Natural History of Chronic Hepatitis B Infection Among Chinese Children and Young Adults: A Single-Center Experience

Abstract

Objectives: 

There are limited data on the natural history of chronic hepatitis B virus (HBV) infection in Asian American children. The aim of the present study was to describe a single-center experience of chronic HBV infection in Chinese American patients in New York City.

Methods: 

A retrospective chart review was conducted for patients with chronic HBV infection who had pediatric visits from 2006 to 2017. Clinical and laboratory data were collected to characterize the status of HBV infection and its disease course both cross-sectionally and longitudinally. Available maternal charts were reviewed.

Results: 

Of the total 353 patients, 72 patients (20%) were US-born. Positive hepatitis B envelope antigen (HBeAg) was documented in 208 patients (58%). Three phases of chronic HBV infection were categorized for 329 patients: immune-tolerant 112 (34%), HBeAg-positive immune-active 47 (14%), and inactive carrier 82 (25%). The remaining 88 patients (27%) did not fit into a particular category with 26 of 88 patients meeting the criteria for inactive carrier except for mildly elevated alanine aminotransferase. Age and liver enzyme levels were significantly different between HBeAg-positive and HBeAg-negative groups (P < 0.05). Among 179 patients followed for ≥5 years, the spontaneous seroconversion rate was 38%. In eight patients with linked maternal data, all children completed the HBV vaccine series and seven of eight received hepatitis B immunoglobulin. All mothers were HBeAg-positive with high HBV DNA and had no anti-viral therapy during pregnancy.

Conclusions: 

Both immune-tolerant and inactive carrier phases were common for chronic HBV infection with a spontaneous seroconversion rate of 38%. All US-born patients were born in the era of implemented universal immune-prophylaxis.

What Is Known/What Is New

What Is Known

• Hepatitis B virus (HBV) infection in children is mainly transmitted vertically, and the spontaneous hepatitis B envelope antigen seroconversion varies depending on age and genotype among other factors.
• Asian Americans account for >50% of over (be more consistent with introduction statement) 1 million Americans with chronic HBV infection.
• There are limited data on the natural history and disease progression of HBV infection in foreign and US-born children of Chinese ethnicity.

What Is New

• In a large cohort of foreign and US-born children and young adults of Chinese ethnicity with chronic HBV infection in a single-center, both immune-tolerant and inactive carrier phases were common.
• Spontaneous seroconversion rate was 38% for those who were followed ≥5 years.
• More can be done to promote American Association for the Study of Liver Diseases guidelines of anti-viral therapy for chronic HBV infected women with high HBV viral loads.

Asian Americans comprise only 6% of the US population but half of over 1 million Americans with chronic hepatitis B virus (HBV) infection ^(1,2). The prevalence of HBV infection is approximately 12% in foreign-born Chinese Americans ⁽³⁾. There are limited data on the natural history of the infection in Asian American children in the era of implemented universal immune-prophylaxis.

The natural history of chronic HBV infection in children includes three main phases: immune-tolerant, hepatitis B envelope antigen (HBeAg)-positive immune-active, and inactive carrier ^(4–6). It is known that infected children are usually in the immune-tolerant phase at younger ages, and the chance of HBeAg spontaneous seroconversion increases with age ⁽⁷⁾. It is also known that many factors including viral factors, host factors, and host–virus interactions may influence the onset of HBeAg seroconversion ⁽⁸⁾; however, the timing for HBeAg seroconversion is very difficult to predict, posing challenges in the clinical practice. A recent study reported HBV genotypes and chronic HBV infection phenotypes using new cutoff alanine aminotransferase (ALT) values in children living in the United States and Canada ⁽⁹⁾. The change of cutoff ALT may impact the classification of chronic HBV phases.

In endemic areas such as China, the major route of HBV infection in children is the vertical transmission ^(10,11). Despite universal immune-prophylaxis, vertical transmission remains evident and is associated with maternal HBV DNA levels ^(12,13). In an observational study done within Kaiser Permanente Northern California, the infant HBV infection rate per 100 birth was 3.37 for HBeAg-positive mothers and 3.61 for mothers with high HBV viral loads ⁽¹⁴⁾.

The present study's aim was to describe characteristics of chronic HBV infection in Chinese American children and young adults at Charles B. Wang Community Health Center (CBWCHC) in New York City.

