HBeAg-negative chronic hepatitis B: why do I treat my patients with nucleos(t)ide analogues?


  • Mauro Viganò,

    1. Hepatology Division, Ospedale San Giuseppe, Università degli Studi di Milano, Milan, Italy
    Search for more papers by this author
  • Giampaolo Mangia,

    1. “A.M. e A. Migliavacca” Center for the Study of Liver Disease, 1st Division of Gastroenterology, Fondazione IRCCS Cà Granda Ospedale Maggiore Policlinico, Università degli Studi di Milano, Milano, Italy
    Search for more papers by this author
  • Pietro Lampertico

    Corresponding author
    1. “A.M. e A. Migliavacca” Center for the Study of Liver Disease, 1st Division of Gastroenterology, Fondazione IRCCS Cà Granda Ospedale Maggiore Policlinico, Università degli Studi di Milano, Milano, Italy
    • Correspondence

      Pietro Lampertico MD, PhD, 1st Division of Gastroenterology, Fondazione IRCCS Ca’ Granda - Ospedale Maggiore Policlinico, Università degli Studi di Milano, Via F. Sforza 35, Milan 20122, Italy

      Tel: +39 0255035432

      Fax: +39 0250320700

      e-mail: pietro.lampertico@unimi.it

    Search for more papers by this author


The aim of chronic hepatitis B (CHB) antiviral therapy is to persistently suppress HBV and improve survival by preventing the progression of liver damage to cirrhosis, end-stage liver disease or hepatocellular carcinoma (HCC), thus preventing early liver-related death. In HBeAg-negative patients who do not or will not respond to or be treated with pegylated interferon (PEG-IFN), the administration of third generation nucleot(s)ide analogues (NAs), i.e. entecavir (ETV) and tenofovir disoproxil fumarate (TDF), is the treatment of choice. Long-term administration of ETV or TDF suppresses HBV replication in >95% of patients after 5 years of treatment with high rates of biochemical normalization, regression of fibrosis and cirrhosis at histology as well as preventing clinical decompensation but not HCC, in compensated cirrhosis and improving survival. No major safety issues have been recorded with either drug. The need for long-term, perhaps indefinite, treatment is the main limitation of NA therapy with possible associated costs, unknown long-term safety and the low rates of HBsAg seroclearance. The latter is important since HBsAg seroclearance is still the best stopping rule for HBeAg-negative NA-treated patients, including those with cirrhosis. For this reason new trials based upon a combination of PEG-IFN and third generation NAs in both naïve and NA-responder HBeAg-negative patients are ongoing.


adefovir dipivoxil


adverse events


estimated glomerular filtration rate






hepatitis B virus


hepatocellular carcinoma




nucleot(s)ide analogues


pegylated interferon


sustained virological response




tenofovir disoproxil fumarate

Chronic hepatitis B virus (HBV) infection is still one of the most common infectious diseases and a leading cause of death worldwide with an estimated 400 million chronically infected individuals [1]. HBeAg-negative CHB, which is currently the predominant type of CHB in many countries, is a difficult to cure, progressive disease leading to end-stage liver disease and HCC [2]. The goal of therapy in these patients is to persistently suppress viral replication to halt progression of liver damage and prevent the development of cirrhosis and other liver-related complications, such as HCC and clinical decompensation [3-5]. These end-points can be achieved by either short-term ‘curative’ treatment with pegylated interferon (PEG-IFN) or long-term ‘suppressive’ therapy with nucleot(s)ide analogues (NAs) [6]. According to recently updated European Association for the Study of the Liver (EASL) HBV guidelines, PEG-IFN and third generation NAs such as entecavir (ETV) or tenofovir (TDF) are the first-line drugs recommended for treatment naïve CHB patients. A 48-week course of PEG-IFN is indicated in patients without contraindications because it is practically the only strategy to offer a chance for a sustained off-treatment response after a finite course of therapy compared to NAs, which require long-term administration [5].

Administration of NAs is easy and these drugs are the treatment of choice in patients with compensated cirrhosis, because of the risk of decompensation associated with PEG-IFN–related flares of hepatitis, in patients with decompensated cirrhosis, in older patients and in patients with contraindications or who are unwilling to take PEG-IFN. However, the subcutaneous use of PEG-IFN, the need for frequent clinical and laboratory monitoring, the side effects and the limited effectiveness have made NAs the treatment of choice for most HBeAg-negative CHB patients worldwide.

