PEDIATRICS Vol. 107 No. 4 April 2001, pp. 626-631

A Randomized Trial of Alternative Two- and Three-Dose Hepatitis B Vaccination Regimens in Adolescents: Antibody Responses, Safety, and Immunologic Memory

William M. Cassidy, MD^*, Barbara Watson, MD, Virginia A. Ioli, BA^§, Karen Williams, MD^¶, Steven Bird, MS^§, and David J. West, PhD

From the ^* Department of Medicine, Louisiana State University Health Science Center, Baton Rouge, Louisiana;  Vaccine Evaluation Center, Albert Einstein Medical Center, Philadelphia, Pennsylvania; ^§ Merck Research Laboratories, West Point, Pennsylvania; and ^¶ Department of Pediatrics, Louisiana State University Health Science Center, Baton Rouge, Louisiana.

	ABSTRACT

Top Abstract Methods Results Discussion Conclusion References

Objectives.  Hoping to increase hepatitis B (HB) vaccination of adolescents, we did the following: 1) studied if modified regimens of the recombinant HB vaccine, Recombivax HB (2 or 3 doses of 5 or 10 µg given over 4 or 6 months), induce protective anti-hepatitis B surface antibody [anti-HBsAb] levels (10 mIU/mL) comparable to the recommended regimen (5 µg at 0 and 1, and 6 months); 2) measured early antibody response after a single dose; and 3) assessed immunologic memory after 2- and 3-dose regimens.

Design.  One thousand twenty-six adolescents were randomized to 1 of 5 treatment groups (10 µg at 0 and 4 or 0 and 6 months; 5 µg at 0 and 6 or 0, 2, and 4 or 0, 1, and 6 months) in an open trial. Anti-HBsAb was measured in all participants just before and 1 month after the last dose, and at several other times in a subset of vaccinees. Anti-HBsAb response to a booster dose 2 years later was examined to assess immunologic memory in participants vaccinated with 5 µg at 0 and 6 or 0, 1, and 6 months.

Results.  All regimens induced 10 mIU/mL of anti-HBs in 95% of vaccinees. Geometric mean titers ranged from 674.8 to 3049.4 mIU/mL. Geometric mean titers were higher with regimens using the following: 1) 10 versus 5 µg; 2) 3 versus 2 doses; and 3) vaccination intervals of 6 versus 4 months. After 6 months, 63.8% of vaccinees given one 10-µg dose had 10 mIU/mL of anti-HBsAb versus 41.6% after one 5-µg dose. Participants vaccinated with either two or three 5-µg doses retained robust immunologic memory.

Conclusions.  The results of this study show that a 2-dose regimen of Recombivax HB is as immunogenic and induces immunologic memory as effectively as the recommended 3-dose regimen. A regimen of two 10-µg doses may be of significant benefit for vaccinees who are poorly compliant or deviate from the intended vaccination schedule.  Key words:  hepatitis B, vaccination, immunization, adolescents.

Hepatitis B (HB) virus infection is a significant public health problem throughout the world. In countries such as the United States, where the overall incidence of HB is relatively low, most infections occur in adolescents and adults.¹ The Advisory Committee for Immunization Practices and several professional medical associations have actively endorsed comprehensive HB vaccination of adolescents who were not previously vaccinated as infants.^2,3

The delivery of such care to adolescents is poorly developed. One barrier to vaccination may be the requirement for a 3-dose series. A 2-dose regimen, which is comparably immunogenic to the recommended 3-dose regimen, might improve vaccine coverage of adolescents. It should also reduce the resources needed to vaccinate adolescents.^a

The primary objective of this study was to assess the safety and immunogenicity of the recombinant HB vaccine, Recombivax HB, given to healthy adolescents using 1 of several investigational regimens listed in the left-hand column of Table 1. A second phase considered whether adolescents developing a protective titer of antibody after a 2-dose regimen retained immunologic memory as well as those vaccinated with the standard 3-dose regimen.

	METHODS

Top Abstract Methods Results Discussion Conclusion References

This study was conducted at the Louisiana State University Health Science Center (LSUHSC) in Baton Rouge, Louisiana, and the Children's Hospital of Philadelphia in Philadelphia, Pennsylvania, between July 1994 and May 1995 under a protocol developed by Merck Research Laboratories. The study was reviewed and approved by the institutional review board (IRB) at LSUHSC and Children's Hospital of Philadelphia. A second part of the study, examining immunologic memory in previously vaccinated participants, involved a subset of participants at the Baton Rouge site. It was conducted under a protocol amendment approved by the IRB at LSUHSC. At both sites and for both study phases, written informed assent and consent was obtained from each adolescent and his or her parent or guardian, respectively.

