Long-term protection following a primary series of hepatitis B vaccine in Alaska Native children and adults

Abstract

Background:

Vaccination is effective at preventing chronic HBV infection and its complications, but the duration of protection and need for booster doses have not been determined.

Methods:

We administered a 3-dose primary series of plasma-derived hepatitis B vaccine to Alaska Native persons aged 6 months or above who resided in an HBV-endemic area of western Alaska. Persons with antibody to hepatitis B surface antigen (anti-HBs) ≥10 milli-international units per milliliter (mIU/mL) at 6 months after series completion were followed through 41 years after vaccination. Beginning at 22 years, participants with anti-HBs <10 mIU/mL were offered a booster dose (ie, challenge dose) of recombinant hepatitis B vaccine and retested 4 weeks later as a surrogate for immune memory. Combining these immune-memory data with a survival model estimating anti-HBs persistence, we estimated the overall proportion with serologic evidence of protection at 22, 30, 35, and 41 years. We also described breakthrough hepatitis B infections.

Results:

During 1981–1982, 1351 persons completed the hepatitis B vaccine primary series and had anti-HBs ≥10 mIU/mL at 6 months after vaccination; median age at enrollment was 13 years (range: 0.5–77). The overall proportion with serologic evidence of protection was 95% at 22 years, 95% at 30 years, 88% at 35 years, and 91% at 41 years after vaccination. At 41 years, 32% had circulating anti-HBs ≥10 mIU/mL, and 59% had immune memory. During 41 years of follow-up, 16 (1%) persons had evidence of a breakthrough infection; none developed chronic hepatitis B.

Conclusions:

We found serologic evidence of protection among Alaska Native persons 41 years after primary hepatitis B vaccination. Our findings support the recommendation that booster doses are not needed in endemic areas.

INTRODUCTION

HBV is transmitted by percutaneous or mucosal contact with infected body fluids. ¹ Chronic hepatitis B infection is an important cause of cirrhosis and HCC, and 15%–25% of chronically infected persons will die prematurely from these complications.^1,2 The risk of developing chronic hepatitis B is highest among those infected as infants and children. ¹ Safe and effective antiviral medications can reduce, but not eliminate, chronic hepatitis B infection. Vaccination is effective at preventing both acute and chronic infection and its complications, but the duration of protection and need for booster doses have not been determined.^1–3

Prior to the introduction of hepatitis B vaccines, HBV was highly endemic in Alaska Native villages in western Alaska. ⁴ A serosurvey during 1973–1975 found the seroprevalence of HBsAg was 6%. ⁴ During this time, Alaska Native populations in regions with high prevalence of hepatitis B infection also experienced high rates of chronic hepatitis B complications, including cirrhosis and HCC. ⁵ Universal hepatitis B screening, widespread hepatitis B vaccine use, and introduction of a birth dose of hepatitis B vaccine led to a dramatic reduction in HBV transmission in these populations, with incidence of symptomatic acute HBV infection declining by 93% within 5 years of introduction and elimination of HBV infection and HCC in children within 20 years.^6–8

Following licensure of plasma-derived hepatitis B vaccine in 1981, a phase 4 clinical trial was conducted in 17 remote Alaska Native communities in western Alaska. ⁵ Over the last 4 decades, the cohort of participants in this study has been assessed periodically for serologic evidence of protection from HBV infection.^5,9–15 We summarize the findings of this long-term immunogenicity study of primary hepatitis B vaccination in Alaska Native persons, including the final follow-up of the cohort at 41 years after vaccination.

METHODS

Study design and population

This was a postlicensure, prospective, longitudinal study whose methods have been previously described.^5,9–15 Briefly, during 1981–1982, a 3-dose series of plasma-derived hepatitis B vaccine (Heptavax-B, Merck) was administered to Alaska Native persons aged 6 months or above living in 17 villages in western Alaska. Persons with immunosuppressive disease or therapy were excluded from participation. Vaccine doses were administered at 0, 1, and 6 months. For children aged below 10 years, each dose contained 10 µg HBsAg; children and adults aged 10 years or above received a vaccine dose with 20 µg HBsAg.

