Peginterferon Alfa-2a (40KD) Plus Lamivudine or Entecavir in Children With Immune-Tolerant Chronic Hepatitis B

Abstract

Objective: 

Treatment guidelines for chronic hepatitis B (CHB) do not recommend antiviral therapy for patients in the immune-tolerant phase of the disease, which generally occurs in children who acquire hepatitis B virus (HBV) vertically and may last for decades. On the basis of promising results of a pilot study, we conducted a randomized, controlled, multicenter study to evaluate the efficacy and safety of antiviral therapy in children and adolescents with immune-tolerant CHB.

Methods: 

Fifty-nine children aged 3 to <18 years hepatitis B e antigen-positive with an HBV DNA titer >20,000 IU/mL and persistently normal alanine aminotransferase levels were randomized to 56 weeks of antiviral therapy with an oral nucleoside analogue [entecavir or lamivudine], combined with subcutaneous peginterferon alfa-2a from week 8, or 80 weeks of untreated observation. The primary efficacy outcome was hepatitis B surface antigen loss 24 weeks post-treatment in the antiviral therapy group or at the end of observation in the control group.

Results: 

Enrollment was terminated after the results of two similar studies showed that similar antiviral regimens were ineffective in children and adults with immune-tolerant CHB. At 24 weeks post-treatment, 1 of 26 patients in the antiviral treatment group experienced HBsAg loss (vs none of 33 patients in the control group). No serious treatment-related adverse events were reported, and no patients discontinued treatment because of adverse events.

Conclusions: 

The antiviral regimen evaluated in this trial had an acceptable tolerability profile, but was ineffective in children and adolescents with immune-tolerant CHB.

What Is Known/What Is New

What Is Known

• Guidelines for chronic hepatitis B do not recommend therapy for patients with the immune-tolerant disease.
• Two multicenter, non-comparative, single-arm clinical trials in patients with immune-tolerant chronic hepatitis B reported that treatment with entecavir plus peginterferon alfa-2a did not modify the natural history of the disease in adults or children.

What Is New

• The present study included an untreated observation group.
• The similar rates of hepatitis B e (HBe) loss, anti-HBe seroconversion, and hepatitis B virus DNA suppression in the observation group highlight the lack of efficacy of available antiviral therapies for children with immune tolerant chronic hepatitis B.

The immune-tolerant phase of chronic hepatitis B (CHB), described as “phase I” in the 2017 European Association for the Study of the Liver guidelines ⁽¹⁾ is a “high-replicative, low-inflammatory” ⁽²⁾ state where minimal inflammation is detectable on liver biopsy, alanine aminotransferase (ALT) levels remain in the normal or near-normal range, and high rates of viral replication are shown by hepatitis B e antigen (HBeAg) positivity and high hepatitis B virus (HBV) DNA levels (>10⁷ IU/mL) ⁽³⁾ in serum. Treatment guidelines do not recommend antiviral therapy for immune-tolerant CHB because of the limited extent of fibrosis and inflammation and limited efficacy of available therapies ^(1,4); however, persistently high HBV DNA levels are a risk factor for hepatocellular carcinoma, and the ability to modify the natural history of immune-tolerant CHB may improve long-term outcomes.

Based on the promising results of a small pilot study ⁽⁵⁾, a randomized study was initiated to evaluate the impact of antiviral therapy in pediatric patients with immune-tolerant CHB. The objective was to compare hepatitis B surface antigen (HBsAg) loss rates in treated patients with those in an untreated control group. Although the study was ongoing, the results of two trials in patients with immune-tolerant CHB were presented in October 2017 ^(6,7) and then published in full ^(8,9). These studies reported that treatment with entecavir monotherapy for 8 weeks, followed by 40 weeks of combination therapy with entecavir plus peginterferon alfa-2a, did not modify the natural history of immune-tolerant CHB ^(8,9). Because of the similar study design and treatment regimens used in the present study and in the two trials ^(8,9), the Data Safety Monitoring Board concluded that our trial was unlikely to demonstrate efficacy and on March 22, 2018 recommended that the trial be stopped. The sponsor decided to terminate recruitment on March 28, 2018. Patients enrolled in the trial were allowed to complete treatment and were followed up for 1 year after the end of treatment. We report the outcome of treatment in these patients here.

