HBeAg HBeAg was positive in 19 patients before the trial. Five HBeAg positive patients turned HBeAg negative (26%).

2510

Safadi et al.

AJG – Vol. 98, No. 11, 2003

Four of them developed anti-HBe antibodies. HBsAg and HBcIgM status did not change in any of the patients throughout the study period. Oral Immune Regulation Toward HBV Proteins Improves Liver Histology and HBsAg and HBcAg Histology Scores All patients underwent a liver biopsy within 2 months before treatment. A repeat biopsy was performed in 40 patients 1– 4 wk after completion of feeding during the follow-up period. All biopsy samples were evaluated for histological scores, 39 for HBsAg scores, and 28 for HBcAg scores. Improvement in the HAI score with a decrease of ⱖ2 points occurred in 12/40 patients (30%). Improvement in liver HBsAg score by ⬎1 point was achieved in 16/39 (41.0%) patients, and improvement in liver HBcAg score by ⬎1 point was achieved in 16/28 (57.1%) patients. Oral Immune Regulation Toward HBV Proteins Augments the Anti-HBV T-Cell–Mediated Effective Immune Response Induction of p.o. immune regulation enhanced the effective antiviral immune response in all responders. Immune modulation of the anti-HBV cellular and humoral immune responses was evaluated by IFN␥ and IL10 ELISPOT assays, HBV specific cytotoxicity, proliferation T cell assays, and serum Th1/Th2 cytokine RT-PCR assay. HBV Specific IFN␥ and IL10 ELISPOT Assays IFN␥ and IL10 spot-forming cells (SFC) were determined using an HBsAg specific ELISPOT assay. Results were expressed as means of triplicates of IFN␥ or IL10 secreting cells per 105 PBMC. A favorable augmentation of the number of IFN␥SFC was observed in 17/27 patients (62.9%, Fig. 3). The median time to response was 9 wk (range 4 –24 wk). In four responders, this increase remained throughout the follow-up period (23.5%). In 13/17 (76.4%) responders, IFN␥-SFC numbers decreased upon termination of treatment. A decrease in number of IL10-SFC was observed in 13/27 patients (48.1%, Fig. 4). The median time to response was 9 wk (range 6 –18 wk). In all responders, IL10 SFC numbers remained low upon termination of treatment. These results suggest that p.o. immune regulation toward HBV antigens augments the antiviral immune response through an increase in the Th1 and a decrease in the Th2 responses. HBV Specific T Cell Proliferation Assays An increase in HBV T cell stimulation index was observed in 21/27 (78.0%) of the treated patients (Fig. 5). Stimulation indices of triplicates were calculated from the ratios of incorporated radioactivity of T cell cultures expressed as counts per minute (cpm), in the presence or absence of antigen. The median time to response was 15 wk (range 3–30 wk). In six of 21 (29%) responders, HBV specific T cell stimulation index decreased upon termination of treat-

Figure 3. Effect of p.o. immune regulation toward HBV envelope proteins on HBV specific IFN␥ ELISPOT assay. IFN␥ spot forming cells (SFC) were determined using an HBsAg-specific ELISPOT assay. A favorable augmentation of the number of IFN␥-SFC was observed in 17/27 patients (62.9%). Results are expressed as means of triplicates of IFN␥ secreting cells per 105 PBMC, after subtraction of mean spots from wells without viral antigens. Results are shown for 17 responders before p.o. feeding (filled bars) and at maximal point of increase (open bars). Week of maximal increase is shown at the head of each bar.

ment. These results suggest an increase in the HBV-associated T cell response after p.o. immune regulation toward the viral envelope proteins. HBV Specific Cytotoxic T Cell 51CR Release Assay Cytolytic activity of restimulated T cell cultures was measured using a 51Cr release assay in 27 of the treated patients. Target cells were obtained for eight of these patients. In six of these eight patients, an increase in HBV specific T cell cytotoxicity was observed (75%, Fig. 6). The percentage of HBV specific T cell cytotoxicity increased from 8% to 60%, thus suggesting that p.o. immune regulation effectively augmented the antiviral T cell immunity. The median time to response was 13.5 wk (range 9 – 42 wk). Effect of Induction of Oral Immune Regulation Toward HBV Proteins on Cytokine Tests Th1/Th2 lymphocytes cytokine secretion was assessed by RT-PCR performed on peripheral lymphocytes. All PBMC were tested for IFN␥, IL2 (Th1), IL4, and IL10 (Th2). Expression of IFN␥ was detected in 20/22 patients who were negative on screening (90.9%). Expression of IL10 was observed in only one of 22 patients tested. No significant detection of IL2 and IL4 were observed. Serum IFN␥ increased in 11 of 22 patients (50%) who were negative on

