Article first published online: 19 JUL 2011
Copyright © 2011 American Association for the Study of Liver Diseases
Volume 54, Issue 3, pages 801–807, 2 September 2011
How to Cite
McMahon, B. J., Bulkow, L. R., Singleton, R. J., Williams, J., Snowball, M., Homan, C. and Parkinson, A. J. (2011), Elimination of hepatocellular carcinoma and acute hepatitis B in children 25 years after a hepatitis B newborn and catch-up immunization program. Hepatology, 54: 801–807. doi: 10.1002/hep.24442
Potential conflict of interest: Dr. Singleton received grants from Pfizer.
Supported by the United States Public Health Service/Indian Health Service, Alaska Native Tribal Health Consortium and the Centers for Disease Control and Prevention.
- Issue published online: 25 AUG 2011
- Article first published online: 19 JUL 2011
- Accepted manuscript online: 26 MAY 2011 09:35AM EST
- Manuscript Accepted: 8 MAY 2011
- Manuscript Received: 11 APR 2011
Alaska Native people experience the highest rates of acute and chronic hepatitis B virus (HBV) infection and hepatocellular carcinoma (HCC) in the United States. We examined the effect of a universal newborn immunization with hepatitis B vaccine and mass population screening immunization program initiated in 1984 on rates of HBV and HCC in children 25 years later. During this time, the population of Alaska Native people grew from an estimated 75,000 to 130,000 persons. A surveillance system to detect acute HBV infection in Alaska Native facilities was established in 1981. Cases of HCC in children under 20 years of age were identified using a National Cancer Institute (NCI)-funded Cancer Registry established in 1969 coupled with an active surveillance program of screening persons with chronic HBV semiannually for alpha-fetoprotein since 1982. The incidence of acute symptomatic HBV infection in persons <20 years of age fell from cases 19/100,000 in 1981-1982 to 0/100,000 in 1993-1994. No cases of acute HBV have occurred in children since 1992. The incidence of HCC in persons <20 years decreased from 3/100,000 in 1984-1988 to zero in 1995-1999 and no cases have occurred since 1999. The number of identified hepatitis B surface antigen-positive children <20 years in the Alaska Native population declined from 657 in 1987 to two in 2008. Conclusion: Universal newborn vaccination coupled with mass screening and immunization of susceptible Alaska Natives has eliminated HCC and acute symptomatic HBV infection among Alaska Native children and this approach is the best way to prevent HBV-related disease in children. (HEPATOLOGY 2011;)
One-third of the world's six billion people have been infected with hepatitis B virus1 (HBV) and over 350 million are chronically infected.2 The risk of developing chronic HBV infection is 90% for those who acquire HBV from infected mothers at birth and between 30% and 50% for those infected after birth but before their fifth year of life.3, 4 Thus, most of the persons in the world with chronic HBV infection are infected in early childhood. Up to 25% of those chronically infected with HBV will develop hepatocellular carcinoma (HCC) and/or cirrhosis of the liver, usually after age 40 years. In 1982 plasma-derived hepatitis B vaccine became available and 6 years later recombinant hepatitis B vaccine was introduced. The World Health Organization determined that the best strategy to eliminate HBV infection was universal hepatitis B vaccination for all infants and the Global Alliance for Vaccine Initiative is funding universal hepatitis B vaccine in most countries in the world.5
In the 1970s, Alaska Native (AN) people were found to have a high rate of HBV infection, as evidenced by an overall prevalence of hepatitis B surface antigen (HBsAg) >6% and as high as 20% in some villages in southwest Alaska.6 The incidence of HCC related to HBV infection was also high, with one-third of HCC cases occurring in persons <30 years of age.7 This burden of disease led the Indian Health Service (IHS), AN health corporations, and the Arctic Investigations Program (AIP) of the Centers for Disease Control and Prevention (CDC), beginning in 1984, to implement the most comprehensive HBV control program initiated.8 The control activities, coordinated through the Liver Disease and Hepatitis Program (LDHP) at Alaska Native Medical Center (ANMC), included both universal infant immunization and population-wide (all ages) screening for HBsAg with immunization of seronegative persons, funded by a special United States congressional appropriation through IHS. Although other programs included universal infant immunization and immunization of selected childhood groups or healthcare workers, no other program has focused on population-wide coverage with screening for hepatitis B seromarkers and hepatitis B vaccination of seronegative persons.9-11
In this article we describe the impact of universal childhood vaccination coupled with mass screening and catch-up immunization on control of HBV infection and HCC in children 25 years after program inception. We examined (1) the annual incidence of acute symptomatic HBV infections in children <20 years of age from before the vaccination program (1981) through 2010; (2) the number of HBsAg-positive children in the population who were positive for hepatitis B “e” antigen (HBeAg) before the program and in 2008; (3) the prevalence of HBsAg among children <20 years of age in the AN population at the time of the screening and vaccination program (1984-1987) and again in 2008; and (4) the incidence of HCC in children <20 years of age from 1969-2008.