METHODS

Patients

Children ages 3 to <18 years with untreated immune-tolerant CHB, who were HBsAg-positive, HBeAg-positive and had an HBV DNA titer >20,000 IU/mL were eligible for the trial. Eligible patients were required to have a normal ALT level at screening, and either no or minimal fibrosis on a liver biopsy performed within 2 years before baseline and stable normal ALT levels (defined as < upper limit of normal [ULN] for the local laboratory) on at least two separate occasions (at least 1 month apart) over the 6 months before baseline; or stable normal ALT levels on at least three separate occasions at least 1 month apart (over the 1 year before baseline), and no signs of hepatocellular carcinoma, advanced fibrosis/cirrhosis or splenomegaly by ultrasonography at the screening.

Patients were excluded if they were co-infected with hepatitis A, C, or D virus, human immunodeficiency virus, or if they had decompensated liver disease.

Study Design and Treatment

Patients, stratified by HBV genotype, were assigned randomly to an active treatment group or an untreated control group by an interactive voice or Web-based response system. Those patients, randomized to the active treatment group, received 56 weeks of antiviral therapy with an oral nucleoside analogue (entecavir or lamivudine, investigator choice), combined with subcutaneous peginterferon alfa-2a from week 8 (Supplemental Digital Content 1, https://links.lww.com/MPG/C292). Choice of nucleoside analogue was at the investigator's discretion.

Lamivudine was given at a dosage of 3 mg/kg once daily (maximum 100 mg/day). Entecavir was given at a dosage of 0.015 mg/kg once daily (maximum 0.5 mg/day). Peginterferon alfa-2a was administered subcutaneously once weekly according to body surface area (BSA) [45 μg (BSA 0.51–0.53 m²); 65 μg (BSA 0.54–0.74 m²); 90 μg (BSA 0.75–1.08 m²); 135 μg (BSA 1.09–1.51 m²); or 180 μg (BSA > 1.51 m²)].

Patients were to be followed up for 5 years after the end of treatment; however, after recruitment was terminated prematurely, the duration of follow-up was reduced to 1 year by a protocol amendment.

Outcomes

The primary efficacy outcome of the study was the loss of HBsAg at 24 weeks post-treatment or at the end of the untreated follow-up (study week 80). Secondary efficacy outcomes included: anti-HBs seroconversion, HBeAg loss, anti-HBe seroconversion, and quantitative HBV DNA levels (<20,000 IU/mL, <2000 IU/mL, and undetectable) at 24 weeks post-treatment or the end of the untreated observation period; HBsAg loss, anti-HBs seroconversion, HBeAg loss, anti-HBe seroconversion, and quantitative HBV DNA levels (as above) after 1 year post-treatment in the antiviral treatment group.

Safety was evaluated by adverse events, laboratory test results, vital signs and anthropometric parameters (height and weight) in patients in the active treatment group up to 24 weeks post-treatment and at the end of observation (week 80) in the untreated control group.

Emergence of resistance to lamivudine or entecavir were exploratory endpoints. Resistance assays were performed by DDL Diagnostic Laboratory (Rijswijk, the Netherlands).

Statistics

Based on the results of a pilot study of interferon ⁽⁵⁾, and from reports of spontaneous loss of HBsAg ⁽¹⁰⁾, it was assumed that 17% of patients randomized to peginterferon alfa-2a and 0.6% of those in the untreated control group would experience loss of HBsAg at post-treatment week 24/end of the untreated observation (study week 80). With these assumptions, 50 evaluable patients per group would be required to detect a statistically significant difference using Fisher exact test with an alpha level of 0.05 (two-sided) and a beta level of 0.20 (80% power). Approximately 114 patients were planned to be recruited to ensure at least 50 evaluable patients in each group. The efficacy analysis was planned to be conducted in the intention-to-treat population; however, after recruitment was terminated, the statistical analysis was modified such that no formal hypothesis testing or treatment group comparisons were conducted. Data were summarized using descriptive statistics.

