Nucleos(t)ide Analogues Therapy for Chronic Hepatitis B in Taiwan: Short-Term Versus Long-Term
- First Online:
- Cite this article as:
- Peng, C. Curr Hepatitis Rep (2013) 12: 181. doi:10.1007/s11901-013-0173-7
- 56 Views
Chronic hepatitis B (CHB) remains an important health issue in Taiwan. The major candidates for antiviral therapy are HBeAg-positive and -negative CHB patients, and individuals with compensated or decompensated cirrhosis. Lamivudine, adefovir dipivoxil, entecavir, telbivudine, and tenofovir disoproxil fumarate (TDF) are the currently available nucleos(t)ide analogues (NA). Entecavir and TDF are recommended as the first-line agents for long-term use. The therapeutic endpoints recommended by the Asian Pacific Association for the Study of the Liver treatment guidelines have been adopted for clinical practice. Only a small number of patients qualify for short-term NA therapy; the majority of patients need long-term treatment. Current therapeutic endpoints confer off-therapy durability in approximately 50 % of patients. Patients with cirrhosis should receive long-term NA therapy. The use of serum HBsAg level as a guide for cessation of NA therapy requires further investigation.
KeywordsAdefovir dipivoxilChronic hepatitis BCompensated cirrhosisConsolidation therapyDecompensated cirrhosisEntecavirHBeAg-positiveHBeAg-negativeHBeAg seroconversionHBsAgHBV DNAHepatitis B virusHepatocellular carcinomaLamivudineNucleos(t)ide analogueTelbivudineTenofovir disoproxil fumarate
Chronic hepatitis B virus (HBV) infection has been a major health concern in Taiwan. Although the current rate of chronic HBV infection is <1 % among those receiving vaccination after the implementation of a universal newborn vaccination program against HBV in July 1984 , the prevalence of chronic HBV infection in the general population >29 years of age is estimated between 15 and 20 % . Currently in Taiwan, hepatocellular carcinoma (HCC) is the second leading cause of cancer death, and liver disease is the eighth leading cause of death. Timely antiviral therapy for chronic hepatitis B (CHB) is important to prevent disease progression and mortality. In recent years, nucleos(t)ide analogues (NA) have played an important role in the treatment of CHB. However, the optimal duration of therapy remains unclear. This review discusses some relevant data that impact current clinical practices for anti-HBV therapy in Taiwan.
The natural history of CHB typically consists of three phases: immune tolerant, immune clearance, and residual inactive, and in some patients, the fourth, reactivation phase . During the immune tolerant phase, there is a high level of HBV replication with no or minimal necroinflammation and disease progression in the liver. The immune clearance phase involves recurrent episodes of hepatic necroinflammation with elevations of serum alanine aminotransferase (ALT) levels. The immunologic event for these recurrent ALT elevations is believed to be cytotoxic T lymphocyte-mediated . With recurrent hepatic necroinflammation, the serum HBeAg and HBV DNA levels gradually decrease, which eventually leads to HBeAg seroconversion (SC). The estimated annual HBeAg SC rate is 2 to 15 %, depending on factors such as age, extent of hepatic necroinflammation, ALT level, and HBV genotype [5, 6]. The mean age of HBeAg SC is 30 to 35 years, with 90 % of cases occurring before age 40 years . After HBeAg SC, most patients enter an inactive HBsAg carrier state. A small proportion of patients (≈4 %) may revert back to HBeAg positivity with recurrent hepatitis activity . Inactive HBsAg carrier status is associated with a low risk for the development of cirrhosis or HCC. However, inactive HBsAg carriers may develop HBeAg-negative hepatitis at an annual incidence of 2 to 3 % due to the reactivation of HBV with mutations in the precore or basal core promoter region that abolish or downregulate HBeAg synthesis [7, 8]. Up to 20 % of inactive HBsAg carriers experience a relapse of hepatitis after 25 years of follow-up . A prospective study from Taiwan showed that male sex, genotype C, ALT levels five-fold higher than the upper limit of normal (ULN) during the HBeAg-positive immune clearance phase, and age at HBeAg SC >40 years are factors predictive of HBV reactivation following spontaneous HBeAg SC . Furthermore, a similar prospective study from Taiwan showed that age at HBeAg SC and relapse of hepatitis after HBeAg SC are two factors significantly associated with the development of cirrhosis (0.5 % per year) . The REVEAL-HBV study from Taiwan showed that serum HBV DNA >2,000 IU/mL at baseline is significantly associated with the risk of future development of cirrhosis or HCC in a concentration-dependent manner after adjustment for age, sex, smoking, alcohol, HBeAg status, and serum ALT level [11, 12]. Furthermore, recent studies from Taiwan showed that a serum HBsAg level >1,000 IU/mL increases the future risk of cirrhosis or HCC even in patients with baseline serum HBV DNA <2,000 IU/mL, especially in those inactive HBsAg carriers [13••, 14••].
