HBV exists as ten different genotypes (A–J) and multiple sub-genotype with geographically distinct distributions which are becoming increasingly blurred through human migration [16]. CHB should not be thought of as a single disease, but rather as a “disease state” or spectrum with marked diversity in CHB natural history including differences in age of acquisition (neonate vs. adult), modes of transmission, disease progression, replication phenotype, response to therapy, and disease resolution across HBV genotypes [16]. For example, persons infected with genotype A respond better to treatment with IFN-α than do all other genotypes a nd patients infected with genotype B respond much better than genotype C. However interferon is virtually ineffective against patients infected with genotype D. Genotype C is generally associated with more severe liver disease and a higher propensity for liver cancer than most other genotypes, although there are exceptions with subgenotype A1 associated with rapid progression to cancer without prior cirrhosis in young African males [17, 18] via mechanisms that have not been fully identified. HBV sequences differ by at least 8 % across the complete HBV genome between genotypes, and by between 4 and 8 % between sub-genotypes, suggesting that new treatments targeting specific areas of the HBV genome (such as the HBeAg) may need to be tailored to individual genotypes or subtypes.

Current direct acting antiviral therapies (DAAs ) inhibit the reverse transcription of pgRNA to DNA, i.e., the replication stage of HBV that takes place within the viral nucleocapsid (Fig. 14.1). They include the class of nucleos(t)ide analogs (NA) and act as competitive inhibitors of the HBV pol, inhibiting synthesis of viral DNA from the pgRNA template through chain termination. The clinical activity of these molecules is explained in detail in Chap. 16.

DAAs have no direct impact on the initial formation of cccDNA , although in reducing the amount of RC DNA they should indirectly impact cccDNA that is formed following the recycling of RC DNA to the nucleus. However, since this recycling happens early in the HBV infection cycle and DAA treatment routinely commences following the establishment of persistent infection long after initiation of HBV infection, the impact of DAAs on cccDNA levels is essentially minimal. This is important, as the continual presence of cccDNA in the nucleus, even in the absence of productive viral replication, is one of the main impediments to curing HBV infection. In turn, continual production of hepatitis B surface antigen (HBsAg) originating from cccDNA or integrated DNA is associated with disease progression and liver cancer, meaning that eradication of HBV-associated disease may only be achieved when strategies are developed that eliminate or suppress cccDNA.

To date five drugs belonging to the class of NA have been approved for treatment of CHB. These are lamivudine (LMV), adefovir dipivoxil (ADV), entecavir (ETV), telbivudine (LdT), and tenofovir (TDF). These drugs are divided into three groups, the l-nucleosides (LMV and LdT), acyclic phosphonates (ADV and TDF), and d-cyclopentane (ETV) groups. The effectiveness of many of these drugs is limited by the selection of resistant strains of HBV during treatment [19]. This is because the “error prone” nature of the HBV reverse-transcriptase results in a high mutation rate (1 in every 10⁵ nucleotide substitutions for each cycle of replication [20]) leading to a large population of variant HBV sequences, some of which encode preexisting mutations in the HBV polymerase gene which confer resistance to one or more DAAs. These mutant strains of HBV are typically selected on therapy, giving rise to a dominant population of drug resistant viruses. The different DAAs have different propensities for selection of resistance, with the most widely used drug LMV driving selection of resistant HBV in up to 23 % of patients following 12 months of therapy, rising to 80 % by 5 years of treatment [21]. In contrast it is encouraging that there is no evidence to date for selection of HBV strains that encode resistance to TDF, even following 5 years of treatment [22, 23] (Table 14.1). Although rates of resistance are low for adefovir and entecavir monotherapy, in the setting of prior resistance to LMV, high rates of resistance to these drugs is observed because of cross-resistance (Table 14.1, adapted from Zoulim and Locarnini, 2009 [19]).

The molecular mechanisms of resistance to d rugs for treatment of CHB have been eloquently described previously [19, 30] but are briefly summarized herein.

To date, eight codons within the HBV polymerase are associated with primary drug resistance to NAs. These are codons 169, 180, 181, 184, 202, 204, 236, and 250 [19] (Fig. 14.2).

