Long-term lamivudine therapy for children with HBeAg-positive chronic hepatitis B†
Potential conflict of interest: Dr. Schwarz has received research grants from Gilead and research and unrestricted grants from Roche. Drs. Gardener and Little own stock in GlaxoSmithKline. Dr. Sokal is a consultant for and has received research grants from GlaxoSmithKline.
One year of lamivudine treatment results in increased hepatitis B e antigen (HBeAg) seroconversion and serum hepatitis B virus (HBV) DNA negativity in children with chronic hepatitis B and high serum alanine aminotransferase concentrations. Two hundred seventy-six children who participated in a 1-year randomized, placebo-controlled study of lamivudine were enrolled in a 24-month, open-label extension. Patients were stratified into two groups based on HBeAg status at week 48 of the previous study: 213 HBeAg-positive children were entered into a treatment arm, and 63 HBeAg-negative children were entered into an observation arm to evaluate durability of HBeAg loss. In the treatment arm, 28 of 133 (21%) children previously treated with lamivudine and 23 of 77 (30%) children who previously received placebo achieved the primary end point: virological response (VR) (HBeAg loss and HBV DNA negativity) at month 24. The incidence of YMDD (tyrosine, methionine, aspartate, aspartate) mutations at month 24 was 64% (66/103) in the children previously treated with lamivudine and 49% (34/70) in those previously treated with placebo. The incidence of VR at month 24 was 5% (5/100) for patients with YMDD mutant HBV and 54% (39/72) for patients without. The durability of response in the observation arm was 89% (48/54) at month 24. In conclusion, further clinical response was seen over the 24-month open-label study period in children who had not initially achieved a VR after 12 months of lamivudine treatment. However, the incidence of YMMD mutations increased over time and resulted in lower response rates. VR was maintained in most patients who had initially responded to lamivudine in the first 12 months. (HEPATOLOGY 2006;43:225–232.)
Chronic hepatitis B is a significant global health care problem that affects more than 350 million people worldwide, resulting in approximately 1 million deaths each year.1 Despite the widespread availability of an effective vaccine, large numbers of children are infected. People infected early in life who have evidence of ongoing viral replication are at the highest risk for development of progressive liver disease and are a major source of infection in others.2
Treatment options for children are limited. The goal of treatment is to prevent the progression of chronic hepatitis B to long-term complications, such as cirrhosis and hepatocellular carcinoma, that can result in premature death.3 Interferon α has demonstrated efficacy in achieving hepatitis B e antigen (HBeAg) loss and seroconversion with HBV DNA suppression in children over 1 year old with active hepatitis B. However, only one third of treated patients will respond, and there are significant side effects associated with interferon therapy, including influenza-like symptoms, gastrointestinal disorders, nausea and vomiting, neutropenia, thrombocytopenia, and transient impairment of growth.4, 5
Treatment with lamivudine, an oral nucleoside analog, has been shown to result in similar response rates as interferon in both adults and children.6–9 The drug is well tolerated, but long-term efficacy can be compromised by the emergence of resistance.10 Treatment with lamivudine was previously assessed in a 52-week, phase III randomized, double-blind, placebo-controlled study in 288 pediatric patients with chronic hepatitis B. After 52 weeks, there was a significantly increased rate of virological response (VR) (absence of HBeAg and HBV DNA) in children treated with lamivudine compared with placebo (23% vs. 13%, respectively); the best results were obtained in children with aminotransferase levels more than 2 times the upper limit of normal.11
The optimal length of treatment with lamivudine required to achieve a durable VR in children has not been established. In addition, the durability of VR following treatment with lamivudine in children has not been evaluated. The purpose of the present study was to evaluate the benefit of extended treatment for patients who remained HBeAg-positive at the end of a previous 52-week, randomized, double-blind, placebo-controlled study of lamivudine therapy. In addition, we aimed to evaluate the long-term off-treatment durability of VR.
Patients and Methods
This was a 24-month, uncontrolled, open-label, stratified study in pediatric patients who had previously completed a 52-week placebo-controlled study of lamivudine.11 The study was conducted between August 1999 and January 2003 at 38 centers in North America, South America, and Europe. The academic investigators participated in the design of the study, the interpretation of the results, and the development of the manuscript. The sponsor held the data and conducted the statistical analysis. All laboratory tests, except for prothrombin time and resistance testing, were conducted at Covance Central Laboratory Services (Geneva, Switzerland, and Indianapolis, IN). Prothrombin times were evaluated at each investigational site, and resistance testing was performed by the Clinical Virology Department at GlaxoSmithKline (Research Triangle Park, NC).