METHODS

Study Design and Population

New York City has the largest Chinese American population in the United States. CBWCHC is a non-profit community-based health center established in 1971 that has been a trusted site for primary health care to the Chinese community in New York City. Due to high rates of chronic HBV infection in this community, CBWCHC has established an HBV registry for their patient population and also a Hep B Moms Program for pregnant women with chronic HBV infection.

A retrospective chart review was conducted for patients with chronic HBV infection, defined as the presence of hepatitis B surface antigen (HBsAg) for >6 months. Inclusion criteria for the study population were Chinese ethnicity and active patient status with at least one medical visit to the pediatric clinic at CBWCHC from January 1, 2006 to December 31, 2017. Patients ranged in age from 0 to 21 years at their initial pediatric visits with the diagnosis of chronic HBV infection. Some patients were retained in the study past 21 years of age since they remained active patients with medical visits to the CBWCHC adult service within three years of the study period end date. The patient's age used for analysis was the age at the time of their latest available laboratory data. During the study period, over 42,700 pediatric patients have been seen at the center and 0.8% of them had chronic HBV infection. The study was approved by the Institutional Review Board at the Montefiore Medical Center/Albert Einstein College of Medicine.

CBWCHC has utilized an electronic medical record system (athenahealth/GE Centricity) since 2006. Each patient's demographic characteristics, clinical history, family history of immediate family members, prior HBV vaccination, and laboratory values were extracted from patient charts.

Definition of Chronic HBV Infection Phases

The three phases of chronic HBV infection were analyzed cross-sectionally according to the following laboratory values: HBeAg status, anti-HBe status, ALT, and HBV DNA levels. The normal cutoff value of ALT was based on the American Association for the Study of Liver Diseases (AASLD) hepatitis B guidance 2018 ⁽¹⁵⁾. Specifically, the immune-tolerant phase was defined as HBeAg-positive, DNA ≥20,000 IU/mL, ALT ≤35 U/L (males), and ≤25 U/L (females). HBeAg-positive immune-active phase was defined as: HBeAg-positive, DNA ≥20,000 IU/mL, ALT >35 U/L (males), >25 U/L (females). Inactive carrier was defined as HBeAg(−), anti-HBe(+), DNA <2000 IU/mL, ALT ≤ 35 U/ L(males), and ≤25 U/L (females) ^(4,15). Patients not directly meeting the criteria for the three phases of chronic HBV infection were entered into an “other” category, which were further described and analyzed based on the collected laboratory values.

The age of patients at the time of the latest laboratory values was analyzed between HBeAg-positive and -negative groups. ALT and platelet count were compared between these two groups. The average aspartate aminotransferase (AST) to platelet ratio index (APRI) score with an upper limit of AST 40 U/L was calculated to determine the likelihood of liver fibrosis ⁽¹⁶⁾. Alpha-fetoprotein (AFP) were also documented for those patients with available laboratory values.

Longitudinal Analysis of a Subgroup

A longitudinal analysis was performed for a subset of patients who had follow up and had laboratory data for a minimum of 5 years. The duration of follow up, change of HBeAg status, and history of anti-viral therapy were collected. Completion of HBeAg seroconversion was interpreted as HBeAg seroclearance from positive to negative with the appearance of anti-HBe from negative to positive. To assess the longitudinal change in ALT, the baseline ALT and the latest ALT were compared using a non-parametric paired t-test. P < 0.05 were considered as statistically significant.

Maternal-Child Data Collection

Characteristics of HBV vertical transmission were studied in detail on US-born patients who had linked maternal charts at CBWCHC. The data were collected from their mother's charts including maternal HBV DNA levels, HBeAg status and anti-viral therapy status during/after pregnancy, and the type of delivery. Patient charts were reviewed for HBV vaccination and receipt of hepatitis B immunoglobulin (HBIG).

Statistical Analysis

Descriptive statistics was used to report basic demographics, clinical and laboratory test values of the study patients. Continuous variables were presented as mean and standard deviation (SD) for normally distributed data, and median and interquartile range (IQR) for skewed distributed data. Wilcoxon rank sum test was used to evaluate the association between patient's age and HBeAg status at the time of their serology testing. All statistical analyses were performed using SAS software version 9.4.

RESULTS

Patient Characteristics

There was a total of 353 patients with chronic HBV infection meeting inclusion criteria during the 12-year study period. The demographic characteristics of the study population are listed in Table 1. Of the study population, 263 patients (75%) were born in China, 72 patients (20%) were born in the United States, and 18 patients (5%) were born in countries other than China and the United States. A total of 234 patients (66%) were ≤21 years old and 119 patients (34%) were > 21 years old at the time of their latest available laboratory data. None of the patients were adopted, and no co-infection with HCV, HDV or HIV was noted.

