Present and Future Therapies of Hepatitis B: From Discovery to Cure

Natural History of Chronic Hepatitis B

The course of chronic HBV infection has been grouped into four phases: the “immune tolerant” phase; the “immune active/HBeAg-positive chronic hepatitis” phase; “immune active/HBeAg-negative chronic hepatitis” phase; the “immune active/HBeAg-negative chronic hepatitis” phase. However these terms may not accurately reflect the immunological status of patients in each phase but are useful for prognosis and determining need for therapy (2, 3). The duration of each phase varies from months to decades. Transitions can occur not only from an earlier to a later phase but regressions back to an earlier phase can also occur (4). It should be noted that not all patients go through all four phases. Furthermore, while the cutoff levels of ALT used to define different phases were traditionally based on upper limits of normal determined by clinical diagnostic laboratories, recent studies suggest that the true normal values are lower (5)

HBV Replication: From Basic Science to Drug Development

Advances in understanding the molecular biology and replication cycle of HBV have provided unprecedented insight into the mechanisms of action and treatment response of currently available drugs against HBV as well as potential future targets for therapeutic development (Fig. 1). HBV gains entry into hepatocytes initially through a low-affinity interaction between heparan sulfate proteoglycans (HSPG) on the hepatocytes and the antigenic loop (“a” determinant or antibody neutralization domain) of the HBV envelope proteins (6, 7) and then a high-affinity interaction of the myristoylated pre-S1 domain with the liver-specific receptor, sodium-taurocholate co-transporter (NTCP) (8). NTCP is exclusively expressed on the basolateral/sinusoidal membrane of hepatocytes. Its natural function is to re-transport conjugated bile salts (e.g. taurocholate (TCA)) into hepatocytes as part of the enterohepatic pathway (9). Accordingly, NTCP plays a key role in the liver tropism of HBV (10, 11). NTCP is also crucial for the host specificity of HBV. Two short sequence motifs within NTCP are sufficient to render the respective proteins from cynomolgus monkey and mouse functioning as a HBV receptor (12, 13). Additional host factors are probably required for efficient HBV entry. Fusion of HBV particles and release of nucleocapsids into the cells involves receptor-mediated endocytosis (14, 15).

The HBV genome-containing nucleocapid is then transported into the nucleus via a yet-undefined pathway, probably involving microtubule and nuclear importin machinery (16). In the nucleus, the relaxed circular, partially double-stranded genome (rcDNA) is then repaired to a full-length, circular DNA by covalently attached viral polymerase (P) and other incompletely understood mechanisms probably involving tyrosyl DNA phosphodiesterase of the topoisomerase and DNA repair pathway (17). The circularized protein-free genome then complexes with host histone and non-histone proteins including various histone-modifying enzymes into a minichromosome that functions as the template for transcription (18). Its transcriptional activity is regulated by epigenetic modifications and specific host transcriptional factors, such as HNF4 (19). HBV core and X proteins are also present on the minichromosome and probably play an important role in HBV transcription (18, 20, 21). The cccDNA is transcribed to three classes of HBV RNAs: genomic-length RNAs (pregenomic and precore RNAs coding for core gene products and P protein), S RNAs (S proteins) and X RNA (HBx protein). The pregenomic RNA transcript is reverse-transcribed by the P protein to rcDNA in the core-containing nucleocapsid. The nucleocapsid can either assemble into an infectious virion with the envelope proteins via the multivesicular body pathway (22) or recycle back to the nucleus for cccDNA amplification in a process probably controlled by the pre-S1 envelope protein and other host factors (23). The steady-state population of cccDNA is about 1–10 molecules per infected hepatocyte (24).

Current Therapies of Hepatitis B and Mechanisms of Actions

There are currently two classes of drugs approved for the treatment of hepatitis B: nucleos(t)ide reverse transcriptase inhibitors (NRTIs) and interferon-α. The first-line antiviral HBV medications include a nucleoside analogue - entecavir, a nucleotide analogue - tenofovir, and pegylated interferon-α, used as monotherapy. (25–27). Peginterferon is administered for 48–52 weeks. While it has a weaker antiviral activity than NRTIs, it is associated with a higher rate of HBeAg and HBsAg loss, possibly through a combination of direct antiviral and immunomodulatory effects. By contrast, NRTIs target only reverse transcription of pregenomic RNA to HBV DNA and have no direct effect on cccDNA. Long-term treatment with more potent NRTIs can lead to progressive loss of HBeAg and HBsAg with time.