Participants

Healthy, nonpregnant adolescents 11 to 19 years old who had not previously received HB vaccine and had a negative history of previous HB infection were eligible to participate in the study.

Vaccine

The 5-µg and 10-µg doses of vaccine used in this study were obtained from different lots of Recombivax HB so as to better profile the safety and immunogenicity that can be expected with various regimens when using commercially available lots designed to deliver these dosages. Lot 0220A, containing 10 µg of Hepatitis B surface antigen (HBsAg) in a volume of 1.0 mL (Adult Formulation- ie, approved for use in vaccinees 20 years old and older), was used for all 10-µg doses. Lot 0949W, containing 5 µg of HBsAg in a volume of 0.5 mL (Adolescent Formulation- ie, approved for use in 11- to 19-year-olds), was used for all 5-µg doses. Both lots of vaccine contained thimerosal (mercury derivative) 1:20 000 added as a preservative, plus 0.5 mg of aluminum (as aluminum hydroxide) per mL of vaccine. Injections were given in the deltoid muscle using a 1-inch needle. If the vaccinee was obese, a 1.5-inch needle was used.

Design

In the initial stage of the study, participants were stratified into 2 groups based on their willingness to have 2 extra post-vaccination blood samples drawn to better estimate the kinetics of antibody development. Eligible participants within each group were assigned sequential case numbers that were randomly allocated to 1 of 4 investigational regimens (Table 1) or the standard regimen. Following receipt of an allocation number, 71 participants refused to participate in the study, met protocol exclusion criteria, or did not return for vaccination. These participants were not included in the summaries or analyses. The number of participants who were not vaccinated was evenly distributed across the 5 treatment groups. Participants were followed after each injection for adverse experiences. Blood samples for the determination of anti-hepatitis B surface antibody (anti-HBsAb) were taken just before and 1 month after the last dose of vaccine in the series. From the participants who donated additional samples, blood was drawn either at 2 and 4 months after the initial dose (participants assigned to be vaccinated over a 6-month period) or at 2 and 7 months after the initial dose (participants assigned to be vaccinated over a 4-month period).

In the second part of the study, a subset of vaccinees who received either a 5-µg dose at 0 and 6 months or at 0, 1, and 6 months who had developed 10 mIU/mL of anti-HBsAb had a blood sample drawn ~2 years later to determine the persistence of antibody. They were then given a 5-µg booster dose of vaccine. Blood samples to assess the anamnestic character of the antibody response were drawn 1 and 4 weeks later.

Safety Assessment

Vaccine safety and tolerability were monitored through the use of Vaccination Report Cards supplied to participants that solicited daily recording of body temperature, local injection-site adverse experiences, and systemic adverse experiences on the day of vaccination and for 5 days thereafter. Participants were asked to report any serious adverse experiences occurring within 14 days of vaccination.

Antibody Responses to Primary and Booster Vaccination

All postvaccination sera were tested quantitatively for anti-HBsAb by a radioimmunoassay (Ausab, Abbott Laboratories, Chicago, IL) at the Merck Research Laboratories in West Point, Pennsylvania, with results calculated in milli-International Units per milliliter using an equation derived by Hollinger.⁴

The outcome of primary clinical interest in the first part of the study was prespecified as the proportion of vaccinees having 10 mIU/mL of anti-HBsAb 1 month after the final dose of vaccine. The geometric mean titer (GMT) after the final dose of vaccine and the proportions of vaccinees with any detectable antibody (sample to negative control cpm ratio 2.1) or 10 mIU/mL of anti-HBsAb at all postvaccination times before the final dose of vaccine were outcomes of secondary interest.

In the second part of the study, anamnestic response was measured as described above. An anamnestic response was defined as a fourfold or greater rise in the titer of anti-HBsAb measured ~1 week after booster vaccination in participants with measurable antibody in the prebooster blood sample or the development of 10 mIU/mL of anti-HBsAb in participants without detectable antibody in the prebooster blood sample.

Data Analysis

All participants who returned a Vaccination Report Card after vaccination contributed to the assessment of vaccine safety and tolerability. These data were analyzed by comparing the incidence of injection-site adverse experiences, elevated temperatures (100°F, orally), and other systemic adverse experiences among recipients of an investigational regimen with those among recipients given the standard regimen of 5 µg at 0, 1, and 6 months using Fisher's exact test.