A prospective study in 93 Taiwanese CHB patients who developed cirrhosis at a mean age of 43.6 years during follow-up showed that 30 % of patients were seropositive for HBeAg, and 73 % had serum HBV DNA >2,000 IU/mL at the time of cirrhosis detection . During a median follow-up period of 97 months, hepatitis flare, hepatic decompensation, HCC, and mortality occurred in 34 %, 13 %, 23 %, and 12 % of patients, respectively . Persistent HBeAg seropositivity was associated with a six-fold higher risk of hepatic decompensation, and trended toward the development of HCC during follow-up (P = 0.062) . The prognosis after the development of hepatic decompensation is usually poor, with a 5-year survival rate between 14 and 35 % [16, 17•].
Five oral NAs are approved for the treatment of CHB in Taiwan: lamivudine (LAM), adefovir dipivoxil (ADV), entecavir (ETV), telbivudine (LdT), and tenofovir disoproxil fumarate (TDF) [18••, 19, 20]. LAM is moderately potent for viral suppression and generally well tolerated; however, the major drawback is a high rate of drug resistant mutations in HBV, reaching 70 % after five years of treatment. ADV is less potent than LAM and suppresses HBV DNA slowly; however, there is limited drug resistance during the first two years of treatment. ADV also demonstrates efficacy against LAM-resistant HBV. However, drug resistance eventually develops in a significant proportion of patients (29 %) after five years of treatment in HBeAg-negative CHB patients. Three percent of patients experience renal toxicity at a daily dose of 10 mg after 4 to 5 years of treatment. ETV is a much more potent agent than LAM and demonstrates a very favorable resistance profile (only 1.2 % after five years of treatment in naïve patients). The tumors encountered in preclinical rodent studies have not been observed in human clinical studies with ETV. The safety profile of ETV is similar to LAM. LdT is also more potent than LAM. However, the low genetic barrier to resistance allows rapid development of drug resistance, reaching 25 % and 11 % in HBeAg-positive and HBeAg-negative patients, respectively, after two years of treatment. LdT is well-tolerated except for grade 3–4 creatine phosphokinase (CPK) elevations, which affect 13 % of patients after two years of treatment. Although these cases of CPK elevation were largely asymptomatic, rare cases of myopathy have been observed. TDF is structurally similar to ADV; however, it is much more potent than ADV and has a very favorable resistance profile. Drug resistant mutations have not been reported after 6 years of treatment . TDF is active against LAM-resistant HBV and also against ADV-resistant HBV. The renal toxicity of TDF is less severe than ADV, and <1.5 % of patients experience renal toxicity after 6 years of treatment . Loss of bone mineral density has been observed with long-term TDF treatment in patients with HBV and human immunodeficiency virus co-infection. This issue requires further study in CHB patients.