Interferon-α, as a front-line host defense against viral infections, is known to induce interferon-stimulated genes (ISGs), which have promiscuous antiviral functions against a variety of viruses. Depending on the viruses, these ISGs play a diverse and pleiotropic role in targeting various viral functions at different steps of viral replication cycle and potently suppress viral infection and spread. Interferon-α has a direct anti-HBV effect and acts on multiple steps of the HBV replication cycle (Fig. 1) (28, 29). In addition, it has an immunomodulatory effect that can indirectly inhibit HBV replication by affecting cell-mediated immunity in vivo (30). Studies of the HBV kinetics in interferon-α treated patients suggest a more relevant role of the latter mechanism in mediating interferon-α’s anti-HBV effects (31).

Despite targeting multiple steps of HBV replication, the molecular mechanisms underlying interferon-α’s action remain to be fully defined. Interferon-α is thought to induce specific ISGs that inhibit HBV transcription or prevent the formation of nucleocapsid or target it for degradation (28, 29, 32). The responsible ISGs have not been clearly defined. Interferon-α’s effect on HBV transcription is partly mediated by epigenetic modifications of the cccDNA minichromosome (33). Recent development of infectious HBV cell culture systems provided the much-needed tools and models to study the effects of antivirals, including interferon, on HBV replication (33, 34). A recent study demonstrated that interferon-α and another putative antiviral cytokine, lymphotoxin-β (LT-β), induces the degradation of cccDNA in infectious cell culture systems (35). This effect is mediated by induction of the APOBEC3 family of proteins, specifically APOBEC3A by interferon-α and APOBEC3B by lymphotoxin (LT)-β. APOBEC3 functions to restrict foreign DNAs, such as those from invading microbial genomes, which activate the interferon response including induction of APOBEC3s as ISGs (36). The APOBEC3s are DNA editing enzymes and deaminate foreign double-stranded DNA cytidines to uridines (36). This conversion can lead to either C to T mutations or degradation of foreign DNA. In contrast cellular genomic DNA is unaffected. APOBEC3s are known to target HIV, AAV and possibly other DNA viral genomes for degradation (36). For HBV, the role of APOBEC3 has been controversial. APOBEC3G was shown to inhibit HBV replication in cell culture but the mechanism had been attributed to either direct inhibition of HBV replication or hyper-mutations from DNA editing (37–40). All the earlier studies were performed in HBV DNA transfection systems that could not be used to investigate cccDNA. In an HBV infectious culture system, the induced APOBEC3 interacted with the core protein and translocated to the nucleus to target cccDNA (35).

The development of nucleoside analogs owes much of its success to the comprehensive understanding of how HBV replicates. Based on the model of HBV replication, the P protein has been the primary therapeutic target in HBV drug development (Fig 1). While this advance represents a pivotal step in the chronicle of HBV treatment, knowledge of mechanism of action of this class of anti-HBV drugs also exposes its limitation, as discussed above.

While the second-generation NRTIs, such as entecavir and tenofovir, can potently suppress the DNA synthesis step of HBV replication, they have little effect on the level and activity of cccDNA, which has a long half-life and can persist for decades in the infected liver despite successful antiviral treatment (24). This limitation explains the necessity for a prolonged, possibly indefinite, treatment with this class of anti-HBV drugs. The turnover of cccDNA has been the subject of intense research because of its fundamental importance in HBV replication and therapy. Several mechanisms appear to explain the turnover of cccDNA in vivo. First, the direct cytopathic effect of activated HBV-specific T lymphocytes can cause death of infected cells. Second, gradual loss of cccDNA pool by cell proliferation in injured liver can account partly for gradual loss of cccDNA. Finally, non-cytopathic mechanism of eliminating cccDNA from infected cells contributes to the turnover of cccDNA (41). Interferons and other cytokines have been implicated in this “cell cure” mechanism, but the precise mechanism is unknown (41).