Prior to the first dose of vaccine, study participants were tested for HBsAg, antibody to HBsAg (anti-HBs), and total antibody to hepatitis B core antigen (anti-HBc); those who tested positive for any marker were excluded. Participants were retested for these 3 markers at the time of the first and third vaccine doses and were eligible for inclusion in the follow-up study if they received 3 doses of vaccine and remained negative for HBsAg and anti-HBc throughout the series. The cohort was followed annually for the first 11 years after vaccination, and then at 15, 22, 30, 35, and 41 years.^5,9–15 Beginning at 22 years, participants with anti-HBs <10 milli-international units per milliliter (mIU/mL) were offered a booster dose (sometimes referred to as a “challenge dose” when used to evaluate anamnestic response) of recombinant hepatitis B vaccine and retested for anti-HBs at 4 weeks after vaccination.^1,13–15 Anti-HBs ≥10 mIU/mL after receipt of a booster dose of recombinant hepatitis B vaccine was considered evidence of immune memory.

Data and specimen collection at 41 years after vaccination

At 41 years after the primary series, we recruited eligible participants who resided in 11 of the original 17 villages in western Alaska and those who had moved to the city of Anchorage or Bethel, the largest town in the region. Study personnel reviewed the state immunization registry and participant medical records for receipt of additional doses of hepatitis B vaccine, and these persons were excluded from participation. Study personnel worked with Tribal health staff to actively recruit potential participants, including conducting home visits for those who were unable or unwilling to visit the study site.

For enrolled participants, study personnel collected a blood specimen for anti-HBs and anti-HBc testing. Participants also completed a brief questionnaire regarding demographic information and receipt of additional doses of hepatitis B vaccine. Persons who tested positive for anti-HBc were asked to provide a second specimen for confirmatory anti-HBc, HBsAg, and HBV DNA testing, and study personnel reviewed their medical records for clinical evidence of acute HBV infection. Participants with anti-HBs <10 mIU/mL were offered a booster dose of recombinant hepatitis B vaccine with 20 µg HBsAg (Engerix-B, GlaxoSmithKline) and had another blood specimen collected at 4 weeks after vaccination for anti-HBs testing.

Laboratory testing

For the first 6 years of follow-up, anti-HBs levels were tested using the sample ratio unit method.^5,9,10 Later, these specimens were retested and quantified in mIU/mL for comparability with subsequent results.^10–15 All specimens positive for anti-HBc or HBsAg were tested for HBV DNA and sequenced, as applicable. ¹²

At the 41-year follow-up, blood was centrifuged at the time of collection, with serum aliquoted and frozen until testing. Serology was performed by the Centers for Disease Control and Prevention (CDC) Arctic Investigations Program laboratory for qualitative and quantitative anti-HBs (Monolisa anti-HBs EIA, Bio-Rad Laboratories) and qualitative anti-HBc (Ortho HBc ELISA, QuidelOrtho). An anti-HBs immunoglobulin G linear standard curve ranging from 0 to 1000 mIU/mL was generated with each test run; the range of detection was 5–1000 mIU/mL. Specimens positive for anti-HBc were sent to the CDC Division of Viral Hepatitis laboratory for HBsAg and HBV DNA testing. Specimens that tested positive for HBV DNA were then sequenced.