Ethics

The study was conducted in conformance with guidelines for good clinical practice and the principles of the Declaration of Helsinki. The protocol and all amendments were reviewed and approved by the Institutional Review Board or Ethics Committee at each participating institution. Parents or legal guardians and/or patients who were considered legally to be adults were required to provide informed written consent. Children and young people also provided age/development appropriate assent.

RESULTS

The first patient was enrolled on February 26, 2007, recruitment was terminated on March 28, 2018 after enrollment of 62 patients, and the last enrolled patient completed follow-up on January 29, 2020. The study took place at 22 study centers across 12 countries and territories.

Three patients were randomized to peginterferon alfa-2a monotherapy prior to a protocol amendment that discontinued enrollment in this group. Data from these three patients are not included in this report. Twenty-six patients were randomized and treated with antiviral therapy (nucleoside plus peginterferon alfa-2a), and 33 patients were randomized to observation (Fig. 1). Twelve patients received entecavir plus peginterferon and 14 received lamivudine plus peginterferon. All 26 patients in the antiviral therapy group completed 56 weeks of treatment, and 25 of 26 patients completed 1 year of follow-up. Among the 33 patients in the untreated observation group, 24 completed 80 weeks of observation. Eight patients withdrew from the observation group for non-safety reasons, none of whom completed follow-up. In addition, one patient in the observation group whose last visit fell outside the week 80 visit window was not considered to have completed the follow-up.

FIGURE 1: Patient flow diagram. NA = nucleoside analogue.

The characteristics of patients randomized to antiviral therapy and untreated observation were generally similar (Table 1).

TABLE 1 - Baseline demographic and disease characteristics

Characteristic	Peginterferon alfa-2a plus nucleoside analogue (n = 26)	Untreated observation (n = 33)
Mean age, y (SD)	11.9 (3.2)	10.9 (3.7)
Age group, y, n (%)
<6	2 (7.7)	3 (9.1)
6 to <12	6 (23.1)	14 (42.4)
>12	18 (69.2)	16 (48.5)
Male/female, n (%)	10/16 (38.5/61.5)	19/14 (57.6/42.4)
Race
Asian	11 (42.3)	11 (33.3)
White	9 (34.6)	17 (51.5)
Black	4 (15.4)	2 (6.1)
Other	2 (7.7)	3 (9.1)
Mean weight (kg) (SD)	46.95 (13.50)	42.78 (14.71)
BSA category (m2), n (%)
<0.75	0	0
0.75–1.08	4 (15.4)	9 (27.3)
1.09–1.51	15 (57.7)	10 (30.3)
>1.51	7 (26.9)	14 (42.4)
Mode of HBV acquisition, n (%)
Perinatal	19 (73.1)	23 (69.7)
Transfusion	1 (3.8)	0
Unknown	6 (23.1)	8 (24.2)
Other	0	2 (6.1)
Mean HBV DNA, log10 IU/mL (SD)	8.02 (0.93)	8.22 (1.07)
Mean HBsAg, log10 IU/mL (SD)	4.71 (0.32)	4.89 (0.35)
Mean HBeAg, log10 IU/mL (SD)	1.66 (0.26)	1.70 (0.30)
ALT (IU/L)
Mean (SD)	22.23 (10.07)	25.32 (13.11)
Median (range)	19.50 (13.0, 59.0)	25.00 (9.0, 76.0)
HBV genotype, n (%)
A	1 (3.8)	2 (6.1)
B	5 (19.2)	6 (18.2)
C	5 (19.2)	4 (12.1)
D	10 (38.5)	15 (45.5)
E	4 (15.4)	5 (15.2)
Missing	1 (3.8)	1 (3.0)

ALT = alanine aminotransferase; BSA = body surface area; HBeAg = hepatitis B e antigen; HBsAg = hepatitis B surface antigen; HBV = hepatitis B virus; SD = standard deviation.

