Prevention of hepatitis C virus infection using a broad cross-neutralizing monoclonal antibody (AR4A) and epigallocatechin gallate

The management of chronic hepatitis C virus (HCV) infection continues to rapidly evolve. Novel direct-acting antivirals (DAAs) capable of achieving universally high cure rates carry much promise.1 Equitable access to these agents represents a global challenge with a large proportion of those chronically infected worldwide likely to have very limited access for the considerable future.2, 3

DAAs will also feature prominently in the global resolve to eradicate HCV infection. HCV prevention, however, will remain fundamental to realizing this aspiration.3, 4 Primary prevention via vaccination has been undermined by the lack of a convenient animal model and the challenge presented by the diversity among HCV genotypes (GTs) and quasispecies.4-6 Likewise, capitalizing on the unique opportunity afforded by liver transplant to prevent HCV reinfection has been unsuccessful to date.7-9

HCV-associated liver disease (cirrhosis or hepatocellular carcinoma) is long established as the leading indication for liver transplantation worldwide.10 In those undergoing this lifesaving intervention, HCV reinfection is presently universal and drives inferior posttransplant outcomes.10-12 A window exists peritransplantation to optimize posttransplant outcomes by preventing allograft HCV infection. A precedent exists within liver transplantation whereby hepatitis B virus reinfection can be prevented after transplant.13, 14 To date, replicating this in relation to HCV has proven elusive.15

Strategies to employ existing therapies (pegylated interferon/ribavirin with or without first generation protease inhibitors) in a preemptive role before transplant or soon after transplant have been unsuccessful. The intolerability and contraindications of these agents in highly complex and medically unstable patients severely limits the applicability of such approaches.7, 8 A randomized trial using hepatitis C immune globulin concluded that this was a safe and tolerable agent in liver transplant recipients, but no beneficial effect on the rate of HCV recurrence was observed.9

Very recently, limited data in select populations employing sofosbuvir and ribavirin pretransplantation have proven the feasibility of this approach.16 Generalizing this approach to complex and often compromised patients before and after liver transplant is, however, the subject of ongoing studies.

Many monoclonal antibodies (mAbs) and polyclonal antibodies targeting linear or conformational epitopes within the HCV surface glycoprotein (E1-E2) have been described that affect HCV neutralization in vitro.17-19 E1/E2 interacts with a number of cell-surface molecules to mediate HCV entry, eg, C-type lectins, CD81, Scavenger receptor-B1.20 The in vitro performance of HCV-neutralizing antibody has been more variable.21 Law et al.22 characterized a number of antigenic regions on E2 and identified numerous human mAbs with cross-neutralizing activity. This group has identified further antigenic regions of E1/E2 and generated a human mAb (AR4A) targeting a discontinuous epitope outside the CD81 binding site on HCV E1/E2. This epitope is highly conserved across GTs, and AR4A demonstrates potent, cross-neutralizing activity in vitro.23, 24

Epigallocatechin gallate (EGCG) is the most abundant catechin present in green tea extract. Green tea extracts have a long history of safe human consumption and have many purported benefits.25, 26 Green tea extracts are safe, widely available, and inexpensive.27 Two groups independently published data reporting an in vitro antiviral effect of EGCG against HCV.28, 29 This was primarily attributed to prevention of initial attachment of the viral particle to hepatocytes. EGCG, however, also inhibited the direct cell-cell mode of HCV transmission.28 In vitro evidence of the anti-HCV effect of EGCG is lacking.

Using the HCV cell culture system and the severe combined immunodeficiency/albumin/urokinase plasminogen activator (SCID/uPA) humanized liver mouse model, we examined the anti-HCV effect of AR4A and EGCG alone and in combination. We hypothesized that combining safe, tolerable, and effective HCV therapies acting at different sites of HCV cell entry could reliably protect against HCV challenge in vivo.

MATERIALS AND METHODS

EXPERIMENTAL PROTOCOL

The study outline and protocol employed is illustrated in Fig. 1.