Statistical analysis

For this analysis, we defined the long-term follow-up cohort as eligible participants who responded to the primary vaccine series with anti-HBs ≥10 mIU/mL at 6 months after the third dose.^1,16 Postseries anti-HBs levels are reported as geometric mean concentrations with 95% CI. At each timepoint, participants were excluded from all subsequent analysis if they had received an additional dose of hepatitis B vaccine (ie, a nonstudy dose or a study booster dose). Participants who became anti-HBc–positive were included in subsequent analyses of breakthrough infections but no longer contributed data to analyses of anti-HBs duration and booster-dose response. Categorical variables were compared with the Pearson χ² test for independence or the Fisher exact test if expected cell counts were <5. Two-sided p values <0.05 were considered statistically significant. All analyses were conducted in SAS version 9.4 (SAS Institute).

We used a survival model to estimate the time from primary series completion until anti-HBs fell below 10 mIU/mL. For all participants, the beginning of the follow-up period was the date of the third dose of the hepatitis B vaccine. The end of the follow-up period was the date of the first anti-HBs <10 mIU/mL or, for participants whose anti-HBs level never fell below 10 mIU/mL, the date of the last anti-HBs result. For participants who became anti-HBc-positive, the follow-up period ended on the date of the last anti-HBs result at which they were anti-HBc-negative. We used the Gamma distribution for the survival model, the Akaike information criterion statistic to evaluate model fit, and the Kaplan-Meier method to calculate 95% CI. ¹⁷

At 22, 30, 35, and 41 years after the primary series, we estimated the proportion of participants with serologic evidence of protection from HBV infection. Protection was defined as circulating levels of anti-HBs ≥10 mIU/mL or evidence of immune memory with anti-HBs ≥10 mIU/mL at 4 weeks after a booster dose of recombinant hepatitis B vaccine. The survival model was used to estimate the proportion of participants with circulating anti-HBs ≥10 mIU/mL. The proportion of participants with immune memory was calculated by dividing the number of participants who responded to a booster dose by the number of participants who received a booster dose and provided a follow-up blood specimen; 95% CI were estimated using the Clopper-Pearson exact binomial method. Overall protection from HBV infection was calculated as $x_i+\left(y_i*\left(1-x_i\right)\right)$ , where x _i =the proportion of participants with circulating anti-HBs ≥10 mIU/mL at timepoint i, and y _i =the proportion of participants with anti-HBs ≥10 mIU/mL at 4 weeks after a booster dose of hepatitis B vaccine at timepoint i.

Throughout the 41-year study period, we defined a breakthrough infection as a positive anti-HBc result in a person previously negative, with or without detection of HBsAg or HBV DNA. An active breakthrough infection was defined as the detection of HBsAg or HBV DNA in a participant with anti-HBc. Anti-HBc, HBsAg, and HBV DNA were considered transient if ≥1 positive result was followed by negative results for that marker at all subsequent timepoints.

Ethical considerations

This study was approved by the CDC (#6127) and Alaska Area (#2011-03-011) Institutional Review Boards, the Alaska Native Tribal Health Consortium, Norton Sound Health Corporation, Southcentral Foundation Board of Directors, and Yukon-Kuskokwim Health Corporation. At each of the previous follow-up timepoints, participants aged 18 years or above provided written informed consent, parents of children aged below 18 years provided written parental permission, and children aged 8 years or above provided verbal assent. All participants at the 41-year follow-up provided written informed consent. All research was conducted in accordance with the ethical principles of the Declaration of Helsinki.

RESULTS

Initial cohort and follow-up

During 1981–1982, 1630 participants were enrolled and received 3 doses of hepatitis B vaccine (Figure 1). Of these, 1351 participants remained negative for HBsAg and anti-HBc throughout the primary vaccine series, had anti-HBs ≥10 mIU/mL at 6 months after series completion, and were included in this long-term follow-up study. Among the 1351 persons in the cohort, 697 (52%) were female (Table 1). The median age at enrollment was 13 years (range: 0.5–77 y); 939 (70%) participants were children aged below 18 years and 17 (1%) were aged below 1 year. The initial, postseries anti-HBs geometric mean concentration was 1192 mIU/mL (95% CI: 1083–1312 mIU/mL). Of the 1351 persons in the cohort, 460 (34%) participated at 22 years after vaccination, 370 (27%) at 30 years, 291 (22%) at 35 years, and 180 (13%) at 41 years (Figure 1). During the study period, 563 persons received an additional hepatitis B vaccine dose and were ineligible to participate at the 41-year timepoint, leaving 499 (37%) living participants who remained anti-HBc-negative and were potentially eligible at 41 years. At 41 years, study participants were more likely to be female compared to those who were eligible but did not participate, but the groups did not differ in age at enrollment, anti-HBs after primary vaccination, or 10-year anti-HBs (Table 2).