The reason for the strong barrier to selection of resistance to the acyclic phosphonate TDF is unclear. It remains to be determined whether continued use of this drug as monotherapy leads to selection of TDF resistant HBV. Nonetheless, the high degree of potency exhibited by TDF (and related compounds such as TAF—see below), and the high genetic barrier to selection of HBV resistance shows there is a role for these compounds in first line therapy against CHB. These compounds also have an important role when first line therapy with alternative NA treatments (such as LMV) has led t o the select ion of HBV encoding resistance mutations, or in the setting of failed immunomodulatory therapy.

Although TDF is a very effective antiviral compound with no evidence to date of selecting antiviral resistance, there are reports of small numbers of HBV infected patients who do not respond to TDF therapy [130]. The reasons for this are unclear—it is likely that poor compliance is a contributor in these studies, but where compliance was not an issue it suggests reduced TDF efficacy might be related to quasispecies diversity or coinfection with multiple genotypes, such as genotypes A and G. Tenofovir alafenamide (TAF) is a TDF prodrug that provides efficient delivery of active drug to hepatocytes at reduced dosage, with improved plasma stability [32]. A recent study of 51 CHB subjects with HBeAg negative HBV infection showed that doses as low as 8 mg per day for 4 weeks resulted in similar levels of viral decline as the standard 300 mg daily dose of TDF [33]. Further clinical studies in larger cohorts are currently underway at a dosage of 25 mg per day.

Although the discovery of the NTCP recept or for HBV entry has been relatively recent [3, 34, 35], it has been known for some time that synthetic peptides derived from the large envelope protein block HBV entry, as well as entry of the related hepatitis delta virus (HDV), which utilizes the HBV envelope protein for viral entry [36–39]. This discovery has led to the production of an HBV “entry inhibitor” Myrcludex B, which is a myristlyated PreS1 peptide currently under clinical trial. It has recently been shown that Myrcludex B not only prevents HBV spreading from infected human hepatocytes in vivo, but also hinders amplification of the cccDNA pool in initially infected hepatocytes [40]. This important finding suggests Myrcludex B could be a useful tool in the treatment of CHB, as well as more obvious applications such as the prevention of reinfection following liver transplantation.

The HBV core protein (nucleocapsid) is critical for viral RNA packaging, reverse transcription and intracellular trafficking. It is also an important for cccDNA generation and stability, binding directly to the cccDNA [49, 50]. These properties suggest the core protein is a suitable therapeutic target, enabling regulation of multiple facets of the HBV life cycle. Elucidation of the core crystal structure in 1999 [10] has provided new insights in core protein assembly, including defining the core dimer–dimer interface [10] leading to development of a range of compounds with therapeutic potential.

The continual presence of cccDNA in the hepatocyte nucleus is a major impediment to HBV treatment and cure, as it is not directly targeted by current treatments and is a continual source of the pgRNA transcriptional template and additional viral mRNAs. Although a proportion of cccDNA molecules are removed during resolution of acute (transient) HBV infection, due to immune-mediated clearance of HBV infected cells during hepatocyte turnover [65–71], reactivation of HBV infection in immunosuppressed patients that may have previously resolved their acute infection [72–76] shows that cccDNA is not completely cleared during disease resolution in all patients. Hence the removal of cccDNA, or the suppression of its transcriptional activity, is a desired aim for treatment and cure regimens.

Epigenetic regulation of HBV cccDNA as anti-HBV therapy needs to be carefully considered due to possible off-target effects on acetylation of host DNA and their ability to reactivate HBV infection. Indeed, reactivation of HBV has been observed following treatment of patients with the anticancer HDAC inhibitory drug Romidepsin [91] and in patients receiving immunosuppressive therapy [72–76, 92]. This suggests that not all cccDNA is eliminated in some patients who have previously cleared an acute infection, raising questions about the true definition of HBV cure. With the development of increasingly sensitive diagnostic tools it may become evident that some level of cccDNA persists in all patients, even with the clearance of the majority of infected hepatocytes, meaning that natural eradication of HBV is unachievable. Through specifically targeting cccDNA, it may be possible to induce cure, although this approach is currently not being considered based on current knowledge and clinical experience. A further problem with this approach for controlling HBV infection are “off-target” effects, since they may also affect the epigenome of host chromosomes. Approaches need to be developed that are specific to HBV cccDNA epigenome and not host DNA. Additional important questions include whether such alteration of cccDNA promotes integration of the modified DNA and whether this in turn may lead to expression of novel protein(s) and or a greater propensity for progression to HCC.