Despite differences in antiviral potency, the rates of HBeAg SC after ≥1 year of treatment with these NAs are similar, suggesting the decisive role lies with the host immune response for the successful achievement of HBeAg SC [18••, 19, 20]. It is believed that the cytotoxic T-cell response against HBV is essential in achieving sustained control of the virus once NA therapy is stopped . Without efficient immune control of the virus, HBV is likely to reactivate sometime after the antiviral agent is discontinued. Since NAs act by directly inhibiting HBV replication, long-term therapy may be necessary in cases where there is a lack of an effective immune system and resurgence of hepatitis activity poses a significant risk of disease progression. Long-term therapy leads to viral suppression, which will lead to biochemical remission, histologic improvement, prevention of complications, and decreased mortality [18••, 19, 20]. However, long-term therapy raises the concerns of efficacy, safety, drug resistance, and cost. Considering the favorable drug potency, resistance, and safety profiles, ETV and TDF have been recommended as the first-line agents for long-term use [18••, 19, 20]. Adherence to long-term NA therapy is important in preventing drug resistance and ensuring treatment efficacy. A recent study using a large pharmacy claims database from the USA showed an overall adherence rate of 87.8 % . Another study from Taiwan in treatment-naïve CHB patients receiving NA therapy over 3 years showed an overall adherence rate of 96.8 %, 96.8 % and 97.5 % in year 1, 2 and 3, respectively [RN Chien and WL Chuang, personal correspondence]. The adherence rates to long-term NA therapy deserve further study.
Treatment Guidelines and Reimbursement of NA Therapy in Taiwan
The Asian Pacific Association for the Study of the Liver (APASL) treatment guidelines recommend therapeutic endpoints for CHB patients receiving NA therapy [18••]. In HBeAg-positive patients, treatment can be stopped if HBeAg SC with undetectable HBV DNA levels has been maintained for at least 12 months. In HBeAg-negative patients, discontinuation of treatment can be considered if patients have received treatment for ≥2 years with undetectable HBV DNA levels documented on three separate occasions 6 months apart. However, the American Association for the Study of Liver Diseases (AASLD) and the European Association for the Study of the Liver (EASL) guidelines suggest HBsAg seroclearance as the therapeutic endpoint, which has rarely been achieved in HBeAg-negative patients undergoing NA therapy [19, 20]. Current APASL treatment guidelines suggest that compensated cirrhotic patients with serum HBV DNA >2,000 IU/mL or decompensated cirrhotic patients with any serum HBV DNA levels should receive life-long NA therapy [18••].
In October 2003, a national reimbursement program for antiviral therapy against HBV was launched in Taiwan. The duration of NA therapy reimbursed for HBeAg-positive, -negative, and decompensated patients had been 12 to 18 months and was extended up to 36 months in November of 2009. HBsAg-positive liver transplant patients have been reimbursed for long-term NA therapy since October 2006. Cirrhotic patients with serum HBV DNA >2,000 IU/mL have been reimbursed for long-term NA therapy since July 2010.
Short-Term Versus Long-Term Therapy in HBeAg-Positive CHB
In the Asian trial with HBeAg-positive CHB patients receiving LAM therapy, the baseline serum ALT level was a major determinant of HBeAg SC after one year of therapy. SC occurred in 64 % of patients with a baseline ALT >5 times ULN versus 5 % in patients with ALT <2 times ULN . The serum ALT level is often regarded as a marker of endogenous host immune response against HBV; therefore, this observation suggests that the strength of endogenous host immune response against HBV is an important factor associated with HBeAg SC during LAM therapy. Chien et al. further reported a 52 % rate of HBeAg SC in 186 patients after a median treatment duration of 8.2 (range 3–15 mo) months . This short-term therapeutic strategy was adopted to avoid the emergence of the YMDD mutation during LAM therapy when an effective therapy against YMDD mutant HBV was not readily available. No cases of YMDD mutation were encountered in this study. Tseng et al. reported that 33.6 % of 253 HBeAg-positive patients with baseline serum ALT levels >5 times ULN achieved HBeAg SC after 12 to 18 months of LAM therapy . Patients with baseline ALT levels >10 times ULN showed a higher HBeAg SC rate at 3 and 6 months (but not at 12 to 18 months of therapy) than patients with baseline ALT levels between 5 to 10 times ULN. Peng et al. further examined independent predictors of early HBeAg loss (defined as HBeAg loss during the first year of LAM therapy) in a cohort of 146 patients with baseline serum ALT levels >5 times ULN. Female gender and baseline ALT level >1,000 U/L were significant predictors of HBeAg loss after 3 months of therapy; only female gender was a significant predictor of HBeAg loss after 6 and 12 months of therapy . Thus, female patients and patients with higher baseline serum ALT levels are likely to achieve earlier HBeAg loss during LAM therapy. Although short-term LAM therapy may prevent the emergence of the YMDD mutation, post-treatment durability is a concern. Chien et al. showed that 58 % of 96 patients who achieved HBeAg SC after receiving LAM therapy for a median duration of 8.2 months relapsed with ALT elevations detected within one year post-treatment . Seventy-seven percent of the patients who relapsed demonstrated reappearance of HBeAg. Chien et al. further identified the determinants of post-treatment durability in a consecutive cohort of 82 HBeAg-positive patients receiving LAM therapy for a mean duration of 16 months (range, 3–55 mo) . Genotype B, age ≤36 years, and >8 months of consolidation therapy after HBeAg SC were major determinants for sustained HBeAg response to LAM therapy. Thus, an additional period of consolidation therapy after HBeAg SC is achieved appears to be imperative for minimizing the risk of post-treatment relapse, and short-term therapy may be limited to a small number of patients who achieved very early HBeAg SC following the institution of antiviral therapy.