FIGURE 1.

Participant flowchart in an immunogenicity study of a 3-dose primary series of plasma-derived hepatitis B vaccine in Alaska Native persons, inlcuding participation at 22-41 years after vaccination. The dashed lines represent participants excluded from the analysis of serologic evidence of protection. Abbreviations: anti-HBc, antibody to hepatitis B core antigen; anti-HBs, antibody to hepatitis B surface antigen; mIU/mL, milli-international units per milliliter.

TABLE 1.

Characteristics of participants in the long-term follow-up cohort of a 3-dose primary series of plasma-derived hepatitis B vaccine in Alaska Native persons, overall and at 41 years after vaccination

	N (%)
Characteristic	All participants (N=1351)	Participants at 41 years (n=180)
Sex
Female	697 (52)	96 (53)
Male	654 (48)	84 (47)
Age group at enrollment, y
<1	17 (1)	0 (0)
1–4	187 (14)	27 (15)
5–17	735 (54)	117 (65)
18–39	234 (17)	31 (17)
≥40	178 (13)	5 (3)
Anti-HBs at 6 mo after series completion, mIU/mL
10–199	241 (18)	10 (6)
200–499	158 (12)	10 (6)
500–999	195 (14)	22 (12)
≥1000	757 (56)	138 (77)
Last follow-up timepoint (y) a
<22	566 (42)	—
22	241 (18)	—
30	185 (14)	—
35	179 (13)	—
41	180 (13)	180 (100)
Residence at last follow-up
Rural village	1130 (84)	123 (68)
Anchorage	167 (12)	50 (28)
Bethel	54 (4)	7 (4)

^a Last follow-up timepoint contributing anti-HBs data.

Abbreviations: anti-HBs, antibody to hepatitis B surface antigen; mIU/mL, milli-international units per milliliter.

TABLE 2.

Characteristics of participants in a 41-year follow-up study of serologic evidence of protection following primary hepatitis B vaccination compared to persons who were eligible but did not participate

	N (%)
Characteristic	Participated in 41-year follow-up (N=180)	Eligible but did not participate (N=319)	p a
Sex
Female	96 (53)	137 (43)	0.03
Male	84 (47)	182 (57)
Age group at enrollment, y
<1	0 (0)	3 (1)	0.40
1–4	27 (15)	44 (14)
5–17	117 (65)	221 (69)
18–39	31 (17)	40 (13)
≥40	5 (3)	11 (3)
Anti-HBs at 6 mo after series completion, mIU/mL			0.29
10–199	10 (6)	31 (10)
200–499	10 (6)	23 (7)
500–999	22 (12)	42 (13)
≥1000	138 (77)	223 (70)
Anti-HBs at 10 y after series completion, mIU/mL	[n=114]	[n=173]	0.52
<10	8 (7)	22 (13)
10–199	51 (45)	80 (46)
200–499	28 (25)	36 (21)
500–999	10 (9)	15 (9)
≥1000	17 (15)	20 (12)

^a p-value for Pearson χ² test for independence.

Abbreviations: anti-HBs, antibody to hepatitis B surface antigen; mIU/mL, milli-international units per milliliter.