All NAs belong to the same class (HBV polymerase inhibitors) and can be subdivided into nucleoside analogues including lamivudine (LAM), ETV, telbivudine (TBV) and nucleotide analogues including adefovir dipivoxil (ADV) and TDF [3-5]. However, given the high potency and low resistance rates, only third generation NAs, i.e. ETV or TDF, should be considered as first-line drugs in patients with CHB. LAM, ADV and TBV are no longer recommended because of the limited efficacy and moderate to high resistance rates of these drugs [3-5].

This article reviews the long-term efficacy and safety of ETV and TDF in NAs-naïve HBeAg-negative CHB patients.

Virological, biochemical and serological response

Enctecavir (ETV)

In the VIRGIL Study, including 243 NAs-naïve CHB patients (mean age 43 years old, 24% with cirrhosis, 64% HBeAg-negative) treated with ETV monotherapy, the cumulative probability of achieving a sustained virological response (HBV DNA <80 IU/ml) in HBeAg-negative patients at weeks 48, 96 and 144 was 89, 98 and 99%, respectively, with two patients (1%) who achieved HBsAg loss and none who developed ETV-resistance [7]. A United States study in 153 patients treated with ETV showed high response rates in naïve HBeAg-negative patients that reached 100% at year 3 [8]. In a Hong Kong cohort including 132 NAs-naïve HBeAg-negative patients receiving ETV, HBV DNA undetectability (<12 IU/ml) and alanine aminotransferase (ALT) normalization rates progressively increased to reach 98% (HBV DNA) and 86% (ALT) at year 3, respectively. One patient seroconverted to anti-HBs at year 3 while none developed ETV resistance [9]. In a Japanese study performed in 252 NAs-naïve HBeAg-negative patients treated with continuous ETV for 4 years, 100% of the patients achieved undetectable HBV DNA levels (<2.6 log10 copies/ml) and 91% normalized ALT levels, with no case of ETV resistance [10].

In a field practice study in Italy including 418 NAs-naïve patients (mean age 58 years old, 49% with cirrhosis, 83% HBeAg-negative) treated with ETV for 5 years, the rates of viral suppression and ALT normalization was 100 and 93% among HBeAg-negative patients respectively. One patient (0.7%) who developed ETV resistance at year 3, was successfully rescued by TDF [11].

The efficacy was similar after 1 year of treatment with ETV in decompensated and compensated patients with cirrhosis; patients achieved undetectable serum HBV DNA in 89% (decompensated cirrhosis) and 78% (compensated cirrhosis) respectively [12]. In a randomized, open-label study in 195 CHB patients with decompensated cirrhosis (45% HBeAg-negative), ETV 1 mg/die suppressed viral replication at week 48 more effectively than ADV 10 mg/die (57% vs 20%, P < 0.0001) [13].

Tenofovir (TDF)

The efficacy of TDF treatment was evaluated in field practice studies in Europe. A German multicentre observational study (GEMINIS) included 400 CHB patients (mean age 45 years, 11% cirrhosis, 69% HBeAg-negative, 46% treatment-naïve). Virological response (HBV DNA <169 IU/ml) in the HBeAg-negative naïve patients at 24 months of treatment was 94%. No virological breakthrough and no TDF resistance were reported [14]. In a French multicentre prospective cohort study (Vireal study), 441 HBV patients treated with TDF were included (mean age 45 years, 74% HBeAg-negative, 33% advanced fibrosis/cirrhosis and 42% treatment-naïve). Virological response (HBV DNA <69 IU/ml) at year 2 in naïve patients was 96% with HBsAg loss in eight HBeAg-negative patients [15]. In a 4 year multicentre European clinical practice study, 97% of the 374 NAs-naïve CHB patients (median age 55 years, 35% cirrhosis, 80% HBeAg-negative) treated with TDF achieved a virological response, five clearing HBsAg (1.7%), with no case of TDF resistance [16].

In the 6 year follow-up of the TDF-registration trials, 86% had normal ALT levels, 99% of HBeAg-negative patients achieved undetectable HBV DNA, but only one patient cleared HBsAg [17].

The virolgical efficacy of 1 year of TDF in patients with decompensated liver disease was evaluated in a double-blind study that randomized 112 CHB patients to receive either TDF (n = 45), combination therapy with emtricitabine (FTC) plus TDF (n = 45) or ETV (n = 22). After 48 weeks of treatment, similar response rates (HBV DNA <400 copies/ml) were observed in the three treatment arms (71, 88 and 73% respectively) while ALT normalization occurred in 57, 76 and 55% of patients in the three groups, respectively [18].

