Asian-Pacific consensus statement on the management of chronic hepatitis B: a 2008 update - Hepatology International

HBV, pathogenesis, and natural course

Chronic HBV infection is a serious clinical problem because of its worldwide distribution and potential adverse sequelae. It is particularly important in the Asian-Pacific region where the prevalence of HBV infection is high. In this part of the world, the majority of HBV infection prevalence is acquired perinatally or in early childhood, and some patients may be superinfected with other viruses that may influence the clinical outcomes.

Previous studies revealed the presence of two replication pathways, namely, episomal and integrated forms, and reverse transcription process in HBV infection [5]. It has been recognized that covalently closed circular DNA plays a key role in the maintenance of chronic HBV infection [6]. As HBV is not usually cytopathogenic by itself, chronic HBV infection is a dynamic state of interactions among the virus, hepatocytes, and the host immune system. The natural course of chronic HBV infection in this geographic region can be divided into (i) immune-tolerant phase, (ii) immune clearance phase, and (iii) residual or inactive phase. HBV reactivation and relapse of hepatitis may occur in some patients who are in the residual or inactive phase.

Patients in the immune-tolerant phase are usually young, hepatitis B e antigen (HBeAg) seropositive with high viral loads (>2 × 10⁶ to 2 × 10⁷ IU/ml or >10⁷–10⁸ copies/ml) but normal serum alanine aminotransferase (ALT) and no or minimal clinicopathological changes. The results of a recent 5-year follow-up study have confirmed that adults in the immune-tolerant phase show minimal disease progression [7]. However, HBeAg-positive subjects older than 40 years with persistently “high normal” ALT may have significant hepatic necroinflammation or fibrosis [8]. During the immune clearance phase, hepatitis activity and even acute flares with serum ALT levels over 5 times upper limit of normal (ULN) may occur, and these may sometimes be complicated by hepatic decompensation. These ALT elevations and hepatitis flares are the result of host’s immune responses against HBV, such as HLA-class I antigen restricted cytotoxic T lymphocyte (CTL)-mediated response against HBV antigen(s) expressed on hepatocytes with resultant apoptosis and necrosis. Higher ALT levels, therefore, usually reflect more vigorous immune response against HBV and more extensive hepatocyte damage [9]. This is eventually followed by HBeAg seroconversion to its antibody (anti-HBe) and/or undetectable HBV-DNA. The estimated annual incidence of spontaneous HBeAg seroconversion was 2–15%, depending on factors such as age, ALT levels, and HBV genotype [9, 10]. Some patients may experience only transient and mild elevation of serum ALT levels before HBeAg seroconversion [11]. HBeAg seroconversion is followed by clinical remission (inactive chronic HBV infection) in the majority of patients. However, active hepatitis may relapse due to HBeAg seroreversion or occurrence of HBeAg-negative hepatitis. The estimated annual incidence of hepatitis relapse was 2.2–3.3% [11, 12], being higher in males, genotypes C infected, and those who have HBeAg seroconversion after age 40 [13]. A recent long-term follow-up study (mean = 12.3 years) involving 1,241 incidentally identified subjects with inactive chronic HBV infection showed a lower annual incidence of 1.5%, being significantly much lower in younger patients, especially those younger than 30 years [11]. All these findings suggest that earlier HBeAg seroconversion or shorter HBeAg-positive phase is associated with higher chance of sustained remission. Asymptomatic HBeAg-negative subjects with HBV-DNA > 2,000 IU/ml may also experience hepatitis flares and disease progression such as in HBeAg-positive patients [10–15]. Since the immunopathogenesis of HBeAg-negative hepatitis is similar to that of HBeAg-positive hepatitis, this phase can be viewed as a variant of immune clearance phase.

A prospective study involving 684 patients with chronic HBV infection showed that cirrhosis developed at an estimated annual incidence of 2.1%, and that age, the extent, severity, frequency of flares, and the duration of hepatic lobular alterations were factors for disease outcomes and HBV clearance [16]. Patients with chronic HBV infection with persistent HBeAg seropositivity have an even higher incidence (3.5% per year) of cirrhosis [17]. One study showed that 23% of the patients with HBeAg-negative hepatitis progressed to cirrhosis during a follow-up period of 9 years (range = 1–18.4) [12]. A recent Korean long-term follow-up study (mean = 120 months) involving 188 patients (52 HBeAg-negative patients) showed that age and persistent ALT elevation are independent factors for the development of cirrhosis, decompensation, and hepatocellular carcinoma (HCC) [18]. HCC develops at an annual incidence of 3–6% in patients with cirrhosis and far less frequently in noncirrhotic patients [12, 19, 20]. Seropositivity for HBeAg and/or HBV-DNA > 2,000 IU/ml are significant risk factors for cirrhosis and HCC development, even in asymptomatic subjects with chronic HBV infection [21–25].

Spontaneous HBsAg seroclearance may occur after HBeAg seroconversion. A recent 11-year follow-up study in 1,965 asymptomatic anti-HBe positive subjects [age = 16–76 years (median = 34)] showed an annual HBsAg seroclearance rate of 1.2%. The cumulative HBsAg seroclearance rate was 8% at 10 years, increased disproportionately to 25% at 20 years, and 45% at 25 years of follow-up [26] HBsAg seroclearance usually confers excellent prognosis [27]. However, HCC may still occur, although at a very low rate if cirrhosis has already developed before HBsAg seroclearance [27, 28].