Only 12 to 22 % of HBeAg-positive CHB patients achieve HBeAg SC after one year of oral antiviral therapy [18••, 19, 20]. Prolonged therapy leads to an incremental increase in the HBeAg SC rate, up to 50 % after five years of continuous therapy [29, 30]. Therefore, a finite course of therapy is possible in this patient population. Yeh et al. examined the relapse rate and predictors of relapse in a cohort of 71 HBeAg-positive patients who received LAM therapy for a median duration of 16.5 months with HBeAg SC and HBV DNA <60 IU/mL documented on two occasions at least 6 months apart . After 6 months off therapy, 27 % of patients relapsed with HBV DNA >2,000 IU/mL and ALT levels >2 times ULN. Pre-therapy HBV DNA >2 x 107 IU/mL was the only predictor of relapse. Kuo et al. evaluated the durability of HBeAg SC and its relation to the length of consolidation therapy in a cohort of 124 HBeAg-positive patients who achieved HBeAg SC, ALT normalization, and HBV DNA <40 IU/mL at the end of LAM therapy . The cumulative relapse rates were 54 % at 48 weeks off therapy and 68.4 % at 96 weeks off therapy. Age ≤34 years and consolidation therapy ≥48 weeks were independent predictors of sustained response. In a Korean study on 178 HBeAg-positive patients receiving LAM therapy, age ≤40 years and ≥12 months of consolidation therapy after HBeAg loss or SC were major predictors of a sustained response [33•].
Fung et al. from Hong Kong compared the virological and biochemical relapse rates between patients who stopped LAM therapy after achieving HBeAg SC with ≥6 months of consolidation therapy (N = 22) and patients who continued LAM therapy despite achieving HBeAg SC (N = 79) . At the end of follow-up, none of the patients who stopped therapy had undetectable HBV DNA levels, but 78 % of the patients who continued LAM therapy had undetectable HBV DNA levels (P < 0.001). Of the 14 (64 %) patients who had virological rebound after stopping LAM, 71 % had HBV DNA >2,000 IU/mL. Eight (10 %) patients who continued LAM developed YMDD mutations after a median treatment duration of 79 months. At five years, the cumulative incidence of ALT elevation was 44 % and 16 %, respectively (P < 0.001). Based on these observations, the authors proposed the option of long-term oral antiviral therapy in HBeAg-positive patients who achieved HBeAg SC. Although a significant proportion of patients who achieved HBeAg SC relapse after stopping oral antiviral therapy, continuation of therapy in all patients despite HBeAg SC poses the risk of overtreatment in this patient population.
Quantitative serum HBsAg levels at the end of therapy are associated with a sustained response following the cessation of therapy, particularly for interferon-based treatment [35•]. In a small-scale study of LdT therapy in 17 HBeAg-positive patients over two years, serum HBsAg levels <100 IU/mL at the end of therapy are predictive of a sustained response (defined as HBV DNA <60 IU/mL, HBeAg SC, and normal ALT) after two years off therapy, with a positive predictive value (PPV) of 93 % and a negative predictive value (NPV) of 100 % . The predictive role of the serum HBsAg levels at the end of therapy for a sustained response to oral antiviral therapy in HBeAg-positive patients deserves further study.