Formal analysis of the anti-HBsAb response to vaccination in the first stage of the study was restricted to participants 11 to 19 years old at entry, who were confirmed negative for all HB serologic markers in a blood sample taken just before the first dose of vaccine, and who received all designated doses of vaccine within a predefined time frame and had a blood sample tested at the key time point. Similarity between an investigational regimen and the standard recommended regimen was evaluated by testing that the difference between the investigational regimen and the standard recommended regimen was significantly less than a prespecified difference of 10 percentage points in the proportion of vaccinees having 10 mIU/mL 1 month after the final dose of vaccine in the series (using an overall error rate of  = 0.025). A multiplicity adjustment was made for the 4 primary comparisons using the critical level of Dunnett test, which was applied using a step-down multiple testing procedure proposed by Bofinger and Bofinger.⁵ With this procedure, an individual P value less than the critical P value supports a conclusion that the 2 regimens are similar.

No hypotheses were tested for the 2 endpoints of secondary interest.

An analysis was done to ascertain that there was no significant statistical interaction between treatment group, study investigator, or whether participants agreed to additional bleeds.

An analysis was also done to explore the effect of demographic variables including age, gender, smoking status, weight, height, and weight-height index (WHI) on the proportion of vaccinees with 10 mIU/mL and the GMT of anti-HBsAb 1 month after the final dose of vaccine for each regimen studied. Logistic regression models were fit for seroprotection rates using the SAS Generalized Linear Models procedure, and analysis of variance was run for GMT using the SAS General Linear Model procedure. The vaccination regimen was always included in the models that were fit. Models were run looking at each demographic variable separately and all demographic variables together. The WHI was calculated as weight in kilograms divided by (height in meters)^p where p is 2.0 for males and 1.5 for females.⁶

	RESULTS

Top Abstract Methods Results Discussion Conclusion References

Participant Characteristics

One thousand twenty-six adolescents enrolled in the study and received at least 1 dose of vaccine. Of these, 967 (94.2%) received all scheduled doses of vaccine and had a blood sample taken at the key time point after the final dose. The ages of the study participants ranged from 11 to 19 years (except for a single male participant who was inadvertently enrolled at 10 years old), with an average age ranging between regimen of 14.2 to 14.5 years. Participants assigned to each of the 5 vaccination regimens were generally similar with respect to gender (female/male ratio, range: 1.02-1.22), race (primarily white, range: 79.1%-81.0%; or black, range: 15.8%-19.9%), smoking status (range: 5.3%-6.8% smokers), WHI (mean ranged from 21.3-21.9 for males and 27.4-28.1 for females), and initial serostatus (range: 94.5%-98.5% negative for anti-HBsAb at baseline).

Safety

Vaccination Report Cards were returned by 1008 (98.2%) of the study participants following 1 or more injections of vaccine. There were no serious adverse experiences attributable to vaccination. No participant withdrew from the study because of a vaccine-related adverse experience. There were no significant differences between regimens in the incidence of any type of adverse experiences, except for the comparison of 2 doses of 5 µg versus 3 doses of 5 µg (50% vs 65%, P = .002). A posthoc statistical comparison focusing on injection site adverse experiences after the first dose of vaccine in any of the regimens found the frequency of such an event to be higher after a 10-µg dose (30.0%) as opposed to a 5-µg dose (19.9%) (P < .001).

Immunogenicity

Each investigational regimen was compared with the recommended 3-dose regimen of 5 µg given at 0, 1, and 6 months with respect to the proportion of vaccinees achieving an anti-HBsAb titer 10 mIU/mL 1 month after the final dose. Among the 967 study participants who completed the assigned vaccination regimen, 897 (92.8%) met all other inclusion criteria and had valid serology test results at the primary time point which could be used for the immunogenicity analysis. Of those vaccinated with 1 of the investigational regimens, 95.1% to 97.9% developed 10 mIU/mL of anti-HBsAb compared with 98.9% of the vaccinees given the 5-µg 0, 1, and 6 months regimen (Table 1). Analysis showed that there was no significant difference between the regimens in the proportion of vaccinees developing 10 mIU/mL of anti-HBsAb after the last dose of vaccine. There did appear to be differences between regimens in the proportions of vaccinees having any detectable amount or 10 mIU/mL of anti-HBsAb at various times before the last dose, as well as the GMT of anti-HBsAb after the last dose. However, because these outcomes were not the subject of a priori hypotheses, no formal statistical comparisons were done.

In a planned exploratory analyses of demographic factors affecting the proportion of each group achieving seroprotection, univariate analysis suggested an effect of age (P = .002), smoking status (P = .041), weight (P = .004), and WHI (P = .001). Logistic regression analysis including all variables; however, revealed increasing age to be the only variable significantly associated with a lower proportion of vaccinees achieving 10 mIU/mL of anti-HBsAb (P = .033) and with a lower GMT (P < .001) after the final dose of vaccine. The effect of age on the antibody response is illustrated in Table 2, which shows the proportions of younger (11 to 15 years old) and older (16 to 19 years old) adolescents developing 10 mIU/mL along with the associated GMT of anti-HBsAb after the last dose of vaccine. In the analysis of GMT, smoking appeared to significantly reduce GMT (P = .039). There were so few smokers in the study, however, (5.3% to 6.8%), it seems that age is the primary demographic factor affecting GMT.