Hepatitis B virus has been classified into at least eight genotypes on the basis of an intergroup divergence of 8% or more in the complete genome nucleotide sequence. Subtypes are identified within some genotypes, but their clinical significance remains to be determined. Each genotype has its distinct geographical and ethnic distribution, worldwide and within the Asian-Pacific region. HBV genotypes B and C are prevalent in East and South-East Asia, the Pacific Islands, and Pakistan, whereas HBV genotypes D and A are prevalent in India and genotype A in the Philippines. HBV genotype D is also found in the Pacific Islands. HBV genotypes B and C are prevalent in highly endemic areas where perinatal or vertical transmission plays an important role in spreading the virus, whereas genotypes A, D, E, F, and G are frequently found in areas where the main mode of transmission is horizontal. The clinical significance and virologic characteristics of HBV genotypes have only been reliably compared between genotypes B and C or genotypes A and D. In general, genotype B is associated with less progressive liver disease than genotype C, and genotype D has a less favorable prognosis than genotype A [29]. A recent study in 1,536 Alaskan natives with chronic HBV infection has shown that the median age for HBeAg clearance was less than 20 years for genotypes A, B, D, and F, but more than 40 years for genotype C, and that patients with genotypes C and F have significantly more frequent HBeAg reversion and higher risk of HCC [30]. Several studies have shown that genotype B is associated with spontaneous HBeAg seroconversion at a younger age, less active liver disease, slower progression to cirrhosis, and less frequent development of HCC than genotype C [10, 22, 29–34]. HBV genotype B has been shown to induce a greater Th₁ and lesser Th₂ response than genotype C, leading to a higher chance of HBeAg seroconversion [35]. A study from India indicated that genotype D is more often associated with HBeAg-negative chronic HBV infection and more severe diseases and may predict the occurrence of HCC in young patients [36]. It has also been shown that recombinant genotypes lead to more severe disease.

Due to the spontaneous error rate of viral reverse transcription, naturally occurring HBV mutations arise during the course of infection under the pressure of host immunity or specific therapy. Several HBV strains including mutations in precore, core promoter, and deletion mutation in pre-S/S genes have been reported to be associated with the pathogenesis of fulminant or progressive liver disease, including cirrhosis and HCC [29]. Patients harboring HBV genotype C have a higher HBV-DNA level, higher frequency of pre-S deletions, higher prevalence of core promoter A1762T and/or G1768/A mutations, and A1762T/G1764A double mutations than patients infected with HBV genotype B and have a significantly higher chance of developing HCC [21, 25, 29, 34, 37, 38]. A recent study revealed that a complex mutation pattern rather than a single mutation was associated with disease progression [38]. The role of these naturally occurring HBV mutations in the pathogenesis of liver disease progression requires further studies.

Concurrent infection with other virus(es)

Hepatitis B virus, hepatitis C virus (HCV), hepatitis delta virus (HDV), and human immunodeficiency virus (HIV) share similar transmission routes. Therefore, concurrent infection with these viruses may occur and complicate the natural course of chronic HBV infection. In general, concurrent infection with these viruses usually results in more severe and progressive liver disease and thus needs treatment [39].

Goals of treatment for chronic HBV infection

It is now clear that active HBV replication is the key driver of liver injury and disease progression, thus sustained viral suppression is of paramount importance [40]. Therefore, the primary aim of treatment for chronic HBV infection is to permanently suppress HBV replication. This decreases infectivity and pathogenicity of the virus. Reducing the pathogenicity of the virus results in reduced hepatic necroinflammation. Clinically, the short-term goal of treatment is to achieve “initial response” in terms of HBeAg seroconversion and/or HBV-DNA suppression, ALT normalization, and prevention of hepatic decompensation; to ensure “maintained/sustained response” to reduce hepatic necroinflammation and fibrosis during/after therapy. The ultimate long-term goal of therapy is to achieve “durable response” to prevent hepatic decompensation, reduce or prevent progression to cirrhosis and/or HCC, and prolong survival.

Currently available treatments

Currently, IFN-α, lamivudine, adefovir, entecavir, telbivudine, and PegIFN-α2a have been licensed globally. Clevudine has been approved in Korea. Thymosin α₁ has also been approved in many countries in Asia.

IFN-based therapy

Conventional IFN

Conventional IFN-α has been used for the treatment of chronic HBV infection for more than two decades. IFN-α has a dual mode of action: antiviral and immunomodulatory. Early controlled studies have shown that a 4- to 6-month course of conventional IFN-α at a dose of 5 MU daily or 10 MU 3 times weekly achieved HBeAg loss in approximately 33% of HBeAg-positive patients in comparison with 12% of controls. Smaller dosage (5–6 MU 3 times weekly) has been used in Asian patients with similar efficacy. Treatment of longer than 12 months’duration may improve the rate of HBeAg seroconversion, particularly in those with lower HBV-DNA levels after 16 weeks of treatment. Retreatment of relapsed patients with IFN-α showed a response rate of 20–40%. When HBeAg seroconversion to anti-HBe is achieved, it is sustained in more than 80% of cases [41].

Children with chronic HBV infection and high ALT levels respond to IFN-α at rates similar to adults. A recent study involving 108 Italian children, however, showed that there was no significant difference in the overall long-term outcomes in IFN-α-treated and untreated patients, and no patient developed end-stage liver disease or HCC during 12 years (range = 5–23) of follow-up [42].

The HBeAg seroconversion rate is lower in patients with lower baseline ALT levels. This rate may be improved by corticosteroid priming before IFN therapy. The recovery of immune function following steroid withdrawal may result in ALT flares and enhance the effect of IFN. A meta-analysis involving 790 patients in 13 randomized trials showed that this approach was associated with significantly more frequent loss of HBeAg (P = 0.03) and HBV-DNA (P = 0.0008), particularly in Asian patients with lower ALT levels and when lower dose of corticosteroid was used [43]. Severe adverse effects have been reported with this approach in patients with advanced liver disease.

IFN-α therapy resulted in end-of-treatment biochemical and virological response in up to 90% of patients with HBeAg-negative hepatitis. Sustained response rates, however, were disappointing: 10–15% with 4–6 months of treatment; 22% with 12 months of treatment; and 30% with 24 months of treatment. A study from Taiwan showed that 6–10 months’ IFN therapy in HBeAg-negative patients had an end-of-treatment response of 57% (vs. 18% of controls) and 6 months’ sustained response of 30% (vs. 7%).

Long-term follow-up studies suggest that IFN-induced HBeAg seroconversion is durable, increases over time, results in less cirrhosis development [17], better overall survival, and survival free of hepatic decompensation [17, 44]. The incidence of HCC is also lower in treated patients, especially among responders [17, 40, 45]. High HBsAg loss rate observed after IFN-α therapy in Italian patients was not observed in Asian patients [17].

A meta-analysis involving 1,505 cirrhotic patients in seven trials favored IFN therapy in reducing HCC, although significant heterogeneity of the trials made these results less conclusive. However, it has been shown that IFN-α therapy in compensated cirrhotic patients is safe and even more effective than noncirrhotic patients [20]. This finding suggests that the benefit of IFN therapy in reducing HCC might be evident upon longer follow-up. A subgroup analysis in a recent long-term follow-up study did show that HCC incidence was reduced significantly in IFN-treated cirrhotic patients [17].