The study was conducted in compliance with the Declaration of Helsinki and was approved by the appropriate local regulatory bodies. All patients—or their parents or legal guardians—gave written informed consent to participate.
To have been eligible for the previous study, patients were required to be between 2 and 17 years of age, to be seropositive for hepatitis B surface antigen (HBsAg) for at least 6 months before enrollment, to be seropositive for HBeAg, to have a serum alanine aminotransferase (ALT) level more than 1.3 times the upper limit of normal (but <500 IU/L) for at least 3 months before enrollment, to have evidence of inflammation on liver biopsy, and to have measurable serum HBV DNA as assessed via branched-chain DNA assay (Chiron Quantiplex assay, lower-limit of detection = 0.7 mEq/mL). Patients were enrolled into one of two groups based on their HBeAg status at the week 48 visit of the initial study. Patients who remained HBeAg-positive were enrolled into a treatment arm and received lamivudine by solution at a dose of 3 mg/kg body weight per day (maximum daily dose: 100 mg) or by tablet (100 mg, once daily). Patients who were HBeAg-negative were assigned to an observational arm.
Clinical and Virological Procedures.
To ensure that there was no treatment interruption, patients were screened at week 48 prior to completing the initial study. Baseline was defined at the last on-treatment visit from the initial study, (conducted at week 52). Patients were assessed at baseline and at months 1, 2, 3, 6, 9, 12, 15, 18, 21, and 24. Blood samples were drawn at each visit to monitor blood chemistry and hematology. Female patients of childbearing potential were required to have a pregnancy test at each visit. Commercially available enzyme-linked immunoassays were used to measure HBeAg, anti-HBe, HBsAg, and anti-HBs (HBe[rDNA] Enzyme Immunoassay, AxSym HBe Microparticle Enzyme Immunoassay, HBs Auszyme Immunoassay, and Ausab Enzyme Immunoassay; Abbott Laboratories, Abbott Park, IL). Serum HBV DNA levels were measured using a signal amplification branch assay [(bDNA) Chiron (Bayer) Quantiplex (version 1.0, lower limit of detection = 0.7 mEq/mL)]. Serum samples were collected at baseline, month 12 and month 24, or the end of treatment, and were frozen at −80°C for subsequent genotypic analysis for resistance mutations performed at GlaxoSmithKline Virology Laboratories. Mutations in the YMDD (tyrosine, methionine, aspartate, aspartate) motif of the reverse-transcriptase domain in the HBV polymerase gene were assessed via polymerase chain reaction and restriction fragment length polymorphism assay.
The primary efficacy end point for the treatment group was VR, which was defined as HBeAg negativity and undetectable HBV DNA concentrations (<0.7 mEq/mL) at month 24 of the open-label study. The study medication was discontinued if a subject had a VR present on two consecutive visits. Secondary end points at month 24 included 2-component HBeAg seroconversion (decrease in HBeAg to undetectable levels, and a rise in HBeAb to detectable levels) and 3-component HBeAg seroconversion (decrease in HBeAg to undetectable levels, a rise in HBeAb to detectable levels, and undetectable HBV DNA concentrations), HBeAg loss, HBsAg loss, HBsAg seroconversion, HBV DNA loss, ALT normalization and a combined response end point (HBeAg seroconversion, ALT normalization, and HBV DNA undetectability). For the no-treatment observational group, the primary objective was the proportion of patients who maintained VR at month 24.
All patients were assessed for adverse events at each visit. Levels of serum ALT, albumin, bilirubin, amylase, lipase, creatinine kinase, hemoglobin, white cells, neutrophils, and platelets were assessed. Weight and height were measured.
Patients in the observational group could be retreated with lamivudine if there was a recurrence of hepatitis B disease activity (at two consecutive study visits at least 7 days apart). Disease recurrence was defined as the return of HBV DNA to detectable levels and an elevation of serum ALT to at least 1.5 times the value before the re-emergence of HBV DNA and greater than the upper limit of normal.
All analyses were prespecified. The primary population for analysis was a modified intent-to-treat population consisting of patients with confirmed chronic hepatitis B assigned to the treatment stratum, regardless of whether the study drug was taken or if the subject completed the study. No formal statistical testing was conducted. There was no plan to formally compare the outcomes of the observation and treatment arms for inference purposes. However, for exploratory purposes and hypothesis generation, logistic regression models were constructed to examine the relationship between outcomes, treatment arms, and important covariates. Conclusions reached from these models are limited to this study, and extraction of results beyond this study should be validated against other studies.