The timeline of components of the AN HBV control program, including the screening of pregnant women, universal newborn immunization, population-based screening, and vaccination of children and adults and screening of HBsAg carriers are listed in Table 1 and have been reported.8, 12 In 1980, universal screening of all pregnant women for HBsAg was introduced at two hospitals, ANMC and Yukon Kuskokwim (YK) Delta Regional Hospital. Three doses of hepatitis B immune globulin (HBIG) were given to infants of HBsAg-positive mothers. Hepatitis B vaccine, starting at birth, was added to this HBIG regimen after it became available in 1982, and these components were extended to all AN regional hospitals statewide in 1984.8 HBV seromarkers were tested between 12 and 15 months of age in infants of HBsAg-positive mothers to monitor the effectiveness of newborn vaccination.
|• 1978: Establishment of a registry of HBsAg-positive persons|
|• 1980: Screening of all pregnant women for HBsAg initiated in the two largest Alaska Native Hospitals, in Anchorage and Bethel. In these two hospitals:|
|- 1980-1981: HBIG administered to infants of HBsAg-positive mothers at birth, 3 and 6 months|
|- 1982: Hepatitis B vaccine added to HBIG schedule starting at 3, 4, and 9 months as per FDA regulations at that time|
|• 1981-82: Hepatitis B vaccine demonstration project in Southwest Alaska|
|• Alaska Native Hepatitis B Mass Screening and Vaccination Program:|
|- 1984-87 vaccination campaign: 52,000 Alaska Natives screened (75% of the population at that time) and 40,000 susceptible persons vaccinated|
|- 1984 to present: Hepatitis B vaccination of all infants starting at birth|
|• 1982 to present: All HBsAg-positive persons who remained positive for 6 months or more are screened semiannually alpha-fetoprotein (AFP) and HBV seromarkers. Persons with AFP >10mg/mL are sent for liver ultrasound|
|- Persons with AFP initially above 25 ng/mL and in 1990 above 15 ng/mL are sent for liver ultrasound examination, liver function tests and, if indicated, liver biopsy to detect hepatocellular carcinoma at an early stage|
|- 2001 to present: ALT and AST testing was included and in persons with elevated ALT or AST, HBV DNA was performed and those persons with levels above 2,000 IU/mL were evaluated as possible candidates for antiviral therapy|
In 1981-1982 a HBV vaccine demonstration project providing HBV screening and vaccination was conducted in 16 villages in the highly endemic YK Delta and lower Norton Sound regions of Alaska.7 In 1984 the mass screening and immunization program was extended to AN in all regions statewide. During 1984-1987 all AN persons were offered screening for HBV seromarkers. In all, 52,022 persons of the estimated 75,000 persons in the AN population were screened and the hepatitis B vaccination series was offered to those negative for HBV markers.12 In western Alaska, the area endemic for HBV, almost 90% of the population was covered by this program (screened or vaccinated at birth).