AJG – November, 2003

Figure 4. Effect of p.o. immune regulation toward HBV envelope proteins on HBV specific IL10 ELISPOT assay. IL10 spot forming cells (SFC) were determined using an HbsAg specific ELISPOT assay. A favorable decrease in number of IL10SFC was observed in 13/27 patients (48.1%). Results are expressed as means of triplicates of IL10 secreting cells per 105 PBMC, after subtraction of mean spots from wells without viral antigens. Results are shown for 13 responders before p.o. feeding (filled bars) and at maximal point of decrease (open bars). Week of maximal decrease is shown at the head of each bar.

screening. Serum IL4 increased in five of 10 patients (50%) who tested negative on screening. Effect of Induction of Oral Immune Regulation Toward HBV Proteins on NKT Lymphocytes Numbers The number of peripheral NKT lymphocytes increased by 2-fold or more in all 27 patients tested (Fig. 7). The median time to response was 21 wk (range 15–30 wk). These results suggest that this subtype of T cells may play a role in p.o. immune modulation of the anti-HBV immune response. Correlation Between Primary and Secondary Endpoints Secondary endpoints were tested in 27 patients. The correlation between the response in primary and secondary endpoints is summarized in Table 1. No major differences were observed in T cell proliferation, and IFN␥ SFC between patients with a decrease in HBV DNA levels and those who did not respond. However, IL10 SFC decreased in 56% of

Oral Tolerance for Hepatitis B Virus

2511

Figure 5. Effect of p.o. immune regulation toward HBV envelope proteins on HBV specific T cell proliferation assay. Increase in HBV specific T cell stimulation index was observed in 21/27 (78.0%) of treated patients. Stimulation indices (SI) of triplicates were calculated from the ratios of incorporated radioactivities of T cell cultures expressed as counts per minute (cpm) in the presence or absence of antigen. Results are shown for 21 responders before p.o. feeding (filled bars) and at maximal point of increase (open bars). Week of maximal increase is shown at the head of each bar.

responders compared with only 28% of nonresponders (p ⬍ 0.005). Safety Treatment was well tolerated by all patients. All patients completed the study and no adverse effects were observed with regard to any of the organs that were followed.

DISCUSSION The results of the present study show that induction of p.o. immune regulation toward HBV-antigens through p.o. administration of HBV envelope proteins alleviated the immune-mediated liver damage while enhancing the effective antiviral immunity. A significant decrease in viral load was observed in 35% of patients. HBsAg/HBcAg scores on liver biopsy improved in 41% and 57%, respectively, and the histological necroinflammatory score in 30%. Five of 19 HBeAg positive patients became HBeAg negative. Of the patients with elevated liver enzymes, 80% showed a favorable biochemical response. No major adverse events were noted. Oral immune regulation toward HBV envelope proteins induced a favorable increase in HBV specific T cell proliferation, cytotoxicity, and IFN␥ secreting clones, along with a significant decrease in the anti HBV IL10 secreting T cell clones.

2512

Safadi et al.

Figure 6. Effect of p.o. immune regulation toward HBV envelope proteins on HBV-specific -cytotoxic T cell 51Cr release assay. Cytolytic activity of restimulated T cell cultures was measured using a 51Cr release assay in 12 of the treated patients. In six of eight patients an increase in the HBV specific T cell cytotoxicity was observed. Percent specific lysis was determined using the standard formula: % cytotoxicity ⫽ 100 ⫻ (experimental 51Cr release ⫺ spontaneous release)/(maximal 51Cr release ⫺ spontaneous release). Results are shown for six responders before p.o. feeding (filled bars) and at maximal point of increase (open bars). Week of maximal increase is shown at the head of each bar.