Partial virological response to ETV or TDF

The rates of resistance with ETV or TDF are negligible although the few cases of ETV resistance in NA-naïve patients occurred in patients with a partial virological response [6], defined as detectable HBV DNA at week 48. Optimal management of patients with a partial virological response after 48 weeks of ETV or TDF is the subject of intense debate. HBV DNA levels at week 48 and their kinetics must be taken into account in these patients. Patients with residual viraemia ≤1000 IU/ml or with a continuous decline in serum HBV DNA levels may continue the same treatment because of the progressive increase in virological response over time and the low risk of resistance with long-term monotherapy. In patients with a flat pattern of HBV DNA, with a residual viraemia >1000 IU/ml, or with underlying cirrhosis, a rescue strategy with a non cross-resistant analogue, i.e. TDF for partial response to ETV and vice versa, can be recommended [19].

Safety and tolerability

Entecavir (ETV). Registration and clinical practice trials have shown that NAs are generally well-tolerated and safe. The better tolerance of NAs is an important advantage compared with PEG-IFN, which can negatively affect the patients’ quality of life and is associated with a wide range of adverse events (AEs) [20]. Although NAs have a good safety profile, no drug can be considered to be absolutely safe, especially if it is given for years or even for life as in the case of HBeAg-negative patients [21].

Long-term administration of ETV was associated with low rates of severe AEs and discontinuations because of AEs. After a median of 184 weeks of ETV, the most common (≥10%) AEs were mild to moderate upper respiratory tract infections, headache and nasopharyngitis, not related to antiviral therapy. Nineteen percent of patients reported grade 3–4 AEs which were considered to be related to ETV in 4% of cases ultimately leading to treatment discontinuation in 1% of patients [22]. A few cases of lactic acidosis were reported in patients with decompensated cirrhosis and a baseline MELD score >22 points. These were resolved in most patients after drug withdrawal [23]. Nevertheless lactic acidosis was not reported in any of the 70 Korean patients with decompensated cirrhosis treated with ETV for 2 years [12]. Interestingly, severe AEs spontaneously resolved when antiviral treatment was continued [18]. This suggests that particular care should be taken when administering ETV to patients with severe liver damage and high baseline MELD scores and that treatment should be suspended in patients who develop clinical or laboratory findings suggesting lactic acidosis [24].

In decompensated patients, serum creatinine levels increased by >0.5 mg/dl from baseline over the 96 weeks of treatment in 17 and 24% of patients on ETV and ADV respectively. A dose reduction was necessary in 10 (11%) patients receiving ETV because of a change in renal function at one or more time points, and the drug was discontinued in 7 (7%) because of AEs [18].

Serum creatinine and serum phosphate levels remained unchanged in an Italian field practice study including more than 400 patients treated for 5 years with ETV: from 0.90 to 0.90 mg/dl and from 3.4 to 3.2 mg/dl, respectively, with <1% of patients with low phosphorus levels or significant proteinuria. The maximal tubular reabsorption of phosphate (TmPO4/GFR) was <0.80 mmol/L in ~30% of patients at month 12 and throughout this study [11], although the pretreatment assessment of this variable was not available.

Tenofovir (TDF). TDF is generally well-tolerated although a decline in the estimated glomerular filtration rate (eGFR), acute renal failure and Fanconi syndrome have been described in patients with HIV receiving TDF containing regimens [25]. Less than 2% of patients with HBV monoinfection discontinued TDF because of AEs and ≤1.5% of patients experienced a confirmed renal event, i.e. ≥0.5 mg/dl increase in serum creatinine from baseline, phosphorus <2 mg/dl, or eGFR <50 ml/min, after 6 years of TDF as part of the long-term follow-up of the registration trial [17].