Serologic evidence of protection at 22, 30, 35, and 41 years after primary vaccine series

Based on the survival model, the estimated proportion of participants with circulating anti-HBs ≥10 mIU/mL declined monotonically over time, from 100% at 6 months after completing the primary series to 32% at 41 years after vaccination (Figure 2). However, after accounting for immune memory, ≥88% of initial vaccine responders were estimated to still have serologic evidence of protection at 22–41 years after vaccination (Figure 3).

FIGURE 2.

Kaplan-Meier curve from a survival model estimating the percentage of a cohort of Alaska Native persons (N = 1351) with anti-HBs ≥ 10 milli-international units per milliliter after a 3-dose primary series of plasma-derived hepatitis B vaccine. Abbreviations: anti-HBs, antibody to hepatitis B surface antigen; mIU/mL, milli-international units per milliliter.

FIGURE 3.

Serologic evidence of protection at 22-41 years after a 3-dose primary series of plasma-derived HBV in Alaska Native persons. ^aCirculating Anti-HBs ≥10 mIU/mL at 4 weeks after a booster dose of recombinant HBV. Abbreviations: anti-HBs, antibody to hepatitis B surface antigen; mIU/mL, milli-international units per milliliter.

At 22 years after vaccination, the first follow-up at which a booster dose of hepatitis B vaccine was offered, an estimated 95% of the initial cohort had serologic evidence of protection. According to the survival model, 66% (95% CI: 63%–69%) were protected by circulating levels of anti-HBs ≥10 mIU/mL. Among the 153 participants without protective circulating anti-HBs who received a booster dose and provided a follow-up blood specimen, 129 (84%, 95% CI: 78%–90%) had evidence of humoral immune memory. Thus, we estimate that an additional 29% of the cohort was protected by immune memory (ie, 84% of the 34% of the cohort without circulating anti-HBs ≥10 mIU/mL).

Similarly, an estimated 95% of the initial cohort had serologic evidence of protection at 30 years after vaccination: 53% (95% CI: 50%–56%) had circulating anti-HBs ≥10 mIU/mL and 42% had immune memory, as evidenced by 102/115 (89%, 95% CI: 82%–94%) participants who responded to a booster dose with anti-HBs ≥10 mIU/mL at 4 weeks after vaccination. At 35 years after vaccination, an estimated 88% of the cohort had evidence of protection: 43% (95% CI: 40%–46%) had circulating anti-HBs, and 45% had immune memory, as evidenced by 51/65 (79%, 95% CI: 67%–88%) participants who responded to a booster dose. Finally, an estimated 91% of the cohort had evidence of protection at 41 years after vaccination: 32% (95% CI: 29%–34%) had circulating anti-HBs and 59% had immune memory, as evidenced by 25/29 (86%, 95% CI: 68%–96%) participants who responded to a booster dose.

HBV infections after vaccination

Among the 1351 participants who completed the primary series of hepatitis B vaccine and had a documented initial anti-HBs ≥10 mIU/mL, we identified 16 (1%) persons with breakthrough HBV infections as evidenced by ≥1 positive anti-HBc result (Table 3). Of these 16, anti-HBc was transient in 10 (63%) and persisted in 6 (38%). After administering a questionnaire and reviewing medical records, we found no evidence of symptomatic acute hepatitis. The median age at vaccination was 13 years (range: 1–46 y). The initial postseries anti-HBs geometric mean concentration was 614 mIU/mL (95% CI: 211–1783 mIU/mL), compared to 1204 mIU/mL (95% CI: 1094–1325 mIU/mL) among participants without a breakthrough infection. The median time from series completion to first anti-HBc–positive result was 6 years (range: 4–35 years), and the median age of participants when the breakthrough infection was first detected was 23 years (range: 6–52 years). Participants whose initial anti-HBs was ≥1000 mIU/mL were more likely to have their breakthrough infections detected ≥10 years after vaccination (range: 10–35 y) (4/7 vs. 0/9, p=0.02).

TABLE 3.