Efficacy

HBsAg loss and HBs-seroconversion occurred in one patient in the antiviral therapy group (3.8%, 95% CI 0.10, 19.64%) after 24 weeks and 1 year post-treatment (Table 2). This was a 4-year-old white male with HBV genotype D infection, a baseline HBV DNA level of 8.51 log₁₀ IU/mL, and a baseline ALT level of 20 IU/mL (normal 5–30 IU/mL). Persistent ALT elevations occurred in this patient: with the highest (217 IU/mL) day 224. HBsAg loss was first observed at week 56 (day 352, confirmed on day 395) and the patient remained HBsAg-negative throughout follow-up. Anti-HBs-antibodies were detected at follow-up weeks 24 and 52. HBeAg loss was first observed at follow-up week 4 and the patient remained HBeAg-negative throughout follow-up. HBV DNA titer was undetectable at his latest evaluation (follow-up week 4) and his last two ALT determinations were 11 IU/mL on day 427 and 8 IU/mL on day 569.

TABLE 2 - Primary and secondary efficacy endpoints at 24 weeks post-treatment in the antiviral therapy group and end of observation in the untreated observation group (intention-to-treat population)

Endpoint	Peginterferon alfa-2a plus nucleoside analogue (n = 26)	Untreated observation (n = 33)
HBsAg loss (primary endpoint)	1 (3.8)	0
HBs seroconversion	1 (3.8)	0
HBeAg loss	1 (3.8)	4 (12.1)
HBe seroconversion	0	3 (9.1)
HBV DNA <20,000 IU/mL	2 (7.7)	4 (12.1)
HBV DNA <2000 IU/mL	2 (7.7)	4 (12.1)
HBV DNA undetectable	0	2 (6.1)
HBV DNA <2000 IU/mL plus HBeAg seroconversion	0	3 (9.1)

HBeAg = hepatitis B e antigen; HBsAg = hepatitis B surface antigen; HBV = hepatitis B virus.

Five patients in the antiviral treatment group met at least one efficacy endpoint (Supplemental Digital Content 2, https://links.lww.com/MPG/C292). At 24 weeks post-treatment, one patient (3.8%) had HBeAg loss, and at 1 year post-treatment, three patients (11.5%) had HBeAg loss, and two (7.7%) had HBe seroconversion. Two patients had HBV DNA levels <2000 IU/mL at 24 weeks and 1 year post-treatment, and two patients had HBe seroconversion and HBV DNA <2000 IU/mL at 1 year post-treatment. ALT elevations during treatment occurred in two patients who experienced HBeAg loss and the combined endpoint of HBe seroconversion and HBV DNA <2000 IU/mL.

HBsAg loss was not observed at the end of observation (week 80) in any of the 33 patients in the untreated observation group; however, four patients met at least one efficacy endpoint (Supplemental Digital Content 2, https://links.lww.com/MPG/C292). Four patients (12.1%) lost HBeAg, three patients (9.1%) experienced HBe seroconversion, four patients (12.1%) had HBV DNA levels <2000 IU/mL and three patients (9.1%) had HBe seroconversion and HBV DNA <2000 IU/mL.

At baseline, mean HBV DNA levels were similar in patients randomized to antiviral therapy (8.02 log₁₀/mL) or untreated observation (8.22 log₁₀/mL). In the antiviral therapy group the mean HBV DNA level decreased to 4.74 log₁₀/mL at study week 8 (end of nucleoside analogue monotherapy) and to 2.21 log₁₀/mL at study week 56 (end of combination therapy), but rebounded after the end of treatment (Supplemental Digital Content 3, https://links.lww.com/MPG/C292).

Exploratory Endpoints

During treatment and follow-up HBsAg levels, HBeAg levels and HBV DNA levels were similar in patients who received lamivudine or entecavir (Supplemental Digital Content 4, https://links.lww.com/MPG/C292).

Resistance to lamivudine or entecavir was not detected in any patient.