Short-Term Versus Long-Term Therapy in HBeAg-Negative CHB
In 85 Taiwanese HBeAg-negative CHB patients with a baseline serum ALT >5 times ULN who received LAM therapy for 6 to 12 months, 81 % achieved undetectable serum HBV DNA levels at the end of therapy, and 48 % of them maintained a sustained virological response after one year off therapy . Approximately 64 to 95 % of HBeAg-negative patients achieve undetectable serum HBV DNA levels after one year of oral antiviral therapy [18••, 19, 20]. However, withdrawal of therapy leads to relapses in most patients. In a study from China involving 61 patients who received LAM for ≥24 months and maintained undetectable HBV DNA levels and normal ALT levels for ≥18 months before stopping therapy, cumulative relapse rates at 6 and 12 months off therapy were 26.2 % and 43.6 %, respectively . Recently, another Taiwanese study examined the post-treatment durability in 79 HBeAg-negative patients who received ETV therapy for 735.2 days ± 229.3 days and stopped therapy according to the APASL treatment guidelines [39•]. The relapse rates were 27.8 % and 53.2 % at 6 and 12 months off therapy, respectively. Pre-treatment HBV DNA <2 × 105 IU/mL was the only independent predictor of a sustained virological response at 12 months off therapy. In a related study involving 58 non-cirrhotic patients, a consolidation period of >64 weeks versus <64 weeks was associated with relapse rates of 25 % and 67 %, respectively [40•]. Thus, an additional period of consolidation therapy after achieving undetectable serum HBV DNA levels appears to minimize the risk of post-treatment relapse in HBeAg-negative patients receiving oral antiviral therapy. Short-term therapy may only be possible in patients who achieved undetectable HBV DNA levels very early in treatment. Approximately 50 to 60 % of patients who achieved undetectable HBV DNA levels at the end of therapy will relapse following the cessation of therapy .
In a study from Hong Kong including 53 HBeAg-negative patients who received LAM for a mean duration of 34 months with a mean post-treatment follow-up of 47 months, serum HBsAg ≤2 log IU/mL and reduction by >1 log from baseline at the end of treatment had PPV and NPV for sustained response (defined as HBV DNA <200 IU/mL at 12 months post-treatment) of 78 % and 96 %, respectively [41•]. Thus, on-treatment monitoring of serum HBsAg levels might be useful to determine the appropriate time to stop therapy. Further studies are warranted to assess the role of end-of-treatment serum HBsAg levels for the prediction of sustained response to oral antiviral therapy in HBeAg-negative patients.
Long-Term Therapy in Compensated Cirrhosis
In a randomized, double-blind, placebo-controlled study in Asian patients with advanced fibrosis or early cirrhosis and baseline HBeAg seropositivity or HBV DNA >1.4 x 105 IU/mL, Liaw et al. showed that 7.8 % of the patients receiving LAM versus 17.7 % of the patients receiving placebo had disease progression after a median treatment duration of 32.4 months (hazard ratio [HR] = 0.45, P = 0.001). Furthermore, patients developing YMDD mutations (7 %) were more likely to have an increase in the Child-Pugh score than patients without YMDD mutations (<1 %) . Thus, LAM treatment in HBV viremic patients with advanced fibrosis or early cirrhosis can delay disease progression, but the development of drug resistant mutations during treatment can negate the therapeutic benefit. This finding suggests that long-term therapy with a potent oral antiviral agent with a high genetic barrier may provide better efficacy. A recent retrospective-prospective study from Hong Kong addressed the efficacy of ETV in the prevention of hepatic decompensation, development of HCC, or liver-related mortality in CHB patients [43••]. ETV therapy significantly reduced the 5-year cumulative incidence of hepatic events (HR = 0.51, P = 0.002), HCC (HR = 0.55, P = 0.049), and liver-related mortality (HR = 0.26, P < 0.001) in cirrhotic patients compared to a historic control group of treatment-naïve patients. Cirrhotic patients who failed to achieve undetectable HBV DNA levels during ETV therapy had a comparable risk of hepatic events as the untreated patients. Thus, antiviral therapy with a potent agent with a low resistance profile can delay disease progression and reduce liver-related mortality in patients with cirrhosis. A retrospective-prospective multicenter study assessing the efficacy of ETV in the prevention of HCC in patients with compensated cirrhosis is currently ongoing in Taiwan.