Entecavir and tenofovir can decrease the level of HBV DNA by 6 logs within 1 year of treatment and have low rates of antiviral drug resistance (0–1% after 5 years of continued treatment) (42, 43). However, rates of HBeAg seroconversion (~20% after 1 year and 40–50% after 5 years) and HBsAg loss (5–10% after 5 years) are low. Therefore, most patients require many years and often lifelong treatment with associated costs and risks of adverse reactions, drug resistance and non-adherence (44). Despite these limitations, antiviral treatment can reverse liver fibrosis and even cirrhosis, prevent cirrhosis complications, and reduce though not eliminate the risk of HCC (45, 46). Derivatives of tenofovir as prodrugs with improved pharmacological properties are being developed and may be of benefit in certain situation (47).

For patients who do not have cirrhosis or do not require immunosuppressive therapy, professional society guidelines recommend treating those in the immune active phase (25–27), although treatment at an earlier stage has been proposed to minimize unrecognized yet significant liver damage (48). However, treatment during the “immune tolerant” phase is associated with a low rate of HBeAg seroconversion and failure to completely suppress HBV DNA to non-detectable levels (49).

The ultimate goal of antiviral therapy would be to eliminate all forms of potentially replicating HBV but this may not be feasible because even in persons who recover from acute HBV infection with HBsAg to anti-HBs seroconversion, HBV persists in the liver in the form of cccDNA and can be reactivated during immunosuppressive therapy. A more realistic goal is a “functional cure” in which HBV DNA is not detectable after the completion of a finite course of treatment with loss of HBsAg and minimization of HCC risk over time. To accomplish this goal, a combination of antiviral drugs that target different steps in the HBV life cycle or immunomodulatory therapies to restore host immune response to HBV will be needed.

Combination Studies of Current Therapies

Given that only 2 classes of anti-HBV agents are currently available, combination therapy consist of two NRTIs or a NRTI plus peginterferon. In the latter case a NRTI and peginterferon may be combined simultaneously, sequentially, starting with either drug first, or as an add-on strategy with either drug first.

Initially, the clinical need to increase the potency of first-generation antivirals and to prevent emergence of antiviral resistance was the primary reason to test combination therapy with NRTIs. Unfortunately, this approach suffered from the fact that all NRTIs have the same virological target, the HBV polymerase. Thus, the treatment response observed in patients was similar to that of the most potent agent in the combination. The issue of antiviral resistance is now greatly diminished with the development of second-generation NRTIs, such as entecavir and tenofovir. The efficacy and safety of entecavir and tenofovir combination therapy was compared to entecavir monotherapy in previously untreated HBV patients (50). A greater proportion of subjects receiving combination therapy achieved viral suppression compared to entecavir alone, but the difference was not statistically significant (50). However HBeAg-positive subjects with baseline HBV DNA ≥ 0⁸ IU/mL receiving combination therapy had a significantly higher rate of virological response compared to those receiving monotherapy (50).

Conceptually, combination of peginterferon with a NRTI would be more likely to result in synergy because the drugs have different mechanisms of action. The concept being that inhibition of viral replication with a NRTI may augment the immune effects of peginterferon. Unfortunately, while studies of peginterferon in combination with first-generation NRTIs did show synergy in achieving viral suppression and reducing the incidence of antiviral resistance, off-treatment responses were similar to that of peginterferon alone (51, 52). The availability of more potent, second-line NRTIs together with a renewed interest in achieving HBsAg clearance has stimulated interest in combining these agents together or in combination with peginterferon.

Simultaneous peginterferon and tenofovir was evaluated in treatment-naïve patients with HBeAg-positive and -negative chronic hepatitis B (43). Patients receiving peginterferon and tenofovir had a higher rate of HBsAg loss than those receiving either drug along (43). Although these results are encouraging, they represent a small increase (6%) in HBsAg loss over peginterferon monotherapy and a benefit was mainly observed in those with genotype A infection.