The main advantage of IFN-α therapy is that a course of finite duration may achieve sustained off-therapy response in a proportion of patients with both HBeAg-positive and HBeAg-negative chronic HBV infections. However, IFN treatment is usually associated with adverse effects, especially influenza-like symptoms, fatigue, neutropenia, thrombocytopenia, and depression. These are usually tolerable, but may require dose modification and premature cessation of treatment [41].

PegIFN-α

In an Asian study, a 24-week course of weekly PegIFN-α2a (40 kD) gave a higher HBeAg seroconversion (33% vs. 25%; P > 0.05) and combination response (HBeAg loss, HBV-DNA < 5 × 10⁵ copies/ml, and normal ALT) rate (24% vs. 12%; P = 0.036) at 6 months after the end of treatment than conventional IFN-α2a. This benefit was noted even in patients with a rather low likelihood of response to conventional IFN [46]. In large-scale phase III international multicenter studies involving 814 HBeAg-positive patients (>85% were Asians) and 564 HBeAg-negative patients (>60% were Asians), PegIFN-α2a (40 kD) monotherapy 180 μg once weekly for 48 weeks resulted in ALT normalization in 41% and 59% patients, HBV-DNA < 80 IU/ml (<400 copies/ml) in 14% and 19% patients, and HBsAg seroclearance in 3% and 3% patients, respectively. HBeAg seroconversion occurred in 32% and HBV-DNA levels were less than 20,000 IU/ml (< 10⁵ copies/ml) in 32% of HBeAg-positive patients, whereas HBV-DNA levels remained less than 4,000 IU/ml (2 × 10⁴ copies/ml) in 43% of HBeAg-negative patients when assessed 6 months after cessation of therapy. PegIFN-α2a was found to be superior to lamivudine, with respect to sustained HBeAg seroconversion and HBV-DNA suppression, in both HBeAg-positive and HBeAg-negative patients [47, 48]. These responses were sustained in up to 90% when assessed 3 years after end of therapy [49]. The 6 months’ sustained HBeAg seroconversion rate is similar to that obtained after 6 months’ therapy in an earlier phase II study. A 4-arm head-to-head randomized control study using 90 and 180 μg of PegIFN-α2a for 6 or 12 months is ongoing.

Several studies using PegIFN-α2b showed similar efficacy [41]. One study showed that peg-IFNα2b was safe and effective in patients with advanced fibrosis or cirrhosis as those with early stage of fibrosis [50]. Patients with chronic HBV infection who are lamivudine refractory and those who are lamivudine naïve respond similarly to PegIFN-α2b therapy [51].

High baseline ALT, low baseline HBV-DNA and HBeAg levels, and high-grade necroinflammatory activity are predictors of response to IFN and PegIFN-α therapy [41]. ALT flares followed by decrease in HBV-DNA levels and decline of HBeAg levels during PegIFN-α therapy were predictors of response at the end of follow-up [52, 53]. Baseline ALT, baseline HBV-DNA, and HBV genotype influence the combined response (ALT normalization and HBV-DNA < 4 × 10³ IU/ml [<2 × 10⁴ copies/ml]) at 24 weeks posttreatment in patients with HBeAg-negative chronic HBV infection with a 48-week course of PegIFN-α2a with or without lamivudine [54].

Studies using conventional IFN therapy have shown that patients with HBV genotypes A and B infection have a higher HBeAg seroconversion rate than patients with HBV genotypes C and D infection, respectively [29]. These findings were confirmed by recent studies using PegIFN, where HBeAg seroconversion occurred more often in patients with HBV genotypes A (40%–47%) and B (30%–44%) than those with HBV genotypes C (28%–30%) and D (20%–25%) infection [46, 55]. Significantly better response in genotype B infected patients (31% vs. 17.5% of genotype C infected; P < 0.05) was observed in earlier 6-month PegIFN-α2a trial [46] but not in the recent 12-month trial [47]. This may suggest that longer PegIFN therapy may be required to enhance the response of the patients with more difficult-to-treat situations such as in genotype C or D infected patients.

IFN combination with other agents

Studies using IFN-α or PegIFN-α and lamivudine combination in comparison with IFN-α or PegIFN-α or lamivudine monotherapy in HBeAg-positive and HBeAg-negative patients showed that combination therapy had greater on-treatment viral suppression and higher rates of sustained response than lamivudine monotherapy, but there was no difference in sustained off-treatment response when compared with IFN-α or PegIFN-α monotherapy [47, 48]. To date, there has been no large clinical trial that confirms the benefits of PegIFN-α plus nucleoside or nucleotide analogue therapy over PegIFN-α monotherapy [41].

Sequential therapy with lamivudine 100 mg daily for 4 weeks followed by PegIFN-α2b 1.0 μg/kg per week for a further 24 weeks (n = 36 patients) compared with PegIFN-α2b monotherapy for 24 weeks (n = 27 patients) in patients with HBeAg-positive chronic HBV infection showed a significantly higher rate of HBV-DNA undetectability (<4,700 copies/ml) (50% vs. 14.8%) and higher rates of HBeAg clearance (38.9% vs. 14.8%) at 6 months posttherapy [56].

A randomized controlled trial in 96 patients showed that lymphoblastoid IFN 5 MU in combination with thymosin α-1 1.6 mg 3 times weekly for 24 weeks increased HBeAg loss 1 year after the end of treatment with marginal significance in comparison with IFN monotherapy (45.8% vs. 28%; P = 0.067) [57].

IFN-based therapy: overall conclusions

The advantages of IFN-based therapy include finite duration of treatment with modest response, long-term benefit, and no resistance. PegIFN may eventually replace conventional IFN because of higher efficacy and more convenient once weekly administration. The optimal duration (6 vs. 12 months) of PegIFN therapy in HBeAg-positive patients is under study. Perhaps patients infected with HBV genotype C or D may require longer treatment.