HCV inoculation was administered by intrajugular injection. The inoculum used (50 μL, 1.5 × 10⁷ IU/mL) was patient-derived HCV GT 1a serum. Blood sampling was conducted weekly by drawing 100 μL via tail bleeds for measurement of HCV titers and serum hAAT levels.

Statistical Analysis

Statistical analyses were performed using Stata, version 12.0 (StataCorp LP, College Station, TX) and GraphPad prism, version 6.0 (GraphPad Software, La Jolla, CA) software. Continuous variables were compared between groups using a Mann-Whitney U test. A P value < 0.05 was considered significant. The drug concentration at which HCV infection was inhibited by 50% inhibitory concentration (IC₅₀) was calculated using the following equations:

In the HCV challenge experiments, animals with HCV RNA detectable above the threshold (1000 IU/mL) by polymerase chain reaction at day 7 or thereafter were considered “infected.” Animals not reaching this threshold were “censored” and considered free from HCV infection. A Kaplan-Meier survival curve (“survival free from HCV infection”) was thus generated. Statistical significance between the groups was calculated using a 2-tailed log-rank test.

RESULTS

AR4A Demonstrates Effective HCV Neutralizing Activity In Vitro

The in vitro neutralizing capability of AR4A against HCVcc expressing surface glycoproteins of GTs 1-6 was examined. Focusing on GT 1a (GT used for mouse challenge experiments) and 2a (parent Japanese fulminant hepatitis [JFH] GT), AR4A exhibited superior neutralizing activity against HCV GT 1a when compared with GT 2a, which is consistent with prior reports.23, 37 IC₅₀ estimates of 1.28 μg/mL and 4.37 μg/mL were obtained for GTs 1a and 2a, respectively (Fig. 2C,D).

Additive Reduction in HCV Infection Combining AR4A and EGCG

AR4A and EGCG primarily act early in the HCV life cycle to inhibit HCV cellular attachment and entry. Using the HCVcc system, we studied the anti-HCV activity of both agents in combination.

Using GT 1a HCVcc, AR4A (2 μg/mL) combined with EGCG (10 μg/mL) demonstrated significantly increased inhibition compared to either agent alone; 88.7% inhibition compared to 61% for AR4A alone (P = 0.01), 56% for EGCG alone (Fig. 3A). Additive efficacy was again demonstrated with HCV GT 2a (Fig. 3B).

High-Dose AR4A With Low-Dose EGCG Completely Inhibits HCV Infection In Vitro

To replicate the clinical scenario whereby therapeutic antibody products are administered at high concentrations and knowing that the in vivo bioavailability of EGCG may prove to be a limiting factor, we next titrated AR4a and EGCG to identify conditions capable of completely inhibiting HCV infection. The mAb AR4A, at a concentration of 10 μg/mL, consistently achieved approximately 95% neutralization of GT 1a HCV. Even at the highest dose of AR4A used (50 μg/mL), residual infection could be detected. Despite 95% neutralization with AR4A (10 μg/mL) alone, the addition of EGCG at 10 μg/mL resulted in a significant further inhibition of HCV (P < 0.001 for AR4A alone versus combination), with complete inhibition of HCV infection attained in a number of experimental repeats (Fig. 4). The activity of AR4A (10 μg/mL) and EGCG (10 μg/mL) in combination was further assessed against JFH-1 chimeric constructs expressing the structural proteins of GTs 2–6. High titer AR4A strongly neutralized GTs 4a, 5a, and 6a, which is a finding that is consistent with prior reports.23

AR4A Containing Treatment Arms Robustly Protected Against HCV Infection in SCID/uPA Mice; EGCG Alone Has No Apparent Protective Effect

HCV infection was established in 8 of 10 control animals, with 5 progressing to high-level replication over 42 days of follow-up (Fig. S3).

AR4A demonstrated clear efficacy. Only 2/11 mice receiving AR4A alone and 1/10 mice in the combination arm had HCV RNA detected above the threshold of 1000 IU/mL throughout study follow-up. “Survival free from HCV infection” was significantly increased in AR4A-treated groups. Given the small number of events in these study groups, no difference was demonstrable between the groups receiving AR4A alone or both AR4A and EGCG.