In the context of the current understanding of immunological mechanisms involved in HBV-mediated liver injury, it is likely that chronic HBV-associated inflammation of the liver results from an imbalance between two types of immune responses. The first is an effective antiviral immune response that clears the virus and suppresses immune-mediated liver damage. The second is an ineffective immune response that enhances viral replication or liver damage, or both (6). Both quantitative and qualitative differences may be held accountable for various types of responses in different patients. Acute HBV infection is associated with the development of a strong polyclonal cytotoxic T lymphocyte response that suppresses viral growth (4). Noncytopathic antiviral mechanisms contribute to viral clearance (10). In contrast, HBV associated-chronic liver damage is mediated by a deviant immune response toward viral or virally associated antigens (3). Chronic HBV-infected patients fail to clear the virus, and a defective immune response prompts liver injury. Several possible causes for immune variance in the anti-HBV response have been suggested, which include the following: efficiency of antigen recognition and responsiveness of B or T cells; alteration of Th1/Th2 balance; “exhaustion” of the im-

AJG – Vol. 98, No. 11, 2003

Figure 7. Effect of induction of p.o. immune regulation toward HBV proteins on NKT lymphocytes numbers. The number of peripheral NKT lymphocytes increased by 2-fold or more in all 27 patients tested. Results are shown for 27 patients before p.o. feeding (filled bars) and at maximal point of increase (open bars). Week of maximal increase is shown at the head of each bar.

mune system; host HLA background; efficiency of viral antigen processing by professional antigen presenting cells; and magnitude of viral load or viral mutations (2–5). Several lines of evidence support the notion that tolerance toward HBV may alleviate the severity of hepatitis. At the cellular level, studies in transgenic mice have shown that HBV viral antigens (e.g., HBeAg in the neonatal period, or HBcAg) may act as tolerogens, leading to induction of an antigen specific suppressor T cell population (31). Human neonates, in whom HBV specific tolerance occurs because of deletion of HBV-recognizing specific T cell clones, clear the virus very slowly (11, 32). After perinatal infection, 95% of subjects become chronic carriers compared with only 5% of adults. Perinatally infected subjects may not develop severe HBV-mediated liver damage (32, 33). Immunosuppressed hosts such as Table 1. Correlation Between Primary and Secondary Endpoints

Primary Endpoint Decreased HBV DNA levels (n ⫽ 9) No decrease in HBV DNA levels (n ⫽ 18) Decrease in HAI score in liver biopsy (n ⫽ 7) No change in HAI score (n ⫽ 18)

Increased T Cell Increased Increased Proliferation IFN␥ SFC IL10 SFC % (n) % (n) % (n) 78 (7)

67 (6)

56 (5)

78 (14)

56 (10)

28 (5)

71 (5) 89 (16)

100 (7) 78 (14)

29 (2) 68 (11)

AJG – November, 2003

patients treated with chemotherapy patients tend to have prolonged viral persistence after HBV infection but may have milder liver injury (2). Treatment with an anti–T cell agent, Campath-1, was reported to alleviate viral hepatitis (34). Some immunosuppressed patients with HBV recurrence after liver transplantation may develop a state of “immune tolerance” to the virus. In these patients, despite recurrence of the viral infection, no liver damage is observed (35). Oral administration of antigens may induce antigen specific immune hypo- or hyperresponsiveness (14). Whether tolerance or immunogenicity is induced is determined by a balance between diffent subtypes of T cells in the microenvironment of the responding lymph nodes. This balance depends on factors such as the proximity of different antigens; changes in number, variety, and magnitude of antigen mimicry among different epitopes exposed; nature of the antigen used; and route of antigen entry (15). Augmentation of the immune response toward p.o. administered antigens, reactivities to multiple autoantigens, and exacerbation of autoimmune disorders have been described (16, 36). In contrast, p.o. tolerance induction can prevent or alleviate immune-mediated disorders in animals and humans (26, 37–39). Tolerization toward adenoviral and HBV antigens can abrogate both humoral and cellular antiviral immune responses (23, 40). Oral administration of low-dose HBV envelope proteins (BioHepB) induced peripheral humoral immune tolerance toward HBV-epitopes in naive and preimmunized animals (23). The results of the present study show that HBV specific T cell immune modulation can be elicited through p.o. administration of HBV envelope proteins to chronically infected individuals. Application of this new concept of p.o. immune regulation toward the virus alleviated the disease, while leaving the general immunological defense of the recipient intact. The antiviral effect achieved in these patients was mediated by augmentation of an anti-HBV Th1 (IFN␥)–mediated response along with a decrease in the Th2 (IL10) response. These data support the concept that viral load is associated with a defective immune response, and that HBV-replication is dependent on the antiviral immune status of the patient. The salutary effects of IFN␣, lamivudine, vaccination therapy, cytokine administration, and adoptive transfer of immunity in chronic HBV patients result from both antiviral and immune modulatory activities (41). A response to therapy involves a quantitative increase in IFN-mediated anti-HBV immune response or a qualitative change in the Th1/Th2 balance (42, 43). Two explanations may elucidate the results of the present study. The first is induction of immune tolerance. Oral immune regulation toward HBV proteins may have altered the immune deviation removing a deleterious T cell population (IL10 producing), uncovering a more efficacious response (antiviral IFN␥ producing). The second possible explanation is induction of immunity. Oral immune regulation may have enhanced the effect of a beneficial subset of