For the primary and secondary end points summarized at month 24, missing data at month 24 were counted as a failure for patients that had not withdrawn. Patients in the observation arm who were retreated were counted as failures from that point forward.
Among the 276 patients who were enrolled, 213 (77%) were assigned to the lamivudine treatment arm, and 63 (23%) were assigned to the no-treatment observational arm (Fig. 1). Of the 213 patients in the treatment arm, 134 (63%) had previously been treated with lamivudine, and 79 (37%) had been treated with placebo. Of the 63 patients in the observational arm, 49 (78%) had previously been treated with lamivudine, and 14 (22%) had been treated with placebo.
The demographics of the two groups were similar (Table 1). Sixty-four percent of the treatment group was male, and the median age was 10 years (range: 2–18 yr). In the observation group, 68% were male, and the median age was 9 years (range: 3–18 yr). In the treatment arm, there were 5 patients who lost detectable HBeAg between the screening visit and study baseline. Three patients in the observation arm regained HBeAg between the screening visit and baseline.
Table 1. Baseline Demographics and Disease Characteristics
|Age (yr), median (range)||8.0 (3–18)||10.5 (2–18)||12.5 (5–16)||9.0 (3–18)|
|Male, n (%)||47 (59%)||89 (66%)||11 (79%)||32 (65%)|
|Race, n (%)|| || || || |
| White||47 (59%)||100 (75%)||11 (79%)||34 (69%)|
| Asian||22 (28%)||22 (16%)||2 (14%)||9 (18%)|
| Black||8 (10%)||8 (6%)||1 (7%)||3 (6%)|
| Hispanic||1 (1%)||2 (1%)||0||1 (2%)|
| Other||1 (1%)||2 (1%)||0||2 (4%)|
|Serum ALT (× ULN), median (range)||2.0 (0.3–14.0)||0.9 (0.4–5.4)||0.6 (0.4–1.4)||0.6 (0.2–2.5)|
|Normal serum ALT*||6 (8%)||7 (5%)||11 (79%)||45 (92%)|
|Serum ALT|| || || || |
| ≤ULN||8 (10%)||75 (56%)||11 (79%)||45 (92%)|
| >1 and ≤2 × ULN||31 (39%)||44 (33%)||3 (21%)||3 (6%)|
| >2 × ULN||40 (51%)||15 (11%)||0 (0%)||1 (2%)|
|Serum HBV DNA (mEq/mL)†, median (range)||707.3 (0.35–20750)||0.35 (0.35–1217)||0.35 (0.35–3.0)||0.35 (0.35–2.0)|
|Undetectable HBV DNA||3 (4%)||75 (56%)||13 (93%)||44 (90%)|
|YMDD status, n (%)||0/73 (0%)||29/121 (24%)||0/13 (0%)||2/45 (4%)|
|HBeAg status, n (%)|| || || || |
| HBeAg-positive||76 (97%)||130 (98%)||0 (0%)||3 (6%)|
| Anti–HBe-positive||8 (10%)||14 (10%)||14 (100%)||46 (94%)|
At baseline, 90% of patients (71/78) in the treatment group who had previously received placebo had elevated serum ALT levels and 96% (76/79) had detectable HBV DNA concentrations. Of the patients in the treatment group who had previously received lamivudine, 44% (59/134) had elevated serum ALT levels and detectable HBV DNA concentrations. In contrast, the majority of patients in the observation group had normal serum ALT and undetectable HBV DNA concentrations at baseline (Table 1).
VR (HBeAg negativity and HBV DNA undetectability) was achieved by 51 of 210 patients (24%) in the treatment group at month 24. VR rates were higher in patients who had received placebo in the initial study (30% vs. 21%; P = .013). Combined response (HBeAg seroconversion, HBV DNA undetectability and serum ALT normalization) was observed in 50 of 213 patients (23%) at month 24. Again, combined response rates were higher in patients who had received placebo in the initial phase of the study (29% vs. 20%; P = .013).
Serum ALT concentrations were normalized in 95 of 213 patients (45%) at month 24. Serum ALT normalization rates appeared to be slightly higher in patients previously treated with placebo compared with those previously treated with lamivudine (54% vs. 39%, respectively). At month 24, serum ALT normalization rates were lower in patients who had developed YMDD mutant HBV (28/93 [30%]) compared with those who had not (51/70 [73%]).