EGCG monotherapy failed to reliably protect from HCV infection, and observed outcomes were no different from the control group. The viral kinetics and a Kaplan-Meier curve of “survival free from HCV infection” are illustrated in Figs. 5 and 6, respectively.

EGCG Levels in Plasma and Liver Tissue of Mice Following 14 Consecutive Days of Administration

In order to assess if repeated high doses of EGCG 200 mg/kg/day were capable of achieving sufficient levels of EGCG (and its metabolites) in plasma and liver, we collected samples after 14 consecutive days of dosing. For this analysis, samples were obtained 4 hours following the final dose. Figure 7 illustrates the levels of detection of EGCG and its metabolites. The parent compound itself or its metabolites were detectable in all treated animals; however, importantly, the levels in both plasma and liver tissue were low (in the nanomolar range) when measured 4 hours after administration.

DISCUSSION

HCV-related liver disease is the leading indication for liver transplantation. Reinfection continues to drive inferior outcomes.12 Numerous unsuccessful attempts to address this discrepancy have been made. Recently, novel approaches provide some cause for optimism,16, 38 potentially shifting the paradigm whereby the treatment of HCV after liver transplantation is restricted to a delayed phase after transplantation when reinfection and histological disease have already been established.15, 39, 40

There is a need for safe, effective, and tolerable therapies that can use the window of opportunity provided by liver transplantation to prevent HCV reinfection. Complex drug-drug interactions require due consideration as does treatment durability in the face of a dynamic virus with high replicative capacity in the setting of impaired host humoral and cell-mediated immunity. Effective inhibition of HCV entry into “naïve” allograft hepatocytes represents a primary objective of preventative strategies.

In this work, we investigated agents acting primarily to inhibit HCV cell entry. Consistent with prior reports, AR4A and EGCG both reliably demonstrate cross-GT inhibition of HCV in vitro. Using these agents in combination was additive and the value of EGCG maintained significance even at high AR4A titers.

SCID/uPA mice receiving AR4A mAb by IP injection were significantly less likely to develop established infection following HCV challenge with a GT 1a inoculum. This is the first data reporting the anti-HCV efficacy of AR4A monotherapy in an animal model capable of sustaining HCV replication. AR4A clearly provides robust protection against the initial establishment of infection, demonstrating durable activity, which compares favorably with prior studies employing immunotherapy.21, 22

Despite clear in vitro inhibition, administration of EGCG alone, although well tolerated by SCID/uPA mice, did not offer protection against HCV challenge in vivo. This is the first report concerning the in vivo anti-HCV activity of EGCG. This lack of efficacy was likely driven by the low bioavailability of orally administered EGCG.41, 42 Previous studies have shown that peak plasma and liver concentrations of unconjugated EGCG in mice were 40 nmol/L and 3.5 nmol/g, respectively, following treatment with IG EGCG (75 mg/kg). EGCG undergoes extensive phase 2 metabolism, resulting in the formation of glucuronide conjugates and methylated metabolites. The levels of these metabolites have been shown in animal models to exceed the levels of the parent compound.41, 43 Studies have shown that EGCG is rapidly metabolized by catechol-O-methyltransferase resulting in the formation of methylated epigallocatechin gallate and dimethylated epigallocatechin gallate.44 These methylated metabolites have been shown to have reduced biological activity in a number of systems compared to the unmetabolized parent compound.45, 46

To date, no studies have examined the effects of methylation on the antiviral activity of EGCG. A recent study of influenza virus reported that methylated (-)-epigallocatechin has reduced inhibitory potency (IC₅₀ = 33.4 μg/mL) compared to unmetabolized (-)-epigallocatechin (IC₅₀ = 13.5 μg/mL) in vitro.47 Given the extensive phase 2 biotransformation of EGCG in vivo, further studies on the antiviral activity of the major metabolites are warranted.