TABLE 1 - Demographic characteristics of Chinese American patients

Characteristics	Study population (n = 353)
Age (y) at the most recent test
0–5	9 (2%)
6–11	32 (9%)
12–17	81 (23%)
18–21	112 (32%)
22–32	119 (34%)
Sex
Male	177 (50%)
Female	176 (50%)
Birth country
United States	72 (20%)
China	263 (75%)
Other	18 (5%)
Health insurance
Medicaid	251 (71%)
Other	102 (29%)
Family history of HBV
With maternal history	92 (26%)
No maternal history	38 (11%)
Family history undocumented	223 (63%)

HBV = hepatitis B virus.

Population Categorized by Chronic HBV Infection Phases

Three phases of chronic HBV infection categorized for 329 of the 353 patients are presented in Table 2. The remaining 24 patients were not analyzed due to missing HBV DNA values. There were 112 (34%) in the immune-tolerant phase, 47 (14%) in the HBeAg-positive immune-active phase and 82 (25%) in the inactive carrier phase. Among 329 patients, 88 patients (27%) in “other” did not fit into one of these main phases of chronic HBV infection. Of these 88 patients, 26 had characteristics similar to those in the inactive carrier phase, but had isolated mild elevation of ALT ranging from 26 to 82 U/L. Ten of the 88 patients also had characteristics similar to those in the inactive carrier phase, with HBV DNA ranging between 2000 to 5000 IU/mL. Two patients at ages >21 years manifested changing from the inactive carrier phase to markedly elevated HBV DNA of 45,480 and 8,670,000 IU/mL, respectively. The remaining 50 of 88 patients had a variety of other features inconsistent with each of the main phases such as partial HBeAg seroconversion (both positive or negative HBeAg and anti-HBe).

TABLE 2 - Chronic hepatitis B phases

	Number of patients (n = 329)	Age (y) Mean ± SD
Immune tolerant	112 (34%)	19 ± 6
HBeAg(+)
HBV DNA ≥20,000 IU/mL
ALT ≤35 U/L (male), ≤25 U/L (female)
Immune active	47 (14%)	19 ± 7
HBeAg (+)
HBV DNA ≥20,000 IU/mL
ALT >35 U/L (male), >25 U/L (female)
Inactive carrier	82 (25%)	20 ± 5
HBeAg(−), anti-HBe(+)
HBV DNA <2000 IU/mL
ALT ≤35 U/L (male), ≤25 U/L (female)
Other	88 (27%)	20 ± 6

ALT = alanine aminotransferase; HBeAg = hepatitis B envelop antigen; HBV = hepatitis B virus.

To further examine chronic HBV infection phases for 72 US-born patients (mean age ± SD 18 ± 7 years old), three main phases of chronic HBV infection were also analyzed separately for 71 of 72 patients. The number of patients in phases of immune-tolerant, HBeAg-positive immune-active, and inactive carrier were 38 (54%), 10 (14%), and 13 (18%), respectively. The remaining 10 patients (14%) did not fit into the three main phases. One patient had missing HBV DNA values.

HBeAg and anti-HBe statuses, other laboratory values specifically ALT, APRI, and AFP were also analyzed in 353 patients. A total of 208 patients (59%) were HBeAg-positive and 145 patients (41%) were HBeAg-negative. The combination of HBeAg-positive with anti-HBe-negative was noted in 203 patients (58%) and HBeAg-negative with anti-HBe-positive were noted in 136 patients (38%). The mean ± SD for the age of the patient was 18 ± 6 years old in the HBeAg-positive group and was 21 ± 5 in the HBeAg-negative group (P < 0.05). The mean ± SD of platelet count was 241 ± 62 × 10³/μL in the HBeAg-positive group and was 230 ± 52 × 10³/μL in the HBeAg-negative group (P > 0.05). No patients had a loss of HBsAg or developed antibody to HBsAg. Of the total 353 patients, the median (IQR) of APRI score was 0.23 (0.11). AFP < 10 ng/mL was documented in 317 patients with mean ± SD 2.1 ± 1.1 ng/mL, AFP ranging between 10 and 15 ng/mL in four patients and missing AFP data in 31 patients. Two patients >21 years old had AFP 54 and 167 ng/mL, respectively.