Characteristics of Alaska Native persons with detection of antibody to hepatitis B core antigen after a 3-dose primary series of plasma-derived hepatitis B vaccine

	Participants (N=16)
Characteristic	N (%)
Sex
Female	9 (56)
Male	7 (44)
Age group at initial vaccination, y
<5	4 (25)
5–17	8 (50)
≥18	4 (25)
Anti-HBs at 6 mo after series completion, mIU/mL
10–199	5 (31)
200–499	1 (6)
500–999	3 (19)
≥1000	7 (44)
Time from series completion to first anti-HBc detection, y
<5	6 (38)
5–10	6 (38)
≥10	4 (25)
Age group at first anti-HBc detection, y
<5	0
5–17	6 (38)
≥18	10 (63)
Evidence of active infection a	2 (13)
One or more HBV DNA detections	2 (13)
One or more HBsAg detections	1 (6)
No HBV DNA or HBsAg ever detected	14 (88)

^a One participant had HBV DNA detected at 5 years after vaccination, and HBV DNA and HBsAg were detected again 5 years later. The second participant had HBV DNA detected at 7 years after vaccination. Both participants had subsequent negative HBsAg and HBV DNA results.

Abbreviations: anti-HBc, antibody to hepatitis B core antigen; anti-HBs, antibody to hepatitis B surface antigen; mIU/mL, milli-international units per milliliter.

Of the 16 participants with evidence of a breakthrough infection, 2 (13%) persons had transient active HBV infection (ie, ≥1 HBsAg or HBV DNA detection). One participant, who was aged 46 years at enrollment, had HBV DNA detected at 5 years after vaccination; this positive result was followed by negative HBV DNA results over the next 4 annual study visits. HBV DNA and HBsAg were detected in this patient 5 years after the first positive DNA result, and DNA sequencing identified a different sequence type than had been identified 5 years before; DNA and HBsAg were negative 1 year later. The second participant, who was aged 11 years at enrollment, had HBV DNA detected at 7 years after vaccination, and was negative for HBV DNA 1 year later and in all subsequent study visits in which they participated.

DISCUSSION

In this long-term, prospective study among Alaska Native persons who responded to a primary series of plasma-derived hepatitis B vaccine, we found that >90% of participants at 41 years after vaccination still had serologic evidence of protection. Anti-HBs declined substantially over time, and at 41 years after vaccination, most participants were protected by immune memory, rather than circulating anti-HBs. Over the 41 years, only 16 (1%) vaccine responders had evidence of a breakthrough HBV infection, and none had evidence of acute hepatitis. Furthermore, while HBsAg or HBV DNA was detected transiently in 2 participants, none developed chronic hepatitis B. These findings provide evidence that the plasma-derived hepatitis B vaccine administered to children and adults in HBV-endemic regions confers long-term protection against symptomatic acute and chronic HBV infection and that a booster dose is not needed for at least 41 years after vaccination.

Our findings are consistent with other long-term hepatitis B vaccine follow-up studies, which have demonstrated that, while anti-HBs levels wane following primary vaccination and are accompanied by occasional breakthrough infections, robust immune memory persists and very few cases of chronic disease occur.^18–23 We defined serologic protection as circulating anti-HBs ≥10 mIU/mL or an anti-HBs ≥10 mIU/mL 4 weeks after receiving a booster dose of hepatitis B vaccine. A previous study among a subset of our cohort found that all 44 participants had evidence of long-term cellular immune response at 32 years after vaccination, including 13 persons with anti-HBs <10 mIU/mL. ²⁴ These results were consistent with an earlier study on cellular immunity among Taiwanese adolescents at 15–18 years after neonatal vaccination and demonstrate additional mechanisms of vaccine-induced protection. ²⁵