Safety

Twenty-four patients in the antiviral therapy group (92.3%) and 15 patients in the untreated observation group (45.5%) reported at least one adverse event (AE). Nineteen patients in the antiviral therapy group (73.1%) reported AEs that were related to study treatment in the opinion of the investigator. Three patients in the antiviral therapy group experienced AEs that led to dose modifications. No patients in the antiviral therapy group experienced serious AEs. One patient in the untreated observation group experienced two serious AEs (pharyngitis and dehydration).

The most common AEs (regardless of cause) in the antiviral therapy group were headache (53.8%), pyrexia (42.3%), and fatigue (19.2%). The most common AEs in the observation group were upper abdominal pain (6.1%) and influenza (6.1%). The most common treatment-related AEs are shown in Table 3.

TABLE 3 - Treatment-related adverse events reported in at least two patients during antiviral treatment and 24 weeks of post-treatment follow-up in the antiviral therapy group (n = 26)

Event	No. of patients (%)
Headache	10 (38.5)
Pyrexia	10 (38.5)
Fatigue	5 (19.2)
Nausea	3 (11.5)
Alopecia	2 (7.7)
Asthenia	2 (7.7)
Decreased appetite	2 (7.7)
Dizziness	2 (7.7)
Epistaxis	2 (7.7)
Hyperthermia	2 (7.7)
Injection site bruising	2 (7.7)
Migraine	2 (7.7)
Myalgia	2 (7.7)
Pain	2 (7.7)
Pain in extremity	2 (7.7)
Vomiting	2 (7.7)

In the antiviral therapy group, four patients (15.4%) experienced Grade 3 ALT elevations and one patient (3.8%) experienced a Grade 4 ALT elevation post-baseline.

No patients discontinued antiviral treatment because of AEs or laboratory abnormalities. Laboratory abnormalities leading to dose modification of peginterferon alfa-2a included low neutrophil counts (two patients) and elevated serum ALT levels (two patients).

Five of 26 patients in the antiviral therapy group (19.2%) and four of 24 patients in the observation group (12.9%) had a decrease of >15% from baseline in height for age percentile at the end of follow-up.

DISCUSSION

The final results of this study indicate that in children in the immune-tolerant phase of CHB, treatment for 8 weeks with a nucleoside analogue followed by 48 weeks of combination therapy with peginterferon alfa-2a plus a nucleoside analogue does not modify the natural history of the disease. HBsAg loss at 24 weeks post-treatment occurred in 1 of 26 treated patients. This result is consistent with the results of the two US NIH-sponsored trials that led to the termination of recruitment in this trial ^(8,9). In the trial by Rosenthal et al ⁽⁹⁾, 2 of 60 children achieved the primary endpoint of HBeAg loss plus HBV DNA <1000 IU/mL 48 weeks post-treatment. The trial by Feld et al ⁽⁸⁾ employed the same primary endpoint; however, none of the 28 adult patients in the trial experienced HBeAg loss and HBV DNA suppression.

The results of the present study and the NIH-sponsored studies ^(8,9) differ markedly from two previous studies. In the study by D’Antiga et al ⁽⁵⁾ five of 23 children experienced HBeAg seroconversion and four experienced HBsAg loss and HBs-seroconversion during treatment, which persisted during long-term follow-up. Poddar et al ⁽¹¹⁾ enrolled 28 children in a case–control study. Six months after completion of therapy, durable HBeAg seroconversion occurred in 10 treated children compared with two children in an untreated control group, and HBsAg loss and seroconversion occurred in six treated children (vs none in the control group). In contrast, HBsAg loss was documented in only one of 26 children (3.8%) in the present trial, in two of 60 children (5%) in the trial by Rosenthal et al ⁽⁹⁾ and in none of 28 adults in the trial by Feld et al ⁽⁸⁾. The reason for the higher response rates in the D’Antiga et al ⁽⁵⁾ and Poddar et al ⁽¹¹⁾ studies is unclear.