Other immunomodulating agents

Thymosin α-1

A few studies have evaluated the efficacy of thymosin α-1 (Tα₁), an immunomodulating agent that enhances the Th₁ immune response, natural killer T cells, and CD8⁺ CTL activity against HBV. A Taiwanese study showed that therapy with subcutaneous Tα₁ 1.6 mg twice weekly for 6 months resulted in a significantly higher HBeAg seroconversion rate (40% vs. 9% in controls) when assessed 12 months after the end of therapy [58]. A 6-month therapy in Chinese HBeAg-negative patients also showed a response rate of 42% (11/26) [59]. Patients infected with genotype B HBV showed a significantly better response (52%) than patients infected with genotype C (24%) HBV [60]. A response rate of 22% was also observed in a Japanese study involving 316 patients, mostly infected with genotype C HBV [61]. A meta-analysis including 353 patients from five trials showed that the odds ratio for virological response to Tα₁ at the end of treatment, 6, and 12 months posttreatment were 0.56 (0.2–1.52), 1.67 (0.83–3.37), and 2.67 (1.25–5.68), respectively, with a significantly increasing virological response over time after discontinuation of thymosin therapy [62]. The number of patients included in thymosin α-1 trials was relatively small in comparison with recent trials using PegIFN or nucleoside analogues. More well-designed, large-scale studies are needed to confirm its efficacy. The main advantages of thymosin α-1 are fixed duration of therapy and minimal side effects.

Therapeutic vaccines

Various therapeutic vaccines were used in an attempt to restore the virus-specific host immune response. However, none of them demonstrated sufficient clinical efficacy. In a recent open-label controlled study, 195 HBeAg-positive patients were randomized to receive 12 doses of HBsAg with AS02 adjuvant candidate vaccine plus lamivudine daily for 52 weeks or lamivudine daily alone. Despite induction of a vigorous HBsAg-specific lymphoproliferative response, cytokine production, and anti-HBs antibodies, therapeutic vaccination with an adjuvanted HBsAg vaccine combined with lamivudine did not demonstrate superior clinical efficacy than lamivudine alone [63].

Direct antiviral agents

Lamivudine, adefovir, entecavir, and telbivudine are highly effective in inhibiting HBV replication and have been approved worldwide for the treatment of chronic HBV infection. These agents are prodrugs and need intracellular activation before they can exert their therapeutic action. The efficacy of treatment with these four drugs is compared in Table 2. Clevudine has been approved only in Korea. Tenofovir and other new nucleoside analogues are in various stages of appraisal.

Lamivudine

Lamivudine, an l-nucleoside analogue, at a daily dose of 100 mg is effective in suppressing HBV-DNA with ALT normalization and histologic improvement in both HBeAg-positive and HBeAg-negative patients [64]. HBeAg seroconversion is achieved in 35–65% of HBeAg-positive patients after 5 years of therapy; the rate being proportional to the levels of ALT prior to treatment and highest in patients with ALT levels over 5 times ULN. This suggests that patients with a more vigorous immune response to HBV respond better to the direct antiviral effect of lamivudine [65]. The HBeAg seroconversion rate is similar in patients with HBV genotype B or C infection. Children treated with lamivudine for 1 year with dosages adjusted for body weight (3 mg/kg) showed similar response to adults, and the drug has been found to be safe during 3 years of continuous therapy [66]. In the absence of HBeAg seroconversion, hepatitis flares may occur if lamivudine is stopped. Lamivudine can be stopped after HBeAg seroconversion. Sustained HBeAg seroconversion to anti-HBe occurs in ∼80% of patients after cessation of lamivudine therapy [64]. The durability of response is particularly low in patients with genotype C HBV infection, older patients, and if treatment is maintained for less than 4–8 months after HBeAg seroconversion [67]. Acute flares of hepatitis may occur in patients with the reappearance of HBeAg and detectable HBV-DNA (HBeAg seroreversion). In pediatric patients, the durability of HBeAg seroconversion increased from 82% to more than 90% in those who had received lamivudine for 52 weeks and more than 2 years, respectively [66].

The antiviral and therapeutic impact of lamivudine in patients with HBeAg-negative chronic HBV infection is similar to that in HBeAg-positive patients. Sustained antiviral response is obtained in only 15–20% of cases after 1 year of treatment [63]. Lamivudine therapy for 6–12 months resulted in 81% maintained virologic response in a study involving 85 Taiwanese HBeAg-negative patients with pretreatment ALT > 5 times ULN, and sustained virologic response was observed in 39% of these patients 12 months after stopping lamivudine therapy [68]. In a study involving 50 Chinese-Canadian patients, 2-year treatment with lamivudine resulted in maintained virologic response in 37 (74%) patients. Therapy was stopped in these 37 patients when undetectable HBV-DNA and normal ALT levels were documented on three separate occasions at least 3 months apart. Relapse was noted in 50% of these patients (86% of them infected with genotype C HBV) 1 year after cessation of therapy [69]. In a Hong Kong Chinese study, 2-year lamivudine treatment in 89 HBeAg-negative patients showed a maintained complete response (normal ALT and HBV-DNA < 2 × 10³ IU/ml [<10⁴ copies/ml]) rate of 56% and a sustained response rate of 26% 6 months posttherapy [70]. These three studies show that about 50% of the patients who achieved maintained response have sustained off-therapy response.

Lamivudine is well tolerated and is safe for use, even in patients with decompensated cirrhosis [20, 64]. Long-term therapy in viremic patients with advanced fibrosis or cirrhosis delays clinical progression by reducing the rate of hepatic decompensation and HCC development, even in patients with low or normal ALT levels [71].

After 6–9 months of lamivudine therapy, viral breakthrough may occur following the emergence of HBV mutations that are resistant to lamivudine. These HBV-variant species have mutations in the YMDD motif of the polymerase gene (rtM204I and rtM204V with or without rtL180M). The incidence increases with increasing duration of therapy, up to 70% among patients treated with lamivudine continuously for 5 years. Other factors associated with the emergence of rtM204 I/V include baseline HBV-DNA, ALT, and/or hepatitis activity, sex and body mass index, and initial virologic response. Recent studies have shown that detectable HBV-DNA at month 6 was associated with higher resistance rate [64, 70]. The emergence of genotypic resistance is usually followed by a more than 1 log increase of HBV-DNA from nadir (viral breakthrough). With continuation of therapy, ALT elevation (biochemical breakthrough) occurs in more than 90% of patients after documented viral breakthrough [72, 73]. Hepatitis flares may develop and can occasionally result in hepatic decompensation [74]. New and distinct mutants may be selected during continuing lamivudine therapy and elicit further hepatitis flares [75]. The initial histologic improvement may be reversed in patients with rtM204 I/V [76]. The benefit of long-term therapy in preventing disease progression in patients with advanced fibrosis or cirrhosis also decreased after emergence of rtM204 I/V [71, 77]. The decision on long-term lamivudine therapy must therefore take into consideration the potential clinical benefits, possible risk associated with drug-resistant mutations, and the durability of response after stopping therapy.