The HBeAg is required for the establishment of persistent infection, with CHB rarely establishing in babies borne to HBeAg-negative mothers, in contrast to 90 % of newborns from HBeAg-positive mothers developing CHB [93]. In turn, low levels of HBeAg in patient serum are associated with HBeAg seroconversion in the setting of IFN-α treatment [94], this being an accepted marker of treatment response that usually precedes HBsAg clearance. Taken together these findings suggest that the HBeAg itself may be a suitable therapeutic target, with reductions in the levels of HBeAg perhaps driving the immune response to clear infection, or reducing mother to baby transmission in newborns. Unlike the related HBcAg, the HBeAg crosses the placenta [95], and is thought to establish tolerance to the HBcAg and HBeAg in the developing fetus [96, 97]. The HBeAg protein sequence differs from the HBV core protein by the addition of ten highly conserved [98] amino acids at the N-terminus and removal of the C-terminal arginine rich core domain [99]. Recent delineation of the HBeAg crystal structure [9] shows that the cysteine molecule at position −7 in the N-terminal 10-mer sequence (SKLCLGWLWG) bonds with cysteine 61 to form a dimer whose structure differs markedly from the HBcAg dimer [10]. It has been suggested that these differences may explain some of the biophysical, biochemical and functional properties of the HBeAg and HBcAg molecules [9, 10]. For example, the HBeAgs inability (usually) to form capsids, instead forming high molecular weight soluble forms in the blood [100], its propensity to initiate tolerogenic rather than immunogenic T-cell responses [101–104], inability to bind and activate B cells without T cell support [102–104], and suppression of innate immune signaling pathways [105–107]. It is hoped that the characterization of the HBeAg crystal may lead to development of molecules such as humanized mAbs which specifically target the HBeAg that may in turn drive HBeAg and HBsAg seroconversion. This approach should also be considered for other HBV proteins as it is becoming increasingly important for treatment of chronic conditions such as rheumatoid arthritis and Crohn’s disease [108]. Our group has recently identified a novel single variable domain (VNAR) of the shark immunoglobulin new antigen receptor (IgNAR) antibody which displayed biologically useful affinity for recombinant and native HBeAg, and recognized a unique conformational epitope [109]. This molecule was subsequently engineered for ER-targeted in vitro delivery to function as an intracellular antibody (intrabody), which effectively regulated precore/HBeAg expression, showing potential as a HBeAg therapeutic that is worthy of further investigation, exploiting the possible importance of FcRn expression on hepatocytes and their role in intracellular antibody “neutralization” [110].

The hepatitis B virus X protein is a non-structural prote in that is essential to initiate and maintain virus replication after infection [111]. Although HBx does not bind directly to cccDNA, its recruitment is mediated through its interaction with a wide range of cellular transcription factors and cofactors [112]. Modulation of cccDNA transcription and HBV replication by “HBx-knockout” strains of HBV which transcribe less pgRNA that wild-type HBV strongly implicates the HBx protein in regulation of cccDNA transcription [58, 111]. These findings suggest that the HBx protein may be a suitable therapeutic target. Indeed numerous studies have shown that siRNA mediated knockdown of the HBx protein reduces HBV replication [113, 114]. The overlapping nature of the HBV genome means that siRNAs targeting the HBX gene also target the major pgRNA transcriptional template, as well as the 3′ end of the HBV polymerase. Consequently it is unclear if HBV replication is reduced in these models due to reduced HBx-mediated transcription of cccDNA, or decreased levels of the pgRNA itself, or both. S tudies using the NTCP infection model as well as in non-transformed hepatocytes are required. It should also be noted that the overlapping nature of the HBV genome means that mutations at positions A176T2/G1764A which frequently occur in the HBV basal core promoter also alter the HBx coding sequence at positions 130 and 131. These changes are associated with increase virulence and disease progression, with Tseng and colleagues showing that the A1762T mutation in particular was associated with increased risk of liver cirrhosis [115]. The role of HBx in this association is unclear. Identifying the crystal structure of the HBx protein may enable identification of the exact location of binding sites with host transcription factors, facilitating the HBx protein as a novel therapeutic target.