In a 4 year European field practice cohort study enrolling 374 TDF treated NAs-naïve patients. serum creatinine and serum phosphate levels remained unchanged (0.90 and 3.3 mg/dl) while eGFR declined from 84 to 80 ml/min during treatment. The proportion of patients with eGFR <50 and <60 ml/min increased from 2 to 3% and from 7 to 11% respectively. Likewise, the percentage of patients with serum phosphate <2.3 mg/dl increased from 2% at baseline to 5% at year 4, while 1% of the patients had phosphate <2.0 mg/dl throughout the study period. Because of renal events, the dose of TDF was adjusted in 19 (5%) patients (eGFR decline in 17; low serum phosphate in 2) and the drug discontinued in additional 7 (2%) cases [16]. In a recent study reporting the safety of a 96-week course of TDF in patients with mild baseline renal impairment, i.e. eGFR 50–80 ml/min [26], none of the patients had a confirmed increase ≥0.5 mg/dl in serum creatinine compared with baseline, whereas nine patients, all with baseline eGFR <61 ml/min, had eGFR declining <50 ml/min that, however, stabilized after dose adjustment. In the Vireal French study, the rate of adverse events was 14%, similar to the 15% reported in elderly (>65 years) patients. Moreover, although 82% of elderly patients had reduced eGFR (<90 ml/min) before therapy, eGFR remained stable or improved in 91% of the cases during treatment [15]. No significant difference in renal safety was found in CHB patients with decompensated liver disease treated with TDF alone or as fixed-dose combination therapy with FTC or with ETV. The proportion of subjects with a confirmed increase in serum creatinine >0.5 mg/dl from baseline or confirmed serum phosphorus <2.0 mg/dl was 9, 7 and 5%, among patients treated with TDF monotherapy, TDF in combination FTC or with ETV [18]. Recently, an Italian field practice study in 156 NAs-naïve patients who received TDF reported that 14% the patients had hypophosphataemia (≤2.5 mg/dl) and 35% had low TmPO4/GFR (≤0.80 mmol/L) at baseline before any antiviral treatment. During 2 years of TDF treatment, de novo hypophosphataemia and low TmPO4/GFR occurred in 4 (6%) and 18 (26%) additional patients [27].

To date, only three cases of TDF-associated Fanconi syndrome have been reported in HBV moninfected patients; all patients recovered completely after switching to ETV [28, 29].

Since all NAs are excreted through the kidneys, dosing should be adjusted in patients with eGFR <50 ml/min according to the Modification of Diet in Renal Disease formula. Therefore, all patients should be tested for serum creatinine levels before starting NAs to estimate pretreatment GFR and define the appropriate dose of antiviral therapy [5]. In addition, the baseline renal risk should be assessed in all patients. The ‘high renal risk’ group includes one or more of the following factors: decompensated cirrhosis, creatinine clearance <60 ml/min, poorly controlled hypertension, proteinuria, uncontrolled diabetes, active glomerulonephritis, concomitant nephrotoxic drugs, solid organ transplantation. All CHB patients receiving TDF should be monitored for eGFR and serum phosphate whereas patients on ETV should only be only strictly monitored for eGFR if they belong to the ‘high renal risk’ group [5] (Fig. 1). However, recent data suggest that TDF-treated patients with low serum phosphate (<2.5 mg/dl) or reduced eGFR (<60 ml/min) at baseline or during treatment as well as those previously exposed to ADV, should be proactively considered for TDF dose reductions to avoid renal complications [30]. Assessment of bone mineral density by DEXA should be considered in patients with a history of stress bone fractures or other risk factors for osteoporosis or bone loss.

Figure 1.

Algorithm for monitoring tenofovir and entecavir renal safety.

Regression of fibrosis

CHB patients with advanced fibrosis or cirrhosis demonstrated histological improvement and reversal of fibrosis and cirrhosis after long-term treatment with both ETV and TDF [31]. A second liver biopsy evaluation after a median of 6 years in 57 patients under long-term ETV treatment showed a significant histological improvement (≥1 point improvement in the Ishak fibrosis score) in 88% of patients, including all 10 patients with advanced fibrosis or cirrhosis at baseline [32]. A reduction in the Ishak fibrosis score to 4 or less was observed in all four patients who had cirrhosis at baseline [33]. A more robust proof of the regression of cirrhosis was reported in NAs-naïve patients treated with TDF enrolled in registration trials. Of the 96 patients with cirrhosis (Ishak score 5 or 6) at baseline, 71 (74%) had histologically reversed cirrhosis [34].

Prevention of liver-related complications

Clinical decompensation is fully prevented during 5 years of ETV or TDF therapy in patients with compensated cirrhosis as none developed ascites, hepatic encephalopathy, jaundice or gastrointestinal bleeding [6, 11, 34]. Survival was significantly improved by antiviral therapy in patients with decompensated liver disease, because persistent HBV DNA suppression led to reversal of clinical decompensation in most patients [13, 18].