Sequential therapy beginning either with a NRTI followed by peginterferon or vice versa for variable durations has been conducted in both HBeAg-positive and -negative subjects. In general these studies have not demonstrated a substantial benefit either in terms of on-treatment or sustained off-treatment HBV DNA suppression or HBeAg and HBsAg loss compared to peginterferon as a historical control (53–55).

Starting with NRTI first and adding peginterferon later would seem to be the most logical approach to combination therapy. The idea being that the NRTI would rapidly lower viral load and restore T-cell responsiveness and then adding peginterferon might hasten the decline of circulating and intrahepatic viral antigens leading to an improvement in the innate immune response (56). Several recent studies seem to support such an approach (57–59). Among HBeAg-positive subjects, higher rates of HBeAg seroconversion were achieved with add-on combination therapy of peginterferon and NRTI (27%) compared to NRTI only (0%) (58). Among HBeAg-negative subjects, HBsAg loss was reported in 6.6% of subjects at the end of therapy in the combination arm vs. 1% in the NRTI-only arm (59). None of these studies included an arm using peginterferon monotherapy and when compared to historical studies of peginterferon monotherapy, the results obtained with combination therapy are comparable.

A recent study compared peginterferon alone to peginterferon followed by add-on entecavir, or entecavir followed by add-on peginterferon (60). Rates of HBeAg seroconversion post-treatment were similar across treatment groups (60). With an add-on strategy, a longer duration of NRTI before add-on peginterferon and longer duration of peginterferon therapy was associated with higher rates of HBeAg and HBsAg loss.

In summary, there is insufficient data at present to recommend the use of combination therapy except in very special circumstances such as in subjects with very high baseline viral levels (>10⁸ IU/ml), or for management of subjects who have failed a first-line agent due to a suboptimal response or the development of multidrug resistance. Further studies are needed to address the benefit of various formats of combination therapy with peginterferon and more potent NRTIs.

HBV Entry Inhibitors

Entry inhibitors have been used successfully in treating viral infections. In particular, small molecules and antibody-based treatments are quite effective in treating acute viral infections (61). For chronic viral infection like HIV, entry inhibitors have also been successfully developed (62). For HBV, entry inhibitors can be applied in two ways. The first is in a preventive setting: entry inhibition blocks de novo HBV infection. This application has been successfully demonstrated in animal models (63) and is clinically a standard of care by using HBsAg-specific immunoglobulins to prevent reinfection after liver transplantation, to avoid vertical transmission of HBV from infected mothers to children, and for post exposure prophylaxis (64). Regarding chronic hepatitis B patients, whether entry inhibition would be a viable therapeutic option is debatable. It is conceivable that potent blockade of HBV reinfection in chronically infected patients can reduce viral load due to the turnover of HBV-infected hepatocytes (65). Previous studies suggested that hepatocyte turnover is indeed much faster in HBV-infected liver than in healthy hepatocytes, because of immune-mediated cytotoxicity (66). As a result, a sustained inhibition of de novo formation of cccDNA in hepatocytes may contribute to the eventual clearance of the virus with prolonged therapy, especially if it is used in combination with other potent anti-HBV drugs. Because hepatitis D virus (HDV) shares the same entry pathway, another potential application of entry inhibitors is in HBV/HDV co-infection.

HBV entry depends on the pre-S1 sequence, more specifically the myristoylated N-terminus of the large envelope protein (LHBsAg). Both the myristoylation and the N-terminal 75 amino acids are required for infectivity of HBV (67, 68). It was shown that synthetic lipopeptides representing this subdomain potently inhibit HBV infection. The mode of action of such peptidic inhibitors (Myrcludex B, for example) could be attributed to specific receptor binding (11). Myrcludex B successfully passed phase I clinical trials (Blank et al, submitted). Moreover, since the natural role of NTCP as a bile salt transporter has been studied in some detail, molecules already known to bind or inhibit the function of NTCP have been tested. Cyclosporine A and its derivatives (e.g. Alisporivir) or approved drugs like Ezetimibe are amongst those that have been demonstrated to inhibit HBV entry (69–71).