Beginning in 1984, AN infants in all AN facilities were offered three doses of hepatitis B vaccine starting at birth. Simultaneously, a vaccine catch-up program targeted children who had not received infant vaccination through screening and immunization campaigns were conducted in preschools, grade schools, and high schools statewide. Seronegative AN adults were also encouraged to be vaccinated and reminders were entered into problem lists in electronic medical records. During the vaccine campaign in 1984-1987, 43,618 and 34,890 AN persons received at least one or two doses, respectively, of hepatitis B vaccine. These measures continued in all AN facilities after the vaccine campaign ended in 1987. During the mass screening and vaccination program during 1983-1987, 1,603 HBsAg-positive AN persons were identified statewide, of whom 1,530 were chronically infected (defined as HBsAg-positive on at least two occasions at least 6 months apart). Of these, 1,428 were tested for HBeAg and antibody to HBeAg (anti-HBe), and 506 (34.5%) were HBeAg-positive. Five HBV genotypes were found in this population: A2, B6, C2, D, and F1.13 Altogether, 20,666 AN children <20 years of age were screened for HBV seromarkers during this period and 657 (3.2%) were found to be HBsAg-positive (Fig. 2). The prevalence of HBsAg in children <20 years of age ranged from 0.1% in southeast to 9.1% in southwest Alaska.12
In November 1991 the Advisory Committee on Immunization Practices recommended universal immunization of infants against HBV, and in 1993 the State of Alaska initiated a universal infant hepatitis B immunization program for infants of all races.14 The state expanded availability of vaccine to all Alaskan children through 18 years in 1997.15 Lastly, in 2000 hepatitis B immunization was required for school and daycare attendance in Alaska.16
Materials and Methods
In 1985 there were approximately 73,798 AN persons (Indian, Aleut, and Eskimo) living in Alaska who received their healthcare from a coordinated healthcare program administered initially by the IHS and subsequently by AN health corporations. The system consists of Community Health Aide/Practitioners in rural village clinics, regional hospitals and clinics, and the tertiary hospital, ANMC, in Anchorage. In 2008 an estimated 130,000 AN people lived in Alaska, approximately 60% of whom lived in rural villages. The AN health service regions include Anchorage, Southeast, Interior, North Slope, and the endemic western regions of YK Delta, Bristol Bay, Kotzebue Sound area (Maniilaq), and Norton Sound.
Estimated vaccine coverage with three doses of hepatitis B vaccine among children 19-35 months of age was reported by the National Immunization Survey for the United States and individual states by race during 1996 through 2008. Data tables for the National Immunization Survey were downloaded (http://www.cdc.gov/vaccines/stats-surv/imz-coverage.htm#nis).18
In Alaska, acute HBV has been a reportable disease to the State of Alaska, Division of Epidemiology since the onset of this program. This surveillance system for monitoring acute HBV was enhanced by the establishment of three central laboratories in the state to perform HBV serology (ANMC from 1987 to present, AIP from 1980 to 1988, and the State of Alaska Laboratories from 1984 until present). Providers were encouraged to send samples for HBV serology to one of these laboratories for testing of suspected HBV infection cases at no cost. A case of acute HBV infection was defined as an illness compatible with hepatitis and an alanine aminotransferase (ALT) level more than 10 times the upper limit of normal, plus a positive IgM antibody to hepatitis B core antigen (anti-HBc). Each anti-HBc IgM-positive laboratory-identified case and each clinical case reported by providers or public health nurses was evaluated by a coordinator from the LDHP at ANMC for adherence to the acute HBV infection diagnostic criteria.
Cases of HCC were identified by two methods. First, a National Cancer Institute (NCI)-funded Cancer Registry was established in 1969 to record all cancer cases in AN people. Second, in late 1982 the LDHP established a HCC surveillance program that tested all HBsAg-positive AN persons for alpha-fetoprotein (AFP) every 6 months to detect HCC early at a surgically resectable stage. HBsAg-positive persons with an AFP result >15 ng/ml underwent a liver ultrasound (US) or computerized axial tomography (CT) scan to identify lesions suspicious for HCC, which were then either biopsied or resected.17 In addition to AFP testing, persons chronically infected with HBV were also followed prospectively for HBV seromarkers including hepatitis HBeAg, anti-HBe. Since 2001, ALT and aspartate aminotransferase (AST) levels were also performed. HBV DNA levels were performed if either ALT or AST liver transaminase levels were elevated.