Combination therapy using lamivudine with adefovir, telbivudine, IFN, or PegIFN has not demonstrated significant efficacy advantage in controlled trials. These combinations, however, lower the rates of resistant mutations than lamivudine monotherapy [64]. A pilot study in 30 Taiwanese patients showed that a short course of prednisolone priming enhanced Th₁ response and efficacy subsequent to lamivudine therapy [78]. “Lamivudine pulse” therapy has resulted in sustained HBeAg and HBV-DNA loss in 31% of 27 patients with chronic HBV infection and normal ALT levels [79]. This approach could be dangerous in patients with advanced fibrosis or cirrhosis.

Adefovir dipivoxil

Adefovir dipivoxil is a synthetic acyclic adenine nucleotide analogue. It is a potent inhibitor of HBV reverse transcriptase in the wild-type HBV as well as in lamivudine-, telbivudine-, and entecavir-resistant mutants.

Two large international multicenter double-blinded, placebo-controlled studies have shown that oral adefovir dipivoxil 10 mg daily for 48 weeks is effective in HBV-DNA suppression, ALT normalization, and histologic improvement in patients with both HBeAg-positive and HBeAg-negative chronic HBV infections. In HBeAg-positive patients, HBeAg loss and HBeAg seroconversion increased from 12% (control 6%) after 1 year to 40% after 3 years’ therapy [64]. Up to 240 weeks of adefovir therapy in naïve HBeAg-negative patients resulted in HBV-DNA < 200 IU/ml in 67% of patients, ALT normalization in 69% of patients, improvement in necroinflammation in 83% of patients, and regression of fibrosis in 73% of patients, respectively [80]. Response to adefovir was similar in Asian and Caucasian patients. Integrated analysis from all phase III clinical trials showed that HBV genotype does not influence virologic response to adefovir dipivoxil regardless of HBeAg serostatus [64].

The safety profile of 10 mg of adefovir dipivoxil was similar to placebo in patients with compensated chronic HBV infection. Renal laboratory abnormalities reported with 30 mg of adefovir dipivoxil were not observed with 10 mg of dosage during the 1-year study period. Reversible increase in serum creatinine of more than 0.5 mg/dL (maximum 1.5 mg/dl) was reported in 3% of patients when therapy was extended to 5 years [80]. Most patients with decompensated chronic HBV infection, including patients with pre- and post-liver transplant, have some degree of underlying renal insufficiency. Studies on these patients showed increases in serum creatinine levels by 0.5 mg/dl or more from baseline in 16% of them by week 48, 31% by week 96, and 1% required discontinuation due to renal failure [64].

Sequenced RT domain of HBV-DNA polymerase identified rtN236T and rtA181T/V mutations with decreased susceptibility to adefovir dipivoxil in patients on adefovir therapy for more than 1 year. The overall incidence of adefovir-resistant mutation is low. Integrated incidence rate was 0%, 3%, and 11%, 18%, and 29% at the end of each successive year of therapy in HBeAg-negative patients. HBV-DNA > 200 IU/ml (10³ copies/ml) at week 48 were predictive of the emergence of adefovir-resistant mutations (49% vs. 6% of those <10³ copies/ml) during 192 weeks of adefovir treatment [80]. Adefovir dipivoxil-resistant rtN236T mutant remains susceptible to l-nucleoside analogues lamivudine, emtricitabine, telbivudine, and entecavir in vitro and may argue for their combination in therapy. The rtA181T/V HBV is resistant to adefovir and all the l-nucleoside analogues, but sensitive to entecavir [73].

One year of adefovir dipivoxil monotherapy or in combination with lamivudine reduced serum HBV-DNA levels in most patients with lamivudine-resistant mutants (median reduction = 3.6 log₁₀ to 4.6 log₁₀ copies/ml). Switching to adefovir dipivoxil monotherapy in lamivudine-resistant patients appeared effective and safe, even in patients with liver decompensation [81]. The rate of resistant mutation, however, is higher under such circumstances (up to 30% by the end of year 2) than adefovir monotherapy in lamivudine-naïve patients [82]. A 3-year study of 145 lamivudine-resistant HBV-infected patients showed that add-on adefovir led to undetectable HBV-DNA in 80% and normal ALT in 84% of patients, and none developed virologic and clinical breakthrough during 12–74 months of therapy [83]. Add-on adefovir in patients with HBV-DNA >10⁷ copies/ml is associated with insufficient virologic response [82, 84], and should therefore be instituted as soon as genotypic resistance is detected and before the serum HBV-DNA levels increase to a level too high to be suppressed successfully [85].

The high genetic barrier to resistance and the ability to suppress most lamivudine-resistant mutants (rtM204 V/I) makes adefovir dipivoxil an attractive drug. Renal toxicity is rare with the dose of 10 mg and few patients had significant elevation of serum creatinine levels of more than 0.5 mg/dl in clinical studies. Caution must be exercised in treating patients with renal impairment.

Entecavir

Entecavir is a cyclopentyl guanosine analogue with potent selective inhibition of the priming, DNA-dependent synthesis, and reverse transcription functions of HBV polymerase. In a viral kinetic study comparing entecavir to adefovir in HBeAg-positive patients with high viral load, entecavir showed significantly greater HBV-DNA reduction as early as day 10, HBV-DNA reduction was −6.23 log versus −4.42 log at week 12 and −7.28 log vs. −5.08 log at week 48, respectively [64]. Pivotal phase III randomized lamivudine controlled trials showed that 1-year entecavir (0.5 mg/day) is superior to lamivudine in reducing HBV-DNA in both HBeAg-positive (−6.9 log vs. −5.4 log; P < 0.0001; HBV-DNA < 300 copies/ml in 67% vs. 36%) [86] and HBeAg-negative patients (−5.0 log vs. −4.5 log; P < 0.001; HBV-DNA < 300 copies/ml in 90% vs. 72%) [87]. HBeAg seroconversion rate was 21% (68% in patients with pretherapy ALT > 5 times ULN). Extending entecavir therapy to 96 weeks for partial responders at week 48 resulted in an increase in the rate of HBV-DNA < 60 IU/ml (<300 copies/ml) to 74%, ALT normalization increased to 79%, and a cumulative HBeAg seroconversion rate of 31% [88]. The corresponding rate was 91%, 86%, and additional 16%, respectively, after extending entecavir therapy to 192 weeks [64].