Myrcludex B, cyclosporine A and other substrate-analogs inhibit bile salt transport by NTCP. Accordingly, these molecules may elevate bile salts and other transported substrates in the serum of patients. This concern may be a clinically manageable problem. First, people with polymorphisms in NTCP resulting in a functional knockdown show very moderate clinical symptoms and do not develop any specific pathology (72). Second, NTCP knock-out mice are viable, show elevated conjugated bile salt levels without symptoms but have a slight retardation in growth during development (73). Most importantly, the antiviral effect of Myrcludex B and cyclosporin A is already apparent at a much lower concentration than that required for inhibiting bile acid transport (>100-fold difference) (12, 70). Thus, entry inhibition should be clinically achievable without significant interference with the transporter function of the receptor.

Myrcludex B is currently being tested in two ongoing clinical trials (74). Preliminary results suggested that Myrcludex B is safe and well tolerated in HBsAg-positive patients with or without HDV co-infection. A decline in the HBV DNA level (>1 log₁₀) was reported in 87% of patients at 12 weeks of treatment (10 mg/day) and the decline continued with extended treatment beyond 12 weeks. Myrcludex B treatment at high doses was associated with some bile acid elevation.

HBV Capsid Inhibitors

Several classes of inhibitors of pgRNA packaging and HBV capsid assembly have been identified. They function to dysregulate or selectively inhibit either pgRNA encapsidation or nucleocapsid assembly or both. The first of these was the phenylpropenamide derivatives AT-61 and AT-130 (75). These compounds selectively inhibit viral pgRNA packaging (76) and are active against both wild-type and lamivudine-resistant HBV (77, 78). As a class and at the molecular level, these agents have been shown to induce tertiary and quaternary structural changes in HBV capsids. AT-130 binds to a promiscuous pocket at the core dimer-dimer interface (79). This binding decreases viral production by initiating virion assembly prematurely in the replication cycle, resulting in morphologically normal capsids that are empty and non-infectious (76).

The second group of inhibitors are the heteroaryldihydropyrimidines, or B-HAPs, that inhibit HBV virion production in vitro and in vivo by preventing capsid formation (80). The best studied of these B-HAPS, Bay 41-4109, has a dual mechanism of action by inhibiting encapsidation directly and causing a concomitant reduction in the half-life of the core protein. Structural studies of this class of inhibitors revealed that they induce inappropriate capsid assembly at low concentrations, and when in excess, promote a misdirected assembly reaction and decreased capsid stability (81, 82). Like the phenylpropenamides, the B-HAPs are active against NRTI-resistant strains of HBV (78).

Other inhibitors targeting the nucleocapsid are being developed by several biotech companies (83) (Table 1). In vitro studies have demonstrated strong synergy when these inhibitors are used in combination with currently approved NRTIs (77, 78)

Inhibition of HBV Gene Expression

Persistence of HBV results from an ineffective antiviral immune response against the virus and one of the ways HBV orchestrates this is via excess production of sub-viral particles containing HBsAg. These non-infectious sub-viral particles (SVPs) may act as a decoy for the immune system (84), especially for mopping up potentially neutralizing anti-HBs. High levels of HBsAg, in the range of 400 μg/mL (0.4% of total serum protein), are commonly found in the blood of patients with chronic hepatitis B (85), and may interfere with HBV-specific immune responses (86, 87).

An alternative approach to inhibit HBV gene expression has been successfully achieved in vitro using molecular-based therapies targeting the viral mRNA. Viral mRNA can be directly targeted using anti-sense oligonucleotides, ribozymes, or RNA interference (RNAi) (88). Of these, RNAi appears most promising since efficient in vivo delivery systems have been developed by a number of biotech companies (Table 1).