The analysis of the impact of this program was approved by tribal health review boards for the Alaska Native Tribal Health Consortium and Southcentral Foundation.
Historical population estimates were derived from U.S. census numbers and, for 2010 estimates, the State of Alaska Bridged Race estimates were used. For comparison of prevalence rates, mid-P exact P-value is reported. HCC rates were compared over time with a linear test for trend. All P-values are two-sided.
Continuous vaccine availability and recommendations for routine infant hepatitis B vaccination resulted in high levels of vaccine coverage in AN children 19-35 months of age as measured by the National Immunization Survey starting in 1996. In 1996, estimated vaccine coverage with three or more doses of hepatitis B vaccine was significantly higher among American Indian or Alaska Native (AI/AN) children residing in Alaska (93.5%; 95% confidence interval [CI]: 87.2-100) than among the general U.S. population (81.8%; 95% CI: 80.9-82.7). From 1996 to 2008, vaccine coverage among AI/AN children in Alaska has remained at a high level, between the lowest vaccine coverage rate of 86.6% (95% CI: 78.2-95) in 1998 and the highest vaccine coverage rate of 99.2% (95% CI: 97.6-100) in 1997), with the point estimate generally significantly higher than the general U.S. population of 81.8% (95% CI: 80.9-82.7) in 1996 to 93.5% (95% CI: 92.8-94.2) in 2008.18
Acute HBV in Children.
In Alaska statewide,12 the annual incidence of acute symptomatic hepatitis B in Alaska Native children decreased from 19 cases/100,000 population in 1981-1982 to 0 cases/100,000 in 1993-1994 (Fig. 1). No cases of acute symptomatic HBV in children have occurred since 1992.
Chronic HBV in Children.
In 1988, 465 (34.3%) of the 1,356 HBsAg-positive persons with chronic HBV infection were children <20 years of age. Figure 2 shows a significant decline in the prevalence of HBsAg-positive AN children in the population between 1988, at the end of the mass HBV serologic screening program, and 2008 (P = 0.001). At the end of 2008, 1,013 ANs were known to be HBsAg-positive; only two of those were under 20 years of age, whereas the remaining 1,011 were 20 years of age or older. Both HBsAg-positive children had been infected after the implementation of universal newborn vaccination and had received HBIG and a complete hepatitis B vaccine course starting at birth. Both were children of mothers who were both HBsAg and HBeAg-positive and infected with HBV genotype C at the time of birth. In 1988, 380 (20.5%) of the 1,356 HBsAg-positive AN persons were also HBeAg-positive. Of the HBsAg-positive AN children, a higher proportion, 181 (38.9%) of 465, were also positive for HBeAg compared to AN adults, 91 of 794 (12.3%). Only 20 (2.0%) of the 1,013 HBsAg-positive AN persons were still HBeAg-positive in 2008, 19 adults and one child under 20.
HCC in Children.
Between 1969 and 2008, 17 cases of HCC were identified in children <20 years of age. The annual incidence of HCC per 100,000 children less than 20 years calculated in 5-year intervals is displayed in Fig. 3. The incidence peaked at almost 3/100,000 in the 1984-1988 period. The incidence thereafter fell to 0 and has remained so since 1999 (test for trend overall, P < 0.001; test for trend using 1969-1988 as baseline, P = 0.003; 1969-1988 versus 1989-2008, P = 0.013). Fourteen children with HCC were HBsAg-positive and three were HBsAg-negative. At the time of diagnosis, the children's ages ranged from between 8 to 19 years. Two of the three HBsAg-negative children were also negative for antibody to hepatitis B core antigen (anti-HBc); both of these children lived in low-endemic areas of Alaska where the prevalence of HBsAg was <1%. The other child was not tested for anti-HBc; however, the child lived in an area endemic for HBV and could have had HBV.