Oral Tolerance for Hepatitis B Virus

2513

T cells toward the antigens that were administered. It is also possible that downregulation of viral growth occurred through augmentation of an effective antiviral immunity or through suppression of a proviral response, or both. Similar approaches were recently described toward Schistosoma mansoni infection, in which the immune response to parasite eggs in the intestine and liver is responsible for the disease (15). Inherent in this concept is the understanding that pathology is not essential for the development of a protective response. As it is not always possible to separate pathology from protection, it might not always be possible to determine the role of different antigens or subsets of T cells in the induction of each response. Oral immune regulation was expected to lead to induction of memory suppressor or antiviral cells. However, the beneficial effect in some of the patients was dependent on continuous antigen administration. After cessation of feeding, HBV DNA levels remained low in only one half of the responders. The antiviral T cell response (IFN␥ ELISPOT and T cell proliferation studies) was lost in 30 –70% of responders. Interestingly, the IL10 SPC number remained low throughout the follow-up period in all responders. It is possible that correction of the anti-HBV immunological imbalance was partial, leading to an enhanced effect on part of T cell subtypes rather than a clearance or irreversible suppression of “unwanted” T cells. NKT lymphocytes are a subgroup of lymphocytes that are characterized as CD4⫹ or CD4 –CD8⫺ and CD16⫹; they express ␣␤ TCRint and share surface molecules with NK cells (44). Upon stimulation they produce significant quantities of IL4 and IFN␥ and exhibit enhanced cytolytic activity (44). We and others have recently shown that this subset of lymphocytes play a role in p.o. immune regulation (44 – 46). The results of the present study showed a significant increase in the number of these cells in the peripheral blood of all patients studied. These results suggest that NKT lymphocytes are involved in immune regulation of the antiHBV immune response. Recently, NKT cells activated by ␣-galactosylceramide were shown to inhibit HBV replication in transgenic mice. This effect was mediated by antiviral cytokines directly produced by activated NKT cells or by other cytokine-producing inflammatory cells that are recruited into the liver (47). In conclusion, induction of p.o. immune regulation toward viral proteins was found to be safe and effective for amelioration of immune-mediated hepatitis in patients with chronic HBV infection. This effect was associated with augmentation of the antiviral immune response and with a change in the Th1/Th2 immune balance.

ACKNOWLEDGMENTS This study was supported by grants from ENZO Biochem, the Hadassit-Horwitz foundation, Israel Science Founda-

2514

Safadi et al.

tion, Israel Cancer Research Fund, and Roaman-Epstein Liver Research Foundation (to Y.I.). Reprint requests and correspondence: Yaron Ilan, M.D., Liver Unit, Department of Medicine, Hadassah-Hebrew University Medical Center, P.O.B 12000, Jerusalem, Israel IL-91120. Received Dec. 30, 2002; accepted Apr. 4, 2003.