Longitudinal Data for a Subgroup of Patients

One hundred and seventy-nine patients were followed for ≥5 years and their HBeAg seroconversion and anti-viral therapy statuses are shown in Figure 1. The mean duration of follow up was 8 years (mean ± SD: 8 ± 2), with 116 patients followed for 5–9 years and 63 patients followed for 10–12 years. Twenty-seven patients (15%) experienced the process of HBeAg loss during the study period. Twenty-two of the 27 patients completed seroconversion with negative HBeAg and positive anti-HBe, and the remaining five patients had partial seroconversion with anti-HBe remained negative and low HBV DNA ranging from undetectable to maximal 1050 IU/mL. Fifteen of the 27 patients had spontaneous seroconversion from positive HBeAg to negative HBeAg, and 13 of these individuals also had seroconversion from negative anti-HBe to positive anti-HBe. Fifty-five patients remained HBeAg-negative/anti-HBe-positive throughout the study period. The total number of patients with the completion of HBeAg spontaneous seroconversion was 68 (38%) including 55 patients who remained completed seroconversion throughout the study period and 13 patients who have completed HBeAg seroconversion during the study period. The anti-viral therapy prescribed to 48 patients in this subgroup included tenofovir, entecavir, adefovir or peg-interferon. The majority of patients on anti-viral therapy were >21 years of age at the time of anti-viral therapy. Of the 179 patients, there was no statistically significant difference between baseline and the latest ALT level during their follow-up period (P > 0.05). None of the 179 patients had AFP >10 ng/mL or platelet count <150 × 10³/μL.

FIGURE 1: Longitudinal data on a subgroup of patients followed for ≥5 years. HBeAg = hepatitis B envelop antigen.

Maternal–Child Data

A total of 72 patients (20%) with chronic HBV infection were born in the United States. Of this group of patients, eight patients were linked with their mothers’ charts at CBWCHC. The remaining 64 patients did not have their identifiable maternal charts at CBWCHC. The details of eight maternal–child data are listed in Table 3. All eight mothers had a history of chronic HBV infection and had laboratory testing done within 9 months before the patients’ birth. All patients had completed the HBV vaccine series. All received HBIG except one (case #7) whose mother declined HBIG at birth. All mothers were HBeAg-positive and had maternal HBV DNA >1 million IU/mL within 9 months before delivery, ranging from 29.6 million to >2 billion IU/mL. None of the mothers were on anti-viral therapy during their pregnancy. Anti-viral therapy was discontinued before pregnancy in two mothers and was started after delivery in six mothers.

TABLE 3 - Vertical transmission: maternal–patient paired data

Case no.	Age (y)	Sex	Birth delivery	HBV vaccine series/HBIG	Maternal HBeAg	Maternal HBV DNA (IU/mL)	Anti-viral therapy during pregnancy
1	3	Female	C/S	Yes/yes	+	>170,000,000	No
2	5	Female	C/S	Yes/yes	+	2,527,307,600	No
3	7	Male	C/S	Yes/yes	+	473,122,839	No
4	7	Female	NSVD	Yes/yes	+	85,100,000	No
5	8	Male	NSVD	Yes/yes	+	125,000,000	No
6	8	Male	NSVD	Yes/yes	+	190,000,000	No
7	8	Female	NSVD	Yes/no	+	301,000,000	No
8	9	Female	C/S	Yes/yes	+	29,600,000	No

C/S = cesarean section; HBeAg = hepatitis B envelop antigen; HBIG = hepatitis B immunoglobulin; NSVD = normal spontaneous vaginal delivery.

DISCUSSION

The study describes Chinese American pediatric patients with chronic HBV infection in a single health center. About half of the population continued care at CBWCHC for at least 5 years and many patients had been followed into early adulthood by the end of the study period. Although the majority were foreign-born, 72 patients (20%) were US-born. We surmise vertical transmission may have been a major mode of infection in the study population. Prior studies from China, Taiwan, and the United States have all shown that vertical transmission is the major source of chronic HBV infection in children ^{(10,14,17,18)}. All patients in our study had no advanced liver disease related to HBV based on their laboratory values such as APRI score, platelet counts and AFP, however, the prediction of chronic HBV disease progression in the present study is limited due to lack of imaging and histology data. It would be valuable to prospectively study this cohort of patients to understand HBV disease progression.