Calland et al.48 have recently proposed a novel mechanism whereby EGCG and related natural compounds disrupt initial HCV attachment in vitro. Through an interaction with surface glycoproteins, such compounds alter the shape of HCV viral particles inhibiting efficient cellular attachment. With a proposed mechanism of action reliant on direct disruption of the initial stages of HCV cellular attachment and direct cell-cell transmission, failing to maintain adequate local concentrations will clearly limit the efficacy of EGCG in vivo.

Prior clinical studies administering passive immunotherapy to HCV patients undergoing liver transplantation have been disappointing. Polyclonal immunoglobulin failed to prevent HCV reinfection.9 Similarly, an anti-E2 mAb (HCV-AbxTL68), although effecting some decline in titers, did not prevent HCV recurrence in patients undergoing liver transplantation.49 Another group administered an anti-E2 mAb (MBL-HCV1) to 6 HCV transplant recipients. All eventually experienced reinfection with viral species harboring mutations in the target epitope.50

AR4A consistently demonstrated a robust ability to protect against HCV challenge with a patient-derived GT 1a inoculum, and its in vitro characteristics compare very favorably with those of other anti-HCV mAbs.23, 37, 51 The increased efficacy of AR4A mAb can be proposed by it uniquely targeting an epitope region abridging both E1 and E2, and containing a residue which is highly conserved across HCV species.

The SCID/uPA humanized liver mouse model is an extremely useful tool for conducting in vivo HCV studies.32, 33, 37 These mice lack an adaptive immune response, thus caution is required when translating findings to the clinical situation, considering that the preexisting host antibody bound to HCV viral particles can competitively inhibit effective neutralization by anti-HCV mAbs.52, 53 Likewise, it is possible that the native viral particle in the experimental systems used is physically different to that which circulates in association with human lipoproteins. An overestimation of efficacy could contribute to the comparatively poor activity observed in patient studies to date.9, 49, 50

Another limitation is the use of a patient-derived GT 1a inoculum only. GT 1 is the predominant HCV GT in Europe and North America; however, the immediate generalizability to individuals with HCV of diverse GTs awaiting liver transplant is somewhat limited. However, clear in vitro cross-GT activity of AR4A was demonstrated. Furthermore, the patient inoculum provided a heterologous HCV species challenge against which AR4A demonstrated impressive preventative capacity.

Despite these limitations, this study has yielded a number of important findings. We have demonstrated the proficiency of the anti-E1/E2 mAb AR4A to prevent the establishment of replicating HCV infection in vivo. On the other hand, EGCG, though an effective in vitro inhibitor, demonstrated a lack of definitive efficacy to protect SCID/uPA mice against HCV infection.

The results identify next generation anti-E1/E2 monoclonal antibodies as a potential therapeutic advance, which can constitute a primary component of future preventative approaches. The cross neutralizing ability of AR4A targeting a highly conserved epitope also provides further proof of principle that effective preformed antibody can contribute to protection against HCV challenge. A candidate vaccine has been shown to elicit such antibodies in vaccinated human volunteers, providing encouragement for demonstration of future preventive efficacy.54

To optimize the efficacy of this mAb in liver transplantation, additional agents to protect against virologic breakthrough and resistant mutants will be a prerequisite. Future clinical studies of anti-HCV mAbs incorporating novel pan-genotypic DAAs in complex transplant candidates with advanced HCV-related liver disease are required.

The application of EGCG, though eminently more available worldwide than DAAs, is limited by poor bioavailability. Further characterization of the interaction between the HCV viral particle and EGCG carries the potential to identify new agents targeting HCV cell entry.48 Moreover, capitalizing upon the unique circumstances of liver transplantation, novel applications of nontoxic compounds such as EGCG or derivatives could provide an opportunity to prime liver allografts ex vivo against HCV reinfection following implantation.

Employing therapies before or immediately at transplantation stands to deliver timely, effective, and tolerable HCV prophylactic combinations. This provides considerable optimism that in the future routine prevention of HCV reinfection after liver transplantation is in fact attainable. In achieving this goal, outcomes for patients with HCV undergoing liver transplant will be brought back in line with that of their non-HCV counterparts while ensuring optimal use of a highly valuable resource—human livers.