Long-Term Therapy in Decompensated Cirrhosis
Several case series involving small numbers of patients with decompensated cirrhosis consistently showed significant improvement in liver function and decrease in the Child-Pugh score following the institution of LAM therapy [17•]. A prospective multicenter study with LAM therapy in 154 North American patients with decompensated cirrhosis for a mean duration of 16 months showed that 21 % of the patients died of liver failure. Seventy-eight percent of the deaths occurred within the first 6 months of therapy, with an estimated 3-year survival of 88 % in patients who survived >6 months . The severity of liver disease at the time treatment began was a determinant of early mortality. Two recent studies showed that virological suppression with LAM did not significantly reduce the incidence of HCC in patients with decompensated cirrhosis [45, 46]. Therefore, it remains to be clarified whether LAM therapy decreases the risk of HCC in patients with decompensated cirrhosis. Liaw et al. reported 48-week results of two clinical trials assessing the safety and efficacy of potent agents such as ETV and TDF [47••, 48•]. One randomized, open-label, multicenter trial compared the safety and efficacy of ETV 1.0 mg daily versus ADV 10 mg daily for 96 weeks in 191 patients with a baseline Child-Pugh score ≥7 [47••]. A significantly greater proportion of ETV-treated patients showed HBV DNA <60 IU/mL (57 % versus 20 %) and ALT normalization at week 48 (63 % versus 46 %) compared to ADV-treated patients. Thirty-five percent of ETV-treated patients versus 27 % of ADV-treated patients showed a reduction of ≥2 points in the Child-Pugh score. The mean reductions from baseline in MELD scores were 2.6 and 1.7 at week 48 for the ETV-treated and ADV-treated patients, respectively. Cumulative HCC and death rates at week 48 were 12 % and 23 %, respectively, for ETV-treated patients. Cumulative HCC and death rates at week 48 were 20 % and 33 %, respectively, for ADV-treated patients. Better virological and biochemical responses did not translate into greater improvements in liver function, HCC occurrence, or mortality in the ETV-treated group compared to the ADV-treated group at 48 weeks of therapy. A follow-up of this cohort with decompensated cirrhosis is needed to show whether more potent HBV DNA suppression early in treatment leads to a better long-term outcome. Another randomized, double-blind, multicenter trial compared the safety of TDF 300 mg/day (N = 45), FTC/TDF 200 mg/300 mg per day (N = 45), and ETV 0.5 or 1.0 mg/day (N = 22) for 168 weeks in CHB patients with a current or past history of hepatic decompensation [48•]. At week 48, the proportions of patients with a serum creatinine increase ≥0.5 mg/dL over baseline or a serum phosphorus <2 mg/dL were 8.9 %, 6.7 % and 4.5 %, respectively. The proportions of patients with HBV DNA <80 IU/mL were 70.5 %, 87.8 % and 72.7 %, respectively. The median reductions from baseline in the MELD score were 2.0 in all three groups, and 25.9 %, 48 % and 41.7 % of patients showed a reduction of ≥2 points in the Child-Pugh score, respectively. Final results are eagerly awaited to evaluate the long-term efficacy, safety, and resistance profiles across the groups to assess the role of combination therapy in patients with decompensated cirrhosis.
Nucleos(t)ide analogues have been widely used in the treatment of CHB in Taiwan. Short-term therapy may be possible in a small proportion of HBeAg-positive or -negative patients who achieve HBeAg SC and/or undetectable HBV DNA levels early in treatment. Consolidation therapy is essential for the prevention of post-treatment relapse. Patients with compensated or decompensated cirrhosis should receive long-term therapy to halt or reverse disease progression. The role of serum HBsAg levels for the prediction of a sustained response to NA therapy needs further investigation.
Conflict of Interest
Cheng-Yuan Peng is a paid board member for Bristol-Myers-Squibb.