Switching to entecavir monotherapy (1 mg/day) is initially effective in lamivudine-resistant patients (−5.11 log vs. −0.48 log reduction in lamivudine controls; P < 0.001) and safe without risk of ALT flares. HBeAg loss was documented in 10% of lamivudine-resistant HBeAg-positive patients (vs. 3% in lamivudine controls; P = 0.028) [89]. Entecavir has a high genetic barrier, and drug resistance requires at least three mutations including rtL180M and rtM204 V, plus a mutation at one of the following codons: rtT184, rtS202, and/or rtM250 [73]. Therefore, entecavir therapy in lamivudine-refractory patients is associated with a higher entecavir resistance rate [88, 89]. The cumulative probability of a virologic breakthrough from entecavir resistance through 4 years is at least 0.8% in lamivudine-naïve patients and 39.5% in lamivudine-refractory patients [64, 73].

Telbivudine

Telbivudine is an orally bioavailable l-nucleoside with potent and specific anti-HBV activity. In clinical trials, telbivudine gave more potent HBV suppression than lamivudine or adefovir. In the phase III randomized lamivudine controlled trial in 1,371 patients (446 HBeAg negative, 1,040 Asians), significantly greater HBV-DNA reduction with telbivudine 600 mg/day was evident by week 12 (−5.71 log vs. −5.42 log in HBeAg-positive patients and −4.36 log vs. −4.08 log in HBeAg-negative patients). HBV-DNA reduction persisted through week 52 with greater histologic response, larger proportions of patients with undetectable HBV-DNA (60.0% vs. 40.4% in HBeAg-positive patients and 88.3% vs. 71.4% in HBeAg-negative patients), and less resistance (5.0% vs. 11% in HBeAg-positive patients and 2.3% vs. 10.7% in HBeAg-negative patients) than lamivudine. The C-domain mutation rtM204I and the B-domain mutation rtA181T/V are the common mutations associated with telbivudine resistance. The HBeAg seroconversion rate was similar between telbivudine- and lamivudine-treated patients. The study also showed that 41% of HBeAg-positive patients with undetectable HBV-DNA at week 24 underwent HBeAg seroconversion by week 52 versus 4% for patients with HBV-DNA > 2,000 IU/ml at week 24. Only 1% of HBeAg-positive patients with undetectable HBV-DNA and 2% of patients with HBV-DNA < 200 IU/ml (<10³ copies/ml) at week 24 developed drug resistance by week 52, whereas 11% of patients with HBV-DNA > 10⁴ copies/ml at week 24 became resistant at week 52. The corresponding figures for drug resistance in HBeAg-negative patients was 0%, 6%, and 30%, respectively [90]. Two-year telbivudine therapy was significantly superior to lamivudine in both HBeAg-positive and HBeAg-negative patients for all direct measures of antiviral effect, including serum HBV-DNA reduction from baseline (−5.7 vs. −4.4 in HBeAg-positive patients and −5.0 vs. −4.2 in HBeAg-negative patients), PCR negativity (56% vs. 39% in HBeAg-positive patients and 82% vs. 57% in HBeAg-negative patients), HBeAg seroconversion in patients with ALT ≥2 times ULN (36% vs. 27%; P = 0.022), and viral resistance (25% vs. 40% in HBeAg-positive patients and 11% vs. 26% in HBeAg-negative patients; P < 0.001). Week 24 HBV-DNA levels also emerged as a strong predictor of week 104 efficacy outcomes [64].

In another 1-year randomized adefovir controlled trial in 135 HBeAg-positive patients, significantly greater HBV-DNA reduction with telbivudine was evident at week 24 (−6.30 log vs. −4.97 log, undetectable HBV-DNA in 39% vs. 12%; ≥3 log copies/ml in 50% vs. 78% of patients) and week 52 (−6.56 log vs. −5.99 log; undetectable HBV-DNA 60% vs. 40% of patients). The HBeAg seroconversion rate at week 52 of treatment was also higher in telbivudine-treated patients than in adefovir-treated patients (28% vs. 19%; P = 0.34). A predictive analysis of response showed that week 24 serum HBV-DNA < 200 vs. ≥ 200 IU/ml (<3 log₁₀ vs. ≥3 log₁₀ copies/ml) correlated with undetectable HBV-DNA (95% vs. 24%) and HBeAg seroconversion rate (41% vs. 14%) at year 1. Patients with viral breakthrough at year 1 had HBV-DNA > 200 IU/ml (>3 log₁₀ copies/ml) at week 24 [91].

Increase in creatine kinase levels was observed more frequently in recipients of telbivudine, of whom 7.5% (vs. 3.1% in lamivudine-treated controls) had grade 3 or 4 elevation (a level of >7 times ULN). Two-thirds of grade 3 or 4 creatine kinase elevations decreased spontaneously to grade 2 or lower during continued treatment. Symptomatic myopathy was reported in 1 patient after 11 months of telbivudine therapy, and resolved over a period of 9–12 months after stopping telbivudine [90].

Other emerging direct antivirals

Clevudine is a pyrimidine analogue with potent and sustained antiviral activity against HBV. Clevudine 30 mg/day for 24 weeks resulted in end-of-treatment HBV-DNA reduction of 5.10 log₁₀ copies/ml, undetectable HBV-DNA in 59%, ALT normalization in 68.2%, and HBeAg loss in 24% of 243 HBeAg-positive patients [92]. The same regimen resulted in end-of-therapy HBV-DNA reduction of 4.25 log₁₀ copies/ml, undetectable HBV-DNA in 92%, and ALT normalization in 75% of 86 HBeAg-negative patients; HBV suppression sustained as HBV-DNA was 3.11 log₁₀ copies/ml, with undetectable HBV-DNA in 80.3% and normal ALT in 70.5% of patients 24 weeks after stopping clevudine [93]. No significant difference was reported in these efficacy parameters among the patients with different pretreatment ALT levels [92]. Substitutions rtA181A/T and rtA181T without viral breakthrough were detected in 5 (2.7%) of the 182 HBeAg-positive patients [92], but none in HBeAg-negative patients [93].