HBsAg loss was observed in 4 of 201 patients (2%) at month 24; one patient had previously been treated with placebo, and three had been treated with lamivudine. Two of the three patients with HBsAg loss who were previously treated with lamivudine also developed antibodies to HBsAg.
Patients with baseline serum ALT concentrations greater than 5 times the upper limit of normal appeared to have higher rates of VR, HBeAg seroconversion and HBeAg loss at month 24 than patients with baseline serum ALT concentrations greater than the upper limit of normal and less than or equal to 5 times the upper limit of normal. None of the patients with normal serum ALT concentrations at baseline achieved VR, HBeAg seroconversion or HBeAg loss at month 24.
In the 2- to 6-year-old and 13- to 17-year-old groups, there appeared to be higher rates of VR, HBeAg seroconversion, and HBeAg loss at month 24 in the patients who had previously received placebo (Table 2), whereas the rates of VR, HBeAg seroconversion, and HBeAg loss at month 24 were similar in both groups of patients in the 7- to 12-year-old group.
Table 2. Virological and Serological Response at Month 24 (Treatment Arm)
|Overall virological response||23/77 (30%)||28/133 (21%)||51/210 (24%)|
| Age 2–6 yrs||14/35 (40%)||4/29 (14%)||18/64 (28%)|
| Age 7–12 yrs||5/21 (24%)||22/70 (31%)||27/91 (30%)|
| Age 13–17 yrs||4/21 (19%)||2/34 (6%)||6/55 (11%)|
|HBeAg seroconversion||26/77 (34%)||34/133 (26%)||60/210 (30%)|
| Age 2–6 yrs||14/35 (40%)||4/29 (14%)||18/64 (28%)|
| Age 7–12 yrs||5/21 (24%)||26/70 (37%)||31/91 (34%)|
| Age 13–17 yrs||7/21 (33%)||4/34 (12%)||11/55 (20%)|
|HBeAg loss||27/77 (35%)||35/133 (26%)||62/210 (30%)|
| Age 2–6 yrs||15/35 (43%)||4/29 (14%)||19/64 (30%)|
| Age 7–12 yrs||5/21 (24%)||27/70 (39%)||32/91 (35%)|
| Age 13–17 yrs||7/21 (33%)||4/34 (12%)||11/55 (20%)|
Impact of YMDD Mutations.
The incidence of YMDD mutations in the treatment group previously treated with placebo increased from 0% at baseline to 19% (13/70) at month 12 and 49% (34/70) at month 24. In the treatment group previously treated with lamivudine, the incidence of YMDD mutations increased from 24% (29/121) at baseline to 59% (68/115) at month 12 and 64% (66/103) at month 24.
Two of the five subjects in the treatment arm with on-treatment serum ALT elevations of at least 2 times the upper limit of normal and greater than 500 IU/L had YMDD mutations at month 24. No patient had hepatic decompensation.
Table 3 details the efficacy results at month 24 for patients based on the presence or absence of YMDD mutations. The VR rates were lower in patients who had developed YMDD mutant HBV (5/100 [5%]) compared with those who had not (39/72 [54%]). For most end points an increase in efficacy was seen with longer treatment for the patients that did not develop YMDD mutations.
Table 3. Efficacy Results at Month 24 by YMDD Variant Status at Month 24
|Virological response||16/35 (46%)||4/34 (12%)||23/37 (62%)||1/66 (2%)|
|ALT normalization||26/34 (76%)||9/31 (29%)||25/36 (69%)||19/62 (31%)|
|Virological response and anti-HBe acquisition||16/35 (46%)||4/34 (12%)||22/37 (59%)||1/66 (2%)|
|HBeAg seroconversion||20/35 (57%)||4/34 (12%)||25/37 (68%)||4/66 (6%)|
|HBeAg loss||20/35 (57%)||4/34 (12%)||26/37 (70%)||4/66 (6%)|
|HBV DNA undetectable*||28/36 (78%)||6/34 (18%)||29/37 (78%)||1/66 (2%)|
Combined Study Results.
The combined efficacy results for the initial 12-month study period and the 24-month extension period are presented in Table 4.
Table 4. Combined Efficacy in All Treated Patients
|Virological response||25/79 (32%)||28/134 (21%)||.013|
|HBeAg seroconversion||28/79 (35%)||34/134 (25%)||.013|
|HBV DNA undetectable||38/79 (48%)||37/134 (28%)||<.001|
|ALT normalization||43/79 (54%)||52/134 (39%)||.002|
|Combined response||23/79 (29%)||27/134 (20%)||.013|
This shows the key efficacy results for patients enrolled in the extension study based on whether they received treatment with lamivudine for 24 months (i.e., patients who received placebo in the initial study) or 36 months (i.e., patients who received lamivudine in the initial study).