REFERENCES 1. Margolis H, Alter M, Hadler S. Hepatitis B: Evolving epidemiology and implications for control. Semin Liver Dis 1991; 11:84 –92. 2. Chisari FV, Ferrari C. Hepatitis B virus immunopathogenesis. Annu Rev Immunol 1995;13:29 –45. 3. Chisari FV. Viruses, immunity and cancer: Lessons from hepatitis B. Am J Pathol 2000;154:1118 –32. 4. Rehermann B. Intrahepatic T cells in hepatitis B: Viral controls versus liver cell injury. J Exp Med 2000;191:1263–8. 5. Koziel MJ. Cytokines in viral hepatitis. Semin Viral Hepatitis 1999;19:157–69. 6. Ilan Y, Roy Chowdhury J. Induction of oral tolerance towards hepatitis B virus: Can we eat the disease and live with the virus? J Med Hypotheses 1999;52:505–9. 7. Maini MK, Boni C, Lee CK, 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. 8. Rehermann B, Lau D, Hoofnagle JH, Chisari FV. Cytotoxic T lymphocyte responsiveness after resolution of chronic hepatitis B virus infection. J Clin Invest 1996;7:1655–65. 9. Guiotti LG, Rochford R, Chung J, et al. Viral clearance without destruction of infected cell during acute HBV infection. Science 1999;284:825–9. 10. Heise T, Guidotti LG, Cavanaugh VJ, Chisari FV. Hepatitis B virus RNA-binding protiens associated with cytokine-induced clearance of viral RNA from the liver of transgenic mice. J Virol 1999;73:474 –81. 11. Tang J, Hsu HY, Lin HH, et al. Hepatitis B surface antigenemia at birth: A long term follow-up study. J Pediatr 1998; 133:374 –7. 12. Marco VT, Iacono OL, Camma C, et al. The long term course of chronic hepatitis B. Hepatology 1999;30:257–64. 13. Liaw Y, Tai D, Chu C, Chem T. The development of cirrhosis in patients with chronic type B hepatitis: A prospective study. Hepatology 1988;8:493–9. 14. Weiner HL. Oral tolerance: Immune mechanisms and treatment of autoimmune diseases. Immunol Today 1997;18:335– 43. 15. McSorley SJ, Garside P. Vaccination by inducing oral tolerance. Immunol Today 1999;20:555–60. 16. Blanas E, Carbone FR, Allison J, et al. Induction of autoimmune diabetes by oral administration of autoantigen. Science 1996;274:1707–9. 17. Takahashi M, Ilan Y, Sengupta K, et al. Induction of tolerance to recombinant adenoviruses by injection into newborn rats: Long term amelioration of hyperbilirubinemia in Gunn rats. J Biolol Chem 1996;271:26536 –42. 18. Ilan Y, Attavar P, Takahashi M, et al. Induction of central tolerance by intrathymic inoculation of adenoviral antigens into the host thymus permits long term gene therapy in Gunn rats. J Clin Invest 1996;98:2640 –7. 19. Ilan Y, Jona VK, Sengupta K, et al. Transient immunosuppression with FK506 permits long term expression of therapeutic genes introduced into the liver using recombinant adenoviruses. Hepatology 1997;26:949 –56.