This study is the first in the world to examine the impact of the introduction of universal newborn hepatitis B vaccine coupled with a simultaneous catch-up vaccination program on HBV infection in children. The AN newborn immunization program was the earliest universal HBV immunization program, occurring 2 years before universal newborn vaccination in Taiwan.19 The impact on HBV transmission in children was dramatic, as can be evidenced by the disappearance in acute symptomatic HBV disease and HCC within a little over one decade of starting this program. Alaska's experience serves as a model for control of HBV in other endemic populations.
Our study shows a dramatic fall in the estimated prevalence of HBsAg in children to nearly zero. In a cluster of seven southwest Alaska villages where prevalence in children younger than 10 years of age was >10% before the campaign, there were no HBsAg- positive children in this age group 10 years after the implementation of this program.20 A dramatic reduction in the prevalence of HBsAg carriage in children has also been seen in other populations where universal infant vaccination has been implemented.19-22 However, in none of these studies has the prevalence decreased below 1%, possibly due to the inconsistent use of HBIG in infants born of HBsAg-positive mothers and the lack of a catch-up vaccination program in children. Only two AN children <20 years are known to be HBsAg-positive in 2008. Because the estimated population of AN children under 20 years of age in 2008 was 48,995 and over 90% of AN children since 1987 have received hepatitis B vaccination, then the prevalence of HBsAg in children must be far less than 1%.
Five separate HBV genotypes have been found in AN people and these HBV genotypes were likely introduced into the population at different times.13 The profound decrease in HBV transmission occurred both in southwest Alaska where horizontal, child-to-child transmission predominated due to the presence of HBV genotypes A2, B6, D2, D3, and F1, as well as in northwest Alaska, where most persons were infected with HBV genotype C and vertical transmission of HBV was frequent.13 Prior to the immunization campaign both symptomatic and asymptomatic infection was found to be spreading rapidly. In southwest Alaska, where vertical transmission of HBV was uncommon due to the absence of HBV genotype C,23, 24 in a large prospective cohort study conducted in the 1970s, we found that children under 5 years of age had the highest infection rate and, when infected, approximately 30% developed chronic HBV compared with 5% to 10% of those older than 20.4 This high transmission and subsequent chronic infection rate was likely facilitated by the high proportion of children who were HBeAg-positive with high viral titers. In the mid-1970s, before HBV DNA detection methods were available, HBsAg was found in abundance on environmental surfaces, including in homes, on school lunch room tables, and filtered out of impetigo sores in children.25 By initiating newborn vaccination and mass vaccination of HBV-seronegative children as well as adults, transmission was rapidly eliminated in susceptible children and, in addition, over time with clearance of HBeAg, chronically infected children became less infectious. Thus, this program virtually halted transmission both perinatal and horizontal of all five HBV genotypes in children in every region of Alaska. The comprehensive nature of the Alaska Program, which included screening pregnant women for HBsAg, routine immunization of all newborns regardless of mothers' status, and vaccination catch-up programs for persons of all ages, is unmatched by other similar hepatitis control programs anywhere in the world. Other programs such as Taiwan, Thailand, and the U.S. have included universal infant vaccination with or without childhood vaccination of specific age groups, such as adolescents, with the goal of eventually, over one or two decades, having a cohort of children protected from HBV.
Coupling a catch-up program with universal immunization likely decreases the risk of perinatal transmission of HBV because teenagers and young women of childbearing age who are negative for HBV seromarkers can be vaccinated and obtain immediate protection against HBV infection. In contrast, in countries where only newborn immunization is used it will take several decades to protect all serosusceptible women from acquiring HBV infection. An ongoing program to screen all pregnant women has the added benefit of identifying all women of childbearing age who are HBsAg-positive.