Tenofovir disoproxil fumarate is an acyclic adenine nucleotide that exerts a strong and early suppression of HBV with or without lamivudine-associated mutations. It has been approved for use in the treatment of HIV infection. Tenofovir 300 mg/day is more potent than adefovir 10 mg/day but without comparable renal toxicity. Clinical studies have shown that administration of tenofovir 300 mg daily has stronger antiviral effect against lamivudine-resistant HBV than adefovir 10 mg daily [94, 95]. Phase III randomized adefovir controlled trial in HBeAg-positive patients has shown that tenofovir has better efficacy than adefovir with respect to histologic improvement (74% vs. 68%), HBV-DNA reduction to less than 400 copies/ml (76% vs. 13%; P < 0.001), ALT normalization (69% vs. 54%; P = 0.02), and HBeAg seroconversion (21% vs. 18%; P = 0.36). Tenofovir also achieved combined virologic and histologic response in a higher proportion of HBeAg-negative patients (71% vs. 49%; P < 0.001) [96]. Tenofovir appears to be a very promising drug and is likely to get approval for use in the treatment of chronic HBV infection and replace adefovir in the near future.

Therapy with direct antiviral agent(s): overall conclusions

The successive generation of nucleos(t)ide analogues has improved potency and raised genetic barrier to resistant mutations (Table 2). Although there is no head-to-head comparison among these 4 drugs, the results of published pivotal trials suggest that entecavir is the most potent agent, followed by telbivudine, lamivudine, and adefovir in terms of HBV-DNA reduction during a 1-year treatment period. Histologic improvement and documented regression of advanced fibrosis and cirrhosis among the responders is an important achievement. Reduction in the progression of disease and HCC development after 3 years of lamivudine therapy for patients with advanced fibrosis is a proof for therapeutic aim. Increased antiviral potency of these drugs, however, does not correlate with increase in HBeAg loss or HBeAg seroconversion. Resistance is a major concern during long-term therapy. The incidence at 1 and 2 years is highest with lamivudine, followed by telbivudine, then adefovir and tenofovir, and almost none with entecavir. In choosing a direct antiviral agent to initiate therapy, resistance profile is a crucial factor to consider other than the potency and cost. The “roadmap” concept for using on-treatment HBV-DNA level as a predictor for drug resistance may be useful when patients are treated with agents with high resistance rate [97]. Pharmacoeconomic studies would be helpful in individual countries in Asia-Pacific region because cost is one of the most important factors in the choice of drug for initial therapy [98].

Special groups of patients

Pregnancy

IFN-based therapy is contraindicated in pregnancy because of its antiproliferative effect. Among the direct antiviral agents, telbivudine is classified as category B drug (no risk in animal studies, but unknown in human), whereas lamivudine, adefovir and entecavir are classified as category C drugs (teratogenic in animal, but unknown in human) by the US FDA [99]. The stage of the mother’s liver disease and potential benefit of treatment must be weighed against the small risk to the fetus. IFN-based therapy is preferable in women in the childbearing age, and pregnancy is discouraged during IFN therapy. No firm recommendation can be made on the use of nucleosi(t)de analogues in the prevention of transmission from viremic mothers because of the lack of sufficient data and conflicting results with regard to efficacy and adverse events. Women with chronic HBV infection who become pregnant while on direct antiviral therapy can continue treatment with category B drugs [3].

Patients with concurrent HCV, HDV, or HIV infection

Patients with concurrent HCV, HDV, or HIV infections tend to have a higher incidence of cirrhosis, HCC, and mortality. Insufficient data exist to reach firm conclusions on the management of patients with HCV and/or HDV infections. However, it is generally agreed that the dominant virus should be identified before designing therapeutic strategy. If HBV is dominant, treatment should be aimed toward this virus. If HCV is dominant, standard IFN or PegIFN therapy in combination with ribavirin can achieve a sustained HCV clearance rate comparable to that in HCV monoinfection. Lamivudine is ineffective in patients with chronic HDV infection. Small randomized controlled trials using 3–9 MU of IFN for 3–24 months showed a biochemical and virologic response in up to 70% of the patients with chronic HDV infection. Sustained response was noted in less than 20% of patients. Higher doses of IFN-α (9 MU thrice weekly) for 12 months have been found to inhibit HDV-RNA, normalize ALT, and improve histology in patients with chronic HDV infection. ALT response sustained in 50% of the patients and the long-term outcomes and survival improved significantly even in patients with liver cirrhosis [100]. IFN in combination with lamivudine therapy tends to increase response rate compared with IFN monotherapy [101]. Two small studies using weight-based PegIFN-α2b (1.5 μg/kg weekly) for 6 and 12 months, respectively, showed discrepant results [100].

In patients with concurrent HIV infection and CD4⁺ counts of more than 500 cells/μL, treatment options include agents without anti-HIV activity: IFN, adefovir, and telbivudine. IFN-based therapy or adefovir is preferred because of the absence of resistance in the former and a low resistance profile in the later. Both lamivudine and tenofovir are active against HBV and HIV and can be used in combination as part of the highly active antiretroviral therapy (HAART) in patients who need both anti-HBV and anti-HIV therapies. In patients with low CD4 count and active liver disease, HBV should be treated first to avoid the risk of immune reconstitution syndrome that usually occurs with HIV treatment.