RNAi is a process by which small interfering RNA molecules of 21–25 nucleotides induce gene silencing at the post-transcriptional level to effectively knock down the expression of the gene(s) of interest. Such short interfering RNA can lead to transcriptional silencing or translational repression (89). These processes are critical in cell growth regulation and tissue differentiation and involve the Drosha and Dicer enzyme complexes, the RNA-induced silencing complex (RISC) and the nuclease Argo (89). The extensive use of overlapping RNAs and ORFs within the HBV genome, makes for an attractive target for inhibition by RNAi (90). Both cell culture and mouse model studies have shown that RNAi, delivered as an expression plasmid, is able to inhibit all steps of HBV replication (91). In transgenic mice, RNAi expression has been shown to significantly reduce the secretion of HBsAg in serum, reduce both HBV mRNAs and genomic DNA in the liver, and eliminate hepatocytes stained positive for core antigen (92). These mouse studies have been extended to a chronically infected chimpanzee (93). Currently, a phase 2 placebo-controlled dose-escalation study with the DPC-NAG-ARC-520 formulation has been initiated in HBeAg-negative chronic hepatitis B whose viremia was controlled by entecavir and showed a 50% drop in HBsAg levels in the treated compared to placebo patients (94). Another RNAi platform has demonstrated similar success in its pre-clinical evaluation with a 2.3-log₁₀ reduction in HBsAg in chronically HBV-infected chimpanzees (Alnylam’s company press release). Other RNAi-based regimens are currently being developed and tested (Table 1).

Inhibitors of HBV cccDNA Formation and Stability

Because the cytoplasmic nucleocapsid DNA is the precursor for cccDNA biosynthesis, complete inhibition of viral DNA replication in the nucleocapsids with polymerase inhibitors should preclude de novo cccDNA formation. However, clinical studies demonstrated that although NRTI monotherapy for 48–52 weeks reduced circulating viremia by ~5 log₁₀ and cytoplasmic HBV DNA levels in the hepatocytes by approximately 2 log₁₀, reduction of cccDNA was much less pronounced, only by 0.11 to 1.0 log₁₀ (24, 95). Moreover, sequential analyses of viral DNA replicative intermediates and core antigen-positive hepatocytes in the livers of woodchuck hepatitis virus (WHV)-infected woodchucks before and during clevudine (a NRTI) therapy revealed that after more than 6 weeks of therapy, all WHV DNA replicative intermediates were markedly reduced, with the exception of cccDNA, which remained as the predominant viral DNA species in the liver (96).

Concerning the failure of prolonged NRTI therapy to eradicate cccDNA, one possibility is that the currently available NRTIs do not completely inhibit viral DNA synthesis in every infected hepatocyte in vivo, allowing for continuous replenishment of cccDNA pool via the intracellular amplification pathway. NRTIs are pro-drugs requiring activation by host cellular nucleoside kinases, the expression and function of which may be heterogeneous in the liver. Therefore hepatocytes may have varying abilities to activate the NRTIs, resulting in incomplete inhibition of HBV DNA replication. Emergence of drug-resistance mutations during apparently effective NRTI therapy suggests that residual HBV replication and de novo cccDNA synthesis still occurs at a low level (97).

Alternatively, failure to eradicate cccDNA by prolonged NRTI therapy may also be due to the extraordinary stability of cccDNA (98). cccDNA may persist in a “latent” state amidst the host chromosomes and remain as a reservoir for later HBV replication. Healthy hepatocytes in the absence of immune response or inflammatory reaction have a half-life of over 6 months (99, 100).

What we have learned from NRTI therapy is that eradication of cccDNA is essential for the cure of chronic hepatitis B. Combination therapies with NRTIs and one or multiple novel antiviral drugs targeting different steps of HBV replication may completely inhibit HBV DNA replication and thus accelerate the reduction of cccDNA. The other approach would be to directly purge the pre-existing cccDNA or permanently silence cccDNA transcription.

Recent strategy to cleave cccDNA molecules or inhibit its transcription by generating cccDNA sequence-specific endonucleases with zinc-finger nuclease, transcription activator-like effector nuclease or CRISPR/cas9 technology has been tested in cell culture or mouse model (101, 102) but efficient and targeted delivery of these antiviral genes to all HBV-infected cells in vivo is a major challenge for clinical application. Another approach is to target the other enzymatic function of HBV polymerase, RNaseH, which is required for HBV replication and cccDNA formation. Recent studies have identified potential inhibitors of HBV RNaseH (103).