Our data revealed that 34% of patients were in the immune-tolerant phase and 25% were in the inactive carrier phase. The projected proportion of patients in the inactive carrier phase is likely to be much higher since an additional 26 of 88 patients in the “other” category had characteristics of the inactive carrier phase with the exception of mild elevation of ALT. We suspect that the mild elevation of ALT may reflect non-HBV related causes of abnormal ALT such as non-alcoholic fatty liver disease, which is highly prevalent in children ⁽¹⁹⁾. At CBWCHC, there is a 21% prevalence of overweight and obese children during the study period ⁽²⁰⁾. In addition, many patients in the “other” group seemed to be in the process of seroconversion or had incomplete seroconversion. The data reflect the natural history of chronic HBV infection in children in endemic areas. The majority of infected children presented with the immune-tolerant phase, followed by the immune-active phase with HBeAg seroconversion, and finally entered the inactive carrier phase. In contrast to our results, a pediatric cohort study through the Hepatitis B Research Network from the United States and Canada comprising younger age patients than the present study reported that a minority of children were in the immune-tolerant phase based on age-specific cutoff ALT values ⁽⁹⁾.

Our finding of the significant difference in age for HBeAg-positive and -negative groups supports the frequency of HBeAg positivity decreases with age. The overall HBeAg spontaneous seroconversion rate is 38% in the longitudinal subgroup. This spontaneous seroconversion rate seems higher than a previously published rate of 15% at ages 5–19 and 23% at ages 20–34 in HBeAg-positive Asian Americans ⁽²¹⁾. The reason is unclear since there were no HBV genotype data to determine its influence on HBeAg seroconversion. Only two patients who were >21 years old experienced reactivation after HBeAg seroconversion supporting a previous report that reactivation is rare in the pediatric population ⁽²²⁾.

A systematic review and meta-analysis showed that anti-viral therapy improved HBV DNA suppression and HBeAg seroconversion; however, no substantial benefit from anti-viral therapy was revealed in children in the immune-tolerant phase ⁽²³⁾. The present study did not produce enough evidence to evaluate the impact of anti-viral therapy on HBeAg seroconversion. In most cases, the indication and duration of antiviral therapy were not well-documented.

Under CDC recommendations, children born to HBeAg-positive mothers should receive passive and active prophylaxis at birth to prevent HBV vertical transmission. A previous study however showed that even with universal prophylaxis at birth, the lowest HBV DNA level associated with the vertical transmission was 6.32 × 10⁷ IU/mL ⁽¹⁴⁾. Our data showed that 72 patients with chronic HBV infection were born in the United States during the time of implemented universal immune-prophylaxis guidelines. All mothers of eight patients had very high HBV DNA levels during pregnancy suggesting universal immune-prophylaxis at birth is not sufficient to prevent vertical transmission in the setting of high maternal HBV DNA levels. The updated AASLD practice guidelines strongly recommend anti-viral therapy for HBeAg-positive pregnant women with a serum HBV DNA >200,000 IU/mL to prevent HBV perinatal transmission ⁽¹⁵⁾. This is further attested that since CBWCHC's Hep B Moms Program fully implemented with the AASLD practice guideline, there has been no subsequent documented HBV vertical transmission among mothers with chronic HBV infection ⁽²⁴⁾.

As with any retrospective study, there were missing data due to variations in clinical practice and documentation. There was a lack of family history recorded in the majority of the study patients. In addition, using ALT as one of the markers for the classification of HBV phases is not without problems since a lower ALT cutoff value may impact the classification of chronic HBV phase with more patients in the immune active phase. In a survey study, a normal ALT of 26 U/L in boys and 22 U/L in girls were reported as more accurate and the old cutoff value of ALT was set too high to screen liver disease in children ⁽²⁵⁾. In the present study, we used the new cutoff ALT 35 U/L for male and 25 U/L for female for chronic HBV infection phase classification as per AASLD updated HBV guideline; however, the ideal cutoff value of ALT to define HBV phases in children remains debatable. This suggests the need to rethink the current concept of the natural course of pediatric chronic HBV infection.

In conclusion, the present study described the natural history of a large number of pediatric and young adult patients with chronic HBV infection at a single health center. Both immune-tolerant and inactive carrier phases were common for chronic HBV infection, with a spontaneous seroconversion rate of 38% in our population. HBV infection in Chinese American children remains challenging as passive and active immuno-prophylaxis at birth is not sufficient to prevent vertical transmission, especially when born to a mother with high maternal HBV viral load. We need to raise awareness of updated AASLD practice guidelines on anti-viral therapy for mothers with chronic HBV infection during pregnancy. The ultimate goal is to decrease the HBV vertical transmission rate to zero and to eradicate HBV infection in children.