Patients with hepatic decompensation

Patients with hepatic decompensation should be considered for treatment because it may both improve their clinical status and even remove them from liver transplant lists. IFN does not benefit patients with Child’s B or C cirrhosis. Moreover, significant adverse effects due to serious bacterial infections and possible exacerbation of liver disease occur even with low doses. Lamivudine is well tolerated and results in clinical improvement or stabilization, especially in patients who have completed a minimum of 6 months’ treatment [102, 103]. Early treatment is recommended to improve outcomes. Selection of resistant mutants with resultant biochemical dysfunction, reduction in efficacy, and rapid clinical deterioration in this group of patients is a real concern with early treatment [20]. Adding adefovir to 128 lamivudine-resistant patients with decompensated cirrhosis and 196 lamivudine-resistant patients with recurrent HBV infection after liver transplantation was associated with 3–4 log reduction in serum HBV-DNA levels throughout the treatment period [104]. However, renal dysfunction is a potential problem in patients with hepatic decompensation. Close monitoring of renal function is, therefore, required if this drug is being used for such patients. Entecavir, telbivudine, and tenofovir are being evaluated as a primary treatment modality in patients with decompensated liver disease. Given the similar mechanisms of action and safety profile, the more potent entecavir, telbivudine, and tenofovir are anticipated to be more effective than or at least as effective as lamivudine in this clinical setting with lower or nearly no incidence of drug resistance and no problem with nephrotoxicity.

Pediatric patients

Children with elevated ALT levels respond to IFN and lamivudine in a similar manner to adults. A small study in children and adolescents (aged 2–17) showed that adefovir is generally well tolerated at a dose of 0.3 mg/day for those aged 2–11, and 10 mg for those aged 12–17 [105]. Newer agents such as PegIFN and other nucleos(t)ide analogues have not yet been studied, but are likely to be as effective in children as in adults with chronic HBV infection. Long-term safety and drug resistance are more important concerns in children than in adults. As already mentioned, recent long-term follow-up study showed that IFN therapy provided little benefit in comparison with untreated children [42]. Therefore, drug therapy is usually not recommended in pediatric patients because of the apparent lack of long-term benefits and attending risks of starting drug therapy, unless there is an absolute indication such as in the setting of ensuing or overt hepatic decompensation.

Patients on immunosuppression or chemotherapy

Reactivation of HBV replication with decompensation has been reported in 20–50% of patients with chronic HBV infection undergoing cancer chemotherapy or immunosuppressive therapy, especially those containing high-dose steroid regimen. Reactivation commonly occurs after the first 2–3 cycles of chemotherapy. High viral load at baseline is the most important risk factor for HBV reactivation [106]. HBV reactivation following transarterial chemoembolization was also observed in 34% of 83 patients with HCC [107]. Lamivudine is effective in the treatment of HBV reactivation in HBsAg-positive organ transplantation recipients and cancer patients undergoing chemotherapy, particularly if it is used preemptively. Prophylactic use of lamivudine within 1 week before the start and continued at least 12 weeks after end of chemotherapy, and when white blood cell count has normalized, can reduce HBV reactivation frequency and severity of flares and improve survival [106].

The impact of immunosuppressive therapy on patients with occult HBV infection is poorly characterized. In a recent study involving 244 consecutive HBsAg-negative lymphoma patients who received chemotherapy, 8 (3.3%) developed de novo HBV-related hepatitis and 3 with fulminant hepatic failure, following a 100-fold increase in serum HBV-DNA levels. These patients responded to lamivudine, but one died of hepatic failure. These findings suggest that even in an HBV endemic area, the occurrence of de novo HBV-related hepatitis after chemotherapy is low. It was suggested that HBsAg-negative patients, especially those receiving rituximab plus steroid-containing regimen, should be closely monitored to facilitate early commencement of nucleoside/nucleotide analogues [108].

Liver transplantation for chronic HBV infection

Liver transplantation has become a cost-effective treatment of liver failure and HCC with excellent 5-year survival. Improving economies and live related liver donation have allowed a rapid expansion of liver transplantation within the Asia-Pacific region where hepatitis B is the most common indication for both acute and chronic liver failure. Acute or chronic HBV infection accounts for most cases of acute liver failure in this region, whereas more than 80% of cases of chronic liver failure and HCC are caused by chronic HBV infection. Although HBV recurrence can be prevented in 60% of cases by high-dose (10,000 U/month) intravenous hepatitis B immunoglobulin (HBIg), this therapy is prohibitively expensive (US$50,000 per annum, lifelong) and is ineffective in transplant candidates with detectable HBV-DNA. Suppression of pretransplant viral replication significantly reduces the risk of posttransplant recurrence. In addition, viral suppression rescues some patients with decompensated cirrhosis, thereby removing the need for future transplant [103].

Antiviral therapy should be commenced in all potential liver transplant candidates with decompensated HBV cirrhosis and detectable HBV-DNA. However, posttransplant HBV recurrence may still occur despite antiviral prophylaxis and is usually due to lamivudine resistance [1]. Adefovir and entecavir are available for rescue therapy for lamivudine resistance, and de novo use of these agents may minimize the problems of drug resistance. Combination lamivudine/HBIg prophylaxis reduces recurrence rates of HBV infection to less than 5% and is associated with 5-year patient and graft survival rates of 85% and 80%, respectively. A recent long-term (median = 62 months) follow-up study involving 147 patients has shown that lamivudine plus low-dose intramuscular HBIg (400–800 U daily for 1 week, then monthly) appears as effective as lamivudine plus high-dose intravenous HBIg, but is less than 10% the cost (US$4,000) [109]. A recent study suggested that late HBIg substitution by adefovir (at least 12 months posttransplant) can prevent late HBV recurrence at less cost [110] In a prospective open-labeled study, lamivudine plus adefovir combination from the time of listing was well tolerated, prevented lamivudine resistance prior to transplant, rescued some patients from the need for transplantation, and prevented recurrent HBV infection following liver transplantation, regardless of baseline HBV-DNA status [111] Both studies demonstrate that lamivudine plus adefovir combination prophylaxis has similar efficacy to current lamivudine plus HBIg prophylaxis but without the cost and inconvenience of long-term monthly HBIg administration. There is emerging data that HBIg ± lamivudine prophylaxis can be replaced by lamivudine monotherapy 12 months posttransplant in certain “low-risk” patient groups. These include patients who were HBV-DNA negative (hybridization assay) before pretransplant lamivudine therapy was started and patients with sustained protective levels of anti-HBs production following posttransplant vaccination.

Adoptive immune transfer may result in de novo anti-HBs production in recipients of live related liver graft from an HBV immune donor. A liver from anti-HBc(+) donor carries a significant risk of de novo HBV infection if transplanted into an HBV-naïve recipient. This risk becomes negligible if the recipient receives long-term prophylaxis with either lamivudine or HBIg or if the recipient is seronegative for HBsAg but positive for anti-HBs.