Patients who were treated with lamivudine for a total of 24 months had significantly higher rates of efficacy than those treated for 36 months. This result appears to be due to a higher incidence of YMDD mutant HBV in patients treated for 36 months compared with those treated for 24 months (64% vs. 49%, respectively).
Durability of Response in the Observational Arm.
Overall, the majority of patients in the observational arm had a virological or combined response at month 24. The proportion of patients with virological or combined response was similar in patients who had been treated previously with lamivudine compared with those who had received placebo (Table 5).
Table 5. Secondary Efficacy End Points at Month 24 (Observation Arm)
|ALT normalization||13/14 (93%)||35/49 (71%)||48/63 (76%)||.74|
|Virological response and anti-HBe acquisition||12/13 (92%)||34/39 (87%)||46/52 (88%)||.554|
|HBeAg seroconversion*||14/14 (100%)||37/44 (84%)||51/58 (88%)||.957|
|HBeAg loss||14/14 (100%)||39/46 (85%)||53/60 (88%)||.958|
|HBV DNA undetectable†||12/13 (92%)||38/44 (86%)||50/57 (88%)||.413|
|Combined response||12/13 (92%)||34/41 (83%)||46/54 (85%)||.341|
Similar response rates were also seen for each of the three age groups. In addition, 48 of 63 patients (76%) had a normal serum ALT level at month 24 (Table 5).
Seven patients (11%) in the observational arm were retreated with lamivudine following recurrence of active hepatitis B, which was characterized by increasing serum HBV DNA and ALT concentrations. Four patients developed detectable levels of HBeAg before being retreated. HBeAg levels subsequently became undetectable in 2 of these patients during treatment with lamivudine. Six of the seven patients treated responded well to retreatment with lamivudine. However, in three cases there was a recurrence of active hepatitis when therapy was stopped.
In the treatment group, 23 of 213 patients (11%) discontinued the study drug prematurely. Of these, 3 (4%) patients had previously received placebo and 20 (15%) patients had received lamivudine. The main reasons for premature discontinuation were lack of efficacy (6/213 [3%]), adverse events (3/213 [1%]), and loss to follow-up (3/213 [1%]).
The most commonly reported adverse events reported on treatment were ear, nose, and throat infections (41/213 [19%]); headaches (38/213 [18%]); cough (37/213 [17%]); viral ear, nose, and throat infections (37/213 [17%]); temperature regulation disturbances (32/213 [15%]); abdominal discomfort and pain (30/213 [14%]); and nausea and vomiting (28/213 [13%]).
One subject died after being accidentally electrocuted. The event was considered unrelated to the study drug by the investigator. Nonfatal serious adverse events were reported in 16 of 213 (8%) treated patients and in 4 of 63 (6%) patients in the observation group. None of these events was considered by the investigator to be related to the study drug.
The overall incidence of adverse events in the observational group was 79% (50 of 63 patients). There were no differences in the frequency and nature of adverse events observed between the treatment and observation groups.
In addition, there were no differences in the frequency and nature of adverse events observed between patients who developed YMDD mutations and those that did not.
There were 7 (3%) patients in the treatment arm and 4 (6%) patients in the observation arm who had a serum ALT elevation of at least 2 times their baseline value and greater than 500 IU/L. Two of the patients in the treatment arm had developed YMDD mutations. None of these events was associated with hyperbilirubinemia or other signs of hepatic decompensation.
When the results of the two study periods were combined, the overall incidence of adverse events observed in patients receiving either 24 or 36 months of lamivudine was similar (85% vs. 81%, respectively).
In the initial phase of this study, it was reported that 52 weeks of therapy with lamivudine resulted in a higher VR rate than placebo, which was similar to responses previously obtained with interferon.5, 12–15 Treatment with lamivudine was not associated with any significant side effects. However, to enable pediatricians to be able to use this agent appropriately, it was important to evaluate treatment over a longer duration, and to evaluate the durability of the treatment response obtained.
The present study shows that the response rates increased with prolonged duration of treatment. However, a longer duration of lamivudine therapy was also associated with a progressive increase in the incidence of YMDD mutations. As observed in adult patients, there was a negative correlation between VR and the incidence of YMDD mutations. Only 5% of patients in the treatment arm who developed YMDD mutations achieved a VR, compared with 54% of patients who retained wild-type HBV. Patients who developed YMDD mutations also had lower rates of ALT normalization, HBeAg loss and seroconversion, HBV DNA undetectability, and combined response.