Further understanding the molecular mechanism of cccDNA metabolism and functional regulation is essential for identifying and validating molecular targets for rational development of antiviral drugs to eradicate or transcriptionally silence cccDNA. As discussed above, recent studies on the molecular mechanism of immune control of HBV infection by IFN-α demonstrated that cccDNA can be specifically targeted for degradation by a cytidine-deamination mechanism (35) and its transcription can be silenced by epigenetic modification (33, 104). These findings raise a potentially exciting possibility of targeting cccDNA via pharmacological activation or augmentation of the host intrinsic antiviral pathways. Moreover, investigation into the role and mechanism of HBx and core protein in cccDNA metabolism and function may reveal viral-host interactions for selective elimination or silencing of cccDNA (20).

Additional efforts have been made to discover cccDNA-targeting compounds via high-throughput cell-based phenotypic screening. This unbiased approach, while attractive, is currently hampered by a lack of efficient HBV infection cell culture systems and convenient assays for high-throughput quantification of HBV replication and cccDNA quantification. Di-substituted sulfonamides had been identified as cccDNA formation inhibitors in a screen of 85,000 small molecular compounds (105). The recent rapid progress in the establishment of efficient HBV infection cell culture system may ultimately allow the development of cell-based assays for high-throughput screening of cccDNA-targeting antivirals.

Immune Mechanisms of HBV Control and Implications for Therapy

The pathogenesis of chronic HBV infection involves not only viral mechanisms by which HBV establishes a persistent infection, but also the host responses to infection. The latter includes the response of hepatocytes to HBV infection as well as the interplay of the virus and infected cells with the other parenchymal and non-parenchymal cells in the liver, i.e. Kupffer cells, endothelial cells, fibroblasts, and non-resident immune cells that are recruited to the site of infection. HBV has evolved mechanisms to counteract and escape these different host responses to establish a chronic infection. Recent studies point out a critical role of the liver microenvironment in the elimination or control of HBV (106, 107) (Figure 2). While much has been learned about the HBV-specific adaptive immunity, the early and innate immune response during acute HBV infection remains largely unknown. In addition, few studies have examined intrahepatic immune responses in patients with chronic HBV infection. Available data suggest impaired responses but the mechanism of this impairment is unclear (107).

In chronic hepatitis B, the antiviral B and T cell responses are quantitatively and/or qualitatively defective. For example, anti-HBsAb is generally undetectable in the setting of excess circulating HBsAg. Furthermore, antiviral T cells show impaired antiviral effector function in vitro. However, this host immune response, despite being dysfunctional, exerts at least partial viral control in vivo since immune suppression with immunosuppressive therapies result in increased viremia (58, 108). HBV persistence with antiviral immune dysfunction is also associated with the induction of immune inhibitory pathways including PD-1, CTLA-4, Bim, arginase and FoxP3+ Tregs (109–114). These pathways, likely induced in response to continued inflammation, viral replication and antigen expression, can dampen both cytopathic inflammatory responses as well as non-cytopathic antiviral effector functions. Thus, the antiviral effector T cell function may be enhanced by blocking one or more of these inhibitory pathways (111, 115), raising the possibility for potential therapeutic application in chronic viral infections such as chronic hepatitis B.

Based on our knowledge of the immune mechanisms of chronic HBV infection, several approaches to restore innate or adaptive immunity or both to control HBV infection in combination with other direct antiviral strategies have been applied (106, 107). These approaches can be broadly divided into viral-nonspecific and – specific modalities. The first involves general immunomodulatory agents and the latter aims to activate HBV-specific immune response by applying the technologies of therapeutic vaccination. As discussed above, the efficacy of interferon-α therapy can be partly attributed to its immunostimulatory effect. A promising approach emerges from the field of Toll-Like Receptors (TLRs). Various TLR agonists with potent immunostimulatory effects have been developed (116). Their administration to HBV patients leads to both intra- and extra-hepatic induction of type I interferons and other cytokines that may contribute directly to antiviral activity or indirectly result in activation of innate and adaptive immune responses. The second approach involves the blockade of negative immunoregulatory pathways (i.e. co-inhibitory signals, inhibitory cytokines, Treg cells), which may induce a partial restoration of HBV-specific T cell. Third, engineering of redirected T cells may result in a de novo reconstitution of functionally active HBV-specific T cells and activation of heterologous T cells. Whether inhibition of suppressive effect(s) of HBV can lead to restoration of HBV-specific innate and adaptive immune responses remains a challenging question. Several lines of evidence suggest that HBV interferes negatively with these host immune responses. A more detailed understanding of the specific mechanisms is mandatory before new ways of restoring immune responses by targeting virus-specific factors can be explored. HBV-specific strategies may prove more effective and safer than virus-nonspecific approaches.