In the VIRGIL study including 372 patients (274 without cirrhosis, 89 with compensated cirrhosis and 9 with decompensated liver disease) treated for 20 months with ETV, a virological response was associated with a lower probability of developing HCC, decompensated cirrhosis or death. However, the benefit of the virological response was only significant in patients with cirrhosis [35]. In Hong Kong, 1446 consecutive CHB patients (mean age 51 years, 33% with cirrhosis, 70% HBeAg-negative) who received ETV for a median of 36 months were compared with a historical control cohort of 424 untreated patients. There was no difference in HCC, liver-related mortality and/or hepatic events defined as any liver-related complications (ascites, spontaneous bacterial peritonitis, hepatic encephalopathy, variceal bleeding, hepatorenal syndrome) between the ETV and control cohort. Overall, the 5-year cumulative rates of HCC in ETV-treated patients and controls were similar (6.6% vs 6.5%, respectively), and were the same in the subgroup of patients without cirrhosis (3.3% vs 3.0%) but were significantly lower in ETV-treated patients with cirrhosis compared with untreated controls with cirrhosis (13.8% vs 26.4%). Interestingly, patients with cirrhosis who received ETV and achieved maintained viral suppression had significantly lower rates of hepatic events, HCC and liver-related mortality compared with ETV-treated patients with cirrhosis without viral suppression and to untreated controls with cirrhosis [36]. In the Italian ETV-treated cohort, the 155 patients with compensated cirrhosis remained clinically stable although 17 developed HCC resulting in a 5-year cumulative rate of 14% despite full suppression of viral replication in most cases. The overall 5-year cumulative survival and liver-related survival were 91 and 95% respectively [11].

The effect of antiviral therapy on the incidence of HCC was recently reported using a predictive model (the REACH-B model) from 6-year follow-up data of the registration trial in naïve patients receiving TDF. This study reported that the number of HCC (13 cases including six in patients with cirrhosis) in TDF-treated patients decreased significantly compared with the predicted risk starting from the third year of therapy [37]. In the field practice European cohort of TDF-treated NAs-naïve patients, HCC rates were 1 and 4%/year in patients with chronic hepatitis and compensated cirrhosis, respectively, while clinical decompensation was completely prevented [16].

Overall, there is no robust clinical evidence that long-term suppression of HBV replication for 5 years with ETV or TDF prevents or reduces the rate of HCC in Asian or Caucasian patients with chronic hepatitis with or without compensated cirrhosis. Thus, monitoring of HCC is mandatory in this clinical setting even when virological responses are obtained.

Can treatment with ETV or TDF be stopped?

The need for long-term, perhaps indefinite, treatment is the main limitation of NAs therapy in HBeAg-negative patients with the possible disadvantage of cost, unknown long-term safety and the low rate (1% at 5 years, 5% at 10 years) of HBsAg seroclearance. Several studies have indeed shown that the decline of HBsAg titres in HBeAg-negative patients under NAs treatment is so limited that over 30 years of treatment are needed to achieve HBsAg loss in 50% of the patients [38-41]. The latter is important because HBsAg seroclearance or seroconversion is the only stopping rule in these patients based on the American Association for the Study of Liver Disease and the EASL guidelines [4, 5]. The Asian Pacific Association for the Study of the Liver (APASL) guidelines recommend that treatment can be discontinued before HBsAg loss if undetectable HBV DNA has been documented by real-time PCR on three occasions at least 6 months apart [3]. However, to date, there is only study describing the outcome of 95 HBeAg-negative patients treated with ETV for 2 years who stopped antiviral therapy according to the APASL guidelines, i.e. confirmed HBV DNA undetectability in the presence of HBsAg [42]. Within 1 year after discontinuation of ETV, an hepatitis flare (ALT >2× upper limit of normal and HBV DNA >2000 IU/ml) occurred in 43 (45%) patients. Low baseline viraemia, i.e. HBV DNA ≤2 × 105 IU/ml, was the only significant independent predictor for a sustained response because the 1-year relapse rate was significantly lower in patients with low viraemia compared with those with baseline HBV DNA >2 × 105 IU/ml (29% vs 53%, P = 0.027).

Another possible strategy to avoid lifelong NAs treatment is to ‘add on’ IFN to NAs-responders to further reduce HBsAg levels and accelerate HBsAg clearance. Although four pivotal studies have tested this strategy with promising results, this ‘add on’ IFN strategy should not be used in clinical practice until more definite results are available from ongoing randomized clinical trials [42-46].


The most popular and effective anti-HBV therapeutic strategy in HBeAg-negative CHB patients is long-term administration of third generation NAs such as ETV and TDF. The advantages of this strategy include excellent tolerance, effective inhibition of HBV replication without the emergence of drug-resistance, high rates of biochemical remission, histological improvement and the prevention of clinical decompensation while it is still unclear whether the risk of HCC is also reduced. NAs are the only treatment for patients with severe liver disease, for older patients, in patients in whom PEG-IFN is contraindicated or who are unwilling to take PEG-IFN and in those with significant concomitant diseases. However, long-term administration of ETV or TDF will not eradicate HBV making long-term therapy necessary in most patients, increasing costs, compliance issues and raising questions on unproven safety profiles.