Importantly, the durability of treatment response in children treated with lamivudine in the initial 1-year study was very high. This is consistent with previous findings in adults with chronic hepatitis B, where HBeAg seroconversion has been demonstrated to be durable in the majority of patients following the cessation of treatment.16 There appeared to be no significant difference in the relapse rates observed between patients who had spontaneously achieved a VR while receiving placebo and those that responded to treatment with lamivudine.
In a recent review, it has been suggested that younger children with chronic hepatitis B may have a better response to interferon treatment than older children.17 In the present study, response rates also appeared to be slightly higher in children aged 2–6 years. However, there appeared to be no correlation observed between age and response to therapy, although VR rates were notably lower in children aged 13–17 years. No firm conclusions regarding the relationship between age and treatment response can be drawn from our study.
In previous studies of lamivudine, a relationship between baseline ALT serum levels and response to therapy has been demonstrated.11 In this study, VR, HBeAg seroconversion, and HBeAg loss response rates appeared to be greater in patients who had a serum ALT concentration greater than 5 times the upper limit of normal at baseline. This phenomenon is closely related to the immunological status of infected patients. Patients with elevated aminotransferase levels are the most logical population for treatment, because they are less tolerant to HBV infection and have a higher chance of response to therapy.
The cumulative efficacy of lamivudine therapy with regard to VR and ALT normalization is comparable to that previously observed in children treated with interferon α for 6 months.5, 11–15 Treatment with interferon α has the advantage of a finite duration and does not result in the emergence of HBV mutations associated with resistance. On the other hand, interferon α treatment is expensive and is associated with significant side effects, including transient growth impairment.4 Lamivudine treatment is more cost-effective over a 1-year period, but longer durations of treatment are associated with an increasing risk of drug resistance and increased cost, which may negate some of the initial benefits.
Because this study was not a randomized trial, no conclusions can be drawn as to the benefit of continued treatment versus cessation of treatment in children who are still HBeAg-positive after 1 year of lamivudine therapy. However, further treatment did result in additional patients achieving a VR, HBeAg seroconversion, HBV DNA undetectability, and/or ALT normalization. Although histology was not assessed in this study, based on the long-term experience of adults treated with lamivudine, it is likely that patients who achieved durable HBV DNA suppression also obtained some improvement in histology.18
Lamivudine was well tolerated in this study. The incidence of adverse events was similar in patients in the treatment group compared with patients in the observation group receiving no treatment. Continued treatment with lamivudine for up to 3 years did not reveal any new or unexpected adverse events or laboratory abnormalities.
The results of this study are similar to those previously reported in adults treated with lamivudine. In addition to the long-term benefits of seroconversion, successful treatment can result in significant social and psychological benefits and can reduce the risk of transmission of HBV among teenagers and young adults.
Because the long-term consequences of YMDD mutation on the natural history of the disease remains unknown, we recommend that treatment with lamivudine is considered only in the subgroup of patients that have the highest chance of a response. Most children do not suffer physical symptoms of the disease and are unlikely to suffer any short-term consequences from not being treated.
On the basis of these results, including those from the initial study period, we recommend that children with HBeAg-positive chronic hepatitis B with elevated serum aminotransferase levels of more than 2 times the upper limit of normal are potential candidates for therapy. Treatment should be continued, for up to 36 months, until VR, with or without seroconversion, is achieved. In view of the poor long-term response, we recommend that treatment is discontinued if YMDD mutations emerge. It is likely that most patients will tolerate the discontinuation of therapy in view of their mild underlying disease, and that they will probably return to their pretreatment disease status. However, because this approach has not been formally evaluated, we recommend that liver function tests be carefully monitored following cessation of treatment.
In conclusion, VR was maintained in most patients who had responded to lamivudine in the initial study. Additional clinical response was seen over the subsequent 24-month study period in children who had not initially achieved a VR after 12 months of lamivudine treatment. However, the incidence of YMMD mutations increased over time, and resulted in lower response rates in these patients. Despite the encouraging rate of response in children treated with lamivudine, more than 60% of them did not meet the study endpoint. This observation suggests the need for better selection of candidates that will benefit from treatment, and secondly the need for new therapeutic strategies, such as combination antiviral therapy, that will provide more potent viral suppression, and present a higher barrier to the emergence of resistance.