A concern of the various immunotherapies is the potential risk of auto-immunity and/or exacerbation of liver damage by immune-mediated death of hepatocytes in vivo. Careful consideration of benefit versus risk and close clinical monitoring would be needed in these approaches.

HBV-nonspecific Immunomodulatory Agents

HBV-Specific Modified T Cells

As discussed above, HBV-specific T cells are either exhausted or nonresponsive in chronic HBV infection. This therapeutic approach is designed to provide genetically engineered T cells to target and eliminate HBV-infected hepatocytes. The strategy to genetically modify patient’s T cells to express HBV-specific T cell receptors (TCRs) and then infuse them into the same patients with HBV-associated HCC showed some promise (124, 125). But the variable and MHC-restricted nature of the interaction between TCR and its ligand and the skepticism that whether one or two such modified T cells would be sufficient to mount an effective T cell-based immune response may limit the clinical application of this approach. The recent emerging technology of “chimeric antigen receptor (CAR)” in the field of cancer therapeutics has been extended to treatment of persistent viral infections (126). The CAR approach is to generate a chimeric receptor expressing an extracellular target-binding domain, a hinge and membrane-anchoring region, and one or more intracellular signaling domain (126). The target-binding domain is derived from the light and heavy chain sequences of a single chain variable fragment (scFv) of the immunoglobulin. In the case of HBV, the target could be the cell-surface form of HBsAg and the single-chain variable fragment derived from a construct with high-affinity anti-HBs activity (127). The binding of the CAR-modified T cells to HBV-infected hepatocytes can trigger proliferating or activating signals to initiate an effective anti-HBV T-cell response. This strategy has been applied to HBV animal models with some promise (128). It remains to be seen whether CAR-modified T cells can achieve a broadly acting and potent anti-HBV response that is sufficient for viral clearance in chronic HBV-infected patients.

Therapeutic Vaccines

The goal for therapeutic vaccination in chronic hepatitis B is to induce sufficient anti-HBV immune responses to eliminate and/or cure infected hepatocytes without undue host cell damage, prevent viral spread to new hepatocytes and promote long-term viral control. These approaches leverage our accumulating knowledge in the adaptive immune responses of HBV infection and focus on restoring or activating endogenous HBV-specific immune responses that initially targeted HBsAg and later expanded to other HBV antigens by using recombinant proteins, CTL epitope vaccine, viral vectors and DNA vaccination. These vaccines are being combined with antiviral drugs and immune modulators to maximize their effects. An intriguing strategy to personalize antigen presentation to induce anti-HBV immune response involving monocytes has been recently proposed (129).

Therapeutic Pipeline and Conclusion

Based on the literature, expert input, publicly disclosed information of various pharmaceutical companies, clinicaltrial.gov website, we generated a table summarizing the current status of various anti-HBV drugs or biologics in the development pipeline (Table 1). While many of them are still in preclinical development, several have advanced to clinical trials. As discussed above, some of them showed early promise and will likely advance to the late clinical trial phase for more definitive proof of the preliminary success. In this review, we have summarized the major therapeutic approaches and novel molecular targets for anti-HBV drug development, and provided a knowledge-based rationale behind these various strategies. It is possible that new and additional technologies may emerge as the field advances. To achieve a more sustained and effective control of HBV infection, a combination of the existing HBV therapies and one or more of the above modalities, either small-molecule drugs or biologics, will be necessary. With the concerted efforts of private and public sectors, the next milestone in the therapy of HBV infection – a functional “cure” that has remained elusive - is likely within our grasp within the next decade.