Our findings are most applicable to HBV-endemic regions, where vaccine recipients may have an enhanced long-term immune response due to continued HBV exposure and natural boosting.^26–29 During the first 10 years of follow-up for this cohort, during which HBsAg positivity ranged from 1% to 11% in the population residing in study villages, 8% of vaccinees were determined to have had natural boosting events (defined as a 4-fold increase in anti-HBs over a 1-year period, without concurrent anti-HBc or HBsAg detection), likely reflecting exposure to HBV. ²⁶ However, while HBV infection was highly endemic in western Alaska when we recruited this cohort in 1981, incidence has fallen dramatically over the last several decades, largely due to widespread immunization. ⁸ The progressive decline in HBV seroprevalence in these communities due to ongoing vaccination starting at birth likely resulted in a reduced risk of HBV exposure over the 41 years of follow-up.

Although we evaluated a primary series of plasma-derived vaccine administered to children and adults, there have been several changes to the hepatitis B vaccine formulation and schedule since this study began.^1,30 For example, most hepatitis B vaccine doses administered since the late 1980s have been yeast-derived recombinant vaccines. ¹ The immune responses immediately following primary vaccination with a plasma-derived and recombinant hepatitis B vaccine are similar.^31–34 However, longer-term studies comparing the formulations have reported varying anti-HBs persistence and anamnestic response to a booster dose.^17,35–38 Differences in study design and setting, including patterns of natural boosting among the study populations, may account for some of this variation. Importantly, despite observed differences in longer-term immunogenicity, plasma-derived and recombinant hepatitis B vaccines appear to confer similarly robust protection from chronic HBV infection.^17,35–38

In addition, although children in the United States and elsewhere now begin hepatitis B vaccination at birth, all participants in our study were aged ≥6 months at enrollment, and only 1% were aged below 1 year.^1,30,39 Although primary hepatitis B vaccination elicits a protective immune response in most immunocompetent persons, regardless of age, several studies have demonstrated less durable anti-HBs among children vaccinated in infancy compared with those vaccinated as older children or adults.^1,30,39 Despite more rapid antibody decay, persons who were vaccinated as infants have shown strong anamnestic responses to booster doses of hepatitis B vaccine and durable protection from infection at 15–31 years after primary vaccination.^21,40–43 Beginning hepatitis B vaccination at birth remains an important tool for preventing perinatal HBV transmission in neonates, who are among those at greatest risk for chronic infection and its long-term complications.^6,44

Our study has several limitations. First, our analysis was limited by loss to follow-up. At 22–41 years after vaccination, 13–35% of the original cohort participated at each timepoint, and it is unknown if those who were lost to follow-up differed from participants in anti-HBs persistence, anamnestic response, or evidence of HBV infection. However, the proportion of eligible persons who were recruited was substantially higher, ranging from 36% to 47%. In addition, our active recruitment strategy and use of survival analysis to incorporate data from all participants until the censoring timepoint likely mitigated the impact of selection bias. Second, because the cohort was restricted to participants with a documented seroprotective response to the primary vaccine series and anti-HBs testing is not routinely performed in clinical practice, the overall proportion of the general population who is protected may be lower than we report. Studies conducted across a range of populations suggest that 90%–98% of healthy persons have a seroprotective response to primary hepatitis B vaccination, although the likelihood of response declines with age. ¹ Third, although persons with immunosuppressive disease or therapy were excluded from enrollment at baseline, participants were not reassessed for the development of these conditions at subsequent follow-ups. Thus, we could not assess evidence of long-term protection in immunocompromised persons nor evaluate the impact of the development of immunocompromise on the durability of protection. Importantly, no participants were found to have symptomatic acute hepatitis or chronic hepatitis B throughout the 41-year follow-up period.

In conclusion, we present serologic evidence of persistent protection in a cohort of persons at 41 years after a 3-dose primary series of plasma-derived hepatitis B vaccine in a region endemic for HBV. Our findings add to the body of evidence suggesting that most people demonstrate serologic evidence of protection for more than 4 decades after initially responding to a primary hepatitis B vaccine series and suggest that a booster dose is not needed for most persons living in HBV-endemic regions.