The incidence of acute symptomatic and fulminant hepatitis B has also declined in countries with universal infant immunization.10, 26, 27 However, we demonstrated not only a more rapid reduction in acute hepatitis B, but the elimination of symptomatic infection in children within just a decade of the AN program initiation, resulting in an entire generation free of chronic HBV. Because 30% of unprotected children <5 years of age infected after birth will become chronically infected if exposed to HBV,4 including a catch-up program will also protect seronegative children from becoming infected and developing chronic HBV infection. The widespread impact of our program was related to the rapid reduction of transmission in the entire population accomplished by mass screening and vaccination of all age groups. Persons who are HBeAg-positive have high HBV DNA viral loads and have a high risk of transmitting HBV to their close contacts. The natural history of persons infected with HBV demonstrates that most persons who are HBeAg-positive will eventually seroconvert from HBeAg to anti-HBe; when this occurs, viral levels of HBV are reduced and they are less likely to infect others.28 Rapid vaccination of the susceptible population in this population, coupled with natural HBeAg to anti-HBe seroconversion over time, has significantly reduced the proportion of persons with high HBV viral levels.13
Finally, our study is the first to show that universal newborn immunization coupled with a child catch-up program can virtually eliminate HBV-associated HCC in children. A reduction in the incidence of HCC was first shown in Taiwan 10 years after introduction of newborn immunization and later in Thailand.29, 30 The incidence of HCC in AN children was more than 3 times higher (3 per 100,000) than in either Taiwan (0.7 per 100,000) or Thailand (0.88 per 100,000). The programs in those countries relied on national cancer registries only, whereas we also included active surveillance of HBsAg-positive children and adults who are tested for AFP every 6 months.
Our study has a few limitations. First, despite a system of both laboratory and clinical surveillance for acute hepatitis B, we cannot be certain that we have captured 100% of all symptomatic occurrences of acute HBV, as is the case in all surveillance systems. Although an extensive population-based serosurvey was conducted from 1984-1987, no recent serosurvey has been conducted, so we cannot ascertain with complete accuracy the prevalence rates of HBsAg in children. However, because 90% of all susceptibles in the most endemic regions were rapidly vaccinated in the 1980s and aggressive vaccination of newborns has continued, as well as continued catch-up screening and vaccination using electronic medical records the past 2 decades, it is highly likely that far less than 1% of children have chronic HBV infection.
In conclusion, our study is the first to show that universal newborn and childhood hepatitis B immunization with high coverage rates vaccine combined with a population-wide catch-up program can halt transmission and maintain control of HBV in a highly endemic population. The Alaska experience can serve a model for control of HBV in other populations. Universal hepatitis B vaccination in infants and catch-up screening and vaccination in children is the most rapid way to achieve the goals of elimination of HCC and chronic HBV infection in children. Because the incidence of HCC does not rise substantially until after age 40 years in adults,31 it may take two more decades to be able to demonstrate a reduction in rates of HCC among adults.
- 8A comprehensive programme to reduce the incidence of hepatitis B virus infection and its sequelae in Alaskan Natives. Lancet 1987; 330: 1134-1136., , , , , , et al.
- 9A comprehensive immunization strategy to eliminate transmission of hepatitis B virus infection in the United States: recommendations of the Advisory Committee on Immunization Practices (ACIP) part 1: immunization of infants, children, and adolescents. MMWR Recomm Rep 2005; 54: 1-31., , , , , , et al.
- 11CDC. Implementation of newborn hepatitis B vaccination—Worldwide, 2006. MMWR 2008; 57: 1249-1252.
- 14State of Alaska. Universal Infant Hepatitis B Immunization: A New Program is Launched. In: Division of Public Health Section of Epidemiology BN, ed. February 1, 1993.
- 15Alaska So. Childhood Vaccination Recommendations Expanded. In: Division of Public Health EBN, ed. December 4, 1997.
- 16State of Alaska. New School and Childcare Recommendations for Fall, 2001. In: Department of Health and Social Services, Division of Public Health, Section of Epidemiology, Bulletin No. 21. December 26, 2000.
- 18National Immunization Survey, Children 19-35 months, Data Tables. Accessed April 24, 2010, at http://www.cdc.gov/vaccines/stats-surv/imz-coverage.htm#nis
- 23Increased prevalence of hepatitis B surface antigen in pregnant Alaska Eskimos. Circumpolar Health 1985; 84: 206-208., , , , , .