AJG – Vol. 98, No. 11, 2003

20. Ilan Y, Droguett G, Roy Chowdhury N, et al. Insertion of the adenoviral E3 region into a recombinant viral vector prevents antiviral humoral and cellular immune responses and permits long term gene expression. Proc Nat Acad Sci USA 1997;94: 2587–92. 21. Ilan Y, Prakash R, Davidson A, et al. Oral tolerization to adenoviral antigens permits long term gene expression using recombinant adenoviral vectors. J Clin Invest 1997;99:1098 – 106. 22. Ilan Y, Sauter B, Roy Chowdhury N, et al. Oral tolerization to adenoviral proteins permits repeated adenovirus-mediated gene therapy in rats with pre-existing immunity to adenovirus. Hepatology 1998;27:1368 –76. 23. Gotsman I, Beinart R, Alper R, et al. 2000. Induction of oral tolerance towards hepatitis B envelope antigens in a murine model. Antiviral Res 2000;48:17–26. 24. Shouval D, Ilan Y, Adler R, et al. Improved immunogenicity in mice of mammalian cell derived recombinant hepatitis B vaccine containing pre-S1 and pre-S2 antigens, compared to conventional yeast derived vaccines. Vaccine 1994;12: 1453–9. 25. Ilan Y, Weksler-Zangen S, Ben-Horin S, et al. Treatment of experimental colitis by oral tolerance induction: A central role for suppressor lymphocytes. Am J Gastroenterol 2000;95: 966 –73. 26. Weiner HL, Mackin GA, Matsui M, et al. Double-blind pilot trial of oral tolerization with myelin antigen in multiple sclerosis. Science 1993;261:1321–4. 27. Knodell RG, Ishak KG, Black WC, et al. Formulation and application of a numerical scoring system for assessing histological activity in asymptomatic chronic active hepatitis. Hepatology 1981;1:431–5. 28. Perrilo R, Rakela J, Dienstag J, et al. Multicenter study of lamivudine therapy for hepatitis B after liver transplantation. Hepatology 1999;29:1581–6. 29. Rahman F, Dahmen A, Herzog-Hauff S, et al. Cellular and humoral immune responses induced by intradermal or intramuscular vaccination with the major hepatitis B surface antigen. Hepatology 2000;31:521–7. 30. Jung MC, Hartmann B, Gerlach JT, et al. Virus-specific lymphokine production differs quantitatively but not qualitatively in acute and chronic hepatitis B infection. Virology 1999;261: 165–72. 31. Milich DR, Schodel F, Peterson DL, et al. Characterization of self-reactive T cells that evade tolerance in hepatitis B e antigen transgenic mice. Eur J Immunol 1995;25:1663–72. 32. Kurose K, Akbar SM, Yamamoto K, Onji M. Production of antibody to hepatitis B surface antigen (anti-HBs) by murine hepatitis B virus carriers: Neonatal tolerance versus antigen presentation by dendritic cells. Immunology 1997;92:494 –500. 33. Bozkaya H, Ulus SA, Brick A, Lok ASF. High degree of conservation in the hepatitis B virus core gene during the immune tolerant phase in perinatally acquired chronic hepatitis B virus infection. J Hepatology 1997;26:508 –16. 34. Nagler A, Ilan Y, Varadi G, et al. In vivo Campath-1 followed by T cell depleted BMT: A potential new mode of therapy for hepatitis associated severe aplastic anemia. Bone Marrow Transpl 1996;18:475–8. 35. Gish RG, Keeffee EB, Lim J, et al. Survival after liver transplantation for chronic hepatitis B using reduced immunosuppression. J Hepatol 1995;22:257–62. 36. Bennett SRM, Carbone FR, Karamalis F, et al. Induction of a CD8⫹ cytotoxic T lymphocyte response by cross-priming requires cognate CD4⫹T cell help. J Exp Med 1997;186:65– 70. 37. Nagler A, Pines M, Abadi U, et al. Oral tolerization amelio-

AJG – November, 2003

38.

39.

40. 41. 42.

rates liver disorders associated with chronic graft versus host disease in mice. Hepatology 2000;31:641–8. Ilan Y, Gotsman I, Pines M, et al. Induction of oral tolerance in splenocyte recipients towards pre-transplant antigens ameliorates chronic graft versus host disease in a murine model. Blood 2000;95:3613–9. Trop S, Samsonov D, Gotsman I, et al. Liver associated lymphocytes expressing NK1.1 are essential for oral tolerance induction in a murine model. Hepatology 1999;29: 746 –55. Ilan Y, Saito H, Thummala NR, et al. Adenovirus-mediated gene therapy of liver diseases. Semin Liver Dis 1999;19:49 – 59. Hoofnagle JH, Di Bisceglie AM. The treatment of chronic viral hepatitis. N Eng J Med 1997;325:347–56. Ilan Y, Nagler A, Adler R, et al. Ablation of persistent hepatitis B by allogeneic bone marrow transplantation from an HBV immune donor. Gastroenterology 1993;104:1818 –21.

Oral Tolerance for Hepatitis B Virus

2515

43. Guidotti LG, Chisari FV. Cytokine mediated control of viral infections. Virology 2000;273:221–7. 44. Godfrey DJ, Hammond KJ, Poulon LD, et al. NKT cells: Facts, functions and fallacies. Imunol Today 2000;21:573–83. 45. Doherty DG, Norris S, Madrigal-Estebas L, et al. The human liver contains multiple population of NK cells, T cells, and CD3⫹CD56⫹ natural T cells with distict cytotoxic activities and Th1, Th2, and Th0 cytokine secretion patterns. J Immunol 1999;163:2314 –21. 46. Samsonov D, Trop S, Alper R, et al. Enhancement of immune tolerance via induction of NK1.1 positive liver-associatedlymphocytes under immunosuppressive conditions in a murine model. J Hepatology 2000;32:812–20. 47. Kakimi K, Lane TE, Chisari FV, Guidotti LG. Cutting edge: Inhibition of hepatitis B virus replication by activated NK T cells does not require inflammatory cell recruitment to the liver. J Immunol 2001;167:6701–5.