Children aged 5 years or younger have been reported to be more likely to respond to interferon-based treatment compared with older children during phase 2 of HBV infection, that is, during the immune-active phase of the disease ⁽¹²⁾. The authors suggested that this could be due to a shorter duration of infection, as studies in adults had reported that patients with a shorter duration of infection and/or a history of acute infection with elevated aminotransaminase levels were more likely to clear HBsAg after IFN treatment ⁽¹³⁾. Whether this is the same for children treated during phase 1 of HBV infection is unknown. The only responder in the present study was 4 years old and the two children who lost HBsAg in the study by Rosenthal et al ⁽⁹⁾ were 5 and 10 years of age. D’Antiga et al ⁽⁵⁾ did not find an association between response and age, although few children in the study were aged younger than 5 years, but also reported that response was associated with the treatment-induced transition from immunotolerance to HBV immune control ⁽¹⁴⁾ and to lower baseline HBsAg plasma levels, lower HBsAg expression in liver and on-treatment decline of plasma HBsAg levels ⁽¹⁵⁾.

The inclusion criteria of the present study were designed carefully to include children with immune-tolerant CHB. D’Antiga et al ⁽⁵⁾ enrolled children with persistently normal or near-normal ALT levels, and Poddar et al ⁽¹¹⁾ enrolled children with two or three consecutive normal or near-normal ALT values in the 6 months before treatment; in both studies, all patients had a pre-treatment liver biopsy to rule out active disease. Although it is unlikely that the marked differences in treatment response are due to differences in the interferon formulations employed (both D’Antiga et al ⁽⁵⁾ and Poddar et al ⁽¹¹⁾ used conventional interferon-alfa 2b, whereas we used peginterferon alfa-2a), they may be related to patients’ ethnicity and HBV genotype. In the present study, five of the 25 (20%) treated patients with known HBV genotype were infected with HBV genotype B, whereas in the study by D’Antiga et al ⁽⁵⁾, 14 of 23 children (61%) were infected with HBV genotype B and 17 (74%) were of Asian ethnicity. HBV genotype B is associated with a higher rate of HBeAg loss in response to treatment in Asian populations ⁽¹⁶⁾.

In the present study, mean HBV DNA levels decreased by approximately 6-log₁₀ from baseline to the end of treatment, but returned to near baseline levels by week 24 of follow-up. This pattern is consistent with the effects of nucleoside analogues. ALT flares may occur during treatment before a response, as observed in our study and in the study by Rosenthal et al ⁽⁹⁾. In addition, ALT flares are not unusual after the withdrawal of nucleoside analogues or after the emergence of drug-resistant mutants in patients with immune-active CHB ^(17,18). The absence of ALT flares in all patients after the withdrawal of nucleoside analogue therapy has been reported in other trials in immune-tolerant patients ^(5,8,9,11).

Both US NIH studies had planned originally to randomize patients to an antiviral treatment group or an untreated observation group. Both studies had difficulty recruiting patients; as a result, the observation group was discontinued and the trials proceeded as non-comparative, single-arm studies ^(8,9). Thus, the inclusion of an untreated observation group in the present study is a strength. The similar rates of HBe loss, anti-HBe seroconversion and HBV DNA suppression in the observation group highlight the lack of efficacy.

Recruitment to this study was challenging as children with immune-tolerant CHB are clinically well; invasive treatment with an unknown outcome and frequent hospital visits were perceived as excessive by several families.

The AE profile of the treatment regimen was consistent with the known safety profiles of nucleoside analogues and peginterferon alfa-2a, which have been reported in previous trials with similar regimens in children ^(5,8,9,11). Of note, the results also showed no difference between treated and untreated patients in growth stunting, a side-effect often associated with interferon treatment ⁽¹⁹⁾.

In conclusion, the results of this trial indicate that 8 weeks of nucleoside analogue monotherapy followed by 48 weeks of peginterferon alfa-2a plus nucleoside analogue combination therapy does not modify the natural history of immune-tolerant CHB in children. Further research is required to develop therapies for patients with immune-tolerant CHB. Development of these agents should include consideration of the needs of children and young people.