In phase 2 of the study, immunologic memory in response to 2 doses was compared with that after 3 doses (Table 3). An anamnestic response was achieved in 92% of the 0 and 6 month group and in 95% of the 0, 1, and 6 month group.

	DISCUSSION

Top Abstract Methods Results Discussion Conclusion References

To fulfill the goal of universal adolescent vaccination against HB, it will be necessary to reduce barriers to acceptance by and delivery of vaccine to this population. A survey has indicated that adolescents average only 1 visit to health care providers per year.⁷ Furthermore, a chart review of a Kaiser Permanente adolescent outpatient clinic revealed that although 90% of adolescents counseled about HB began the vaccine series, only 11% received 3 doses.⁸ A 2-dose alternative vaccination regimen might improve both acceptance of vaccination and completion of the series by adolescents.

One approach to increasing HB vaccine coverage in the adolescent population is through school-based clinics.^9-13 A 2-dose regimen would benefit such clinics by reducing resources needed to deliver a complete course of vaccine. A regimen with an overall vaccination interval as short as 4 months would also be helpful, making it possible to complete the required series of injections within half a school year. At minimum, any alternative regimen will have to induce a protective anti-HBsAb response comparable to the standard 3-dose regimen and be as safe.

This study found that several investigational 2- and 3-dose regimens with Recombivax HB were generally well-tolerated, and similarly induced seroprotection compared with the recommended 3-dose regimen of 5 µg at 0, 1, and 6 months. There were, however, differences between regimens.

The 5-µg at 0 and 6 month regimen had a lower incidence of injection site and systemic adverse events compared with the 5-µg at 0, 1, and 6 month regimen. As these regimens used the same dose of antigen and injection volume, the difference in the rate of adverse experiences is logically attributable to the fact that the 2-dose regimen entailed 1 less injection and 6 fewer days of follow-up than the 3-dose regimen. The 10-µg dose had a higher frequency of injection site adverse experiences relative to the 5-µg dose. As the 10-µg dose is contained in 1.0 mL, whereas the 5 µg injection is only 0.5 mL, the larger injection volume may explain this higher rate.

As shown in Table 2, age affected both the proportion of vaccinees developing a seroprotective titer of 10 mIU/mL and the GMT of anti-HBsAb 1 month after the final dose of each investigational regimen. In vaccinees given the standard recommended regimen of 5 µg at 0, 1, and 6 months, there was no apparent effect of age on the proportion of vaccinees developing 10 mIU/mL of anti-HBsAb. There was a small effect on the final level of anti-HBsAb with the GMT in vaccinees 16 to 19 years old being approximately twofold lower than that in vaccinees 11 to 15 years old.

The proportion of vaccinees developing an early antibody response also differed between the various regimens. This could have clinical relevance for individuals at high risk of immediate exposure to HB and for any adolescent who begins a vaccine series, but fails to complete it. For the latter group, the early antibody response to an incomplete series of vaccine may represent the final level of immunity and so could be clinically important. In this context, it is interesting to note that the proportions of participants with a minimally detectable level or with 10 mIU/mL of anti-HBsAb were substantially larger 2, 4, and 6 months after a single 10-µg dose than after a single 5-µg dose. Somewhat larger proportions of participants given the first two 5-µg doses of a 3-dose series developed a minimally detectable level or 10 mIU/mL of anti-HBsAb before the last dose than did participants given only the first 10-µg dose of a 2-dose series (Table 1). Similar trends regarding both the early and final antibody responses were observed in the 2 age subgroups, as were noted for the total 11- to 19-year-old population (Table 2).

The GMT of anti-HBsAb following the final dose of vaccine also varied from one regimen to another. Taking one variable at a time, a larger GMT was obtained with 1) an antigen dosage of 10 µg versus 5 µg; 2) a 3-dose versus a 2-dose regimen; and 3) an overall vaccination interval of 6 months versus 4 months. One of these factors can compensate for the effect of another. Thus, for example, a 2-dose regimen of 10 µg given at 0 and 6 months induced a GMT as large as that induced by the standard 3-dose regimen of 5 µg at 0, 1, and 6 months (3049.4 mIU/mL vs 2493.2 mIU/mL).

The most important response to vaccination, however, is achieving 10 mIU/mL of anti-HBsAb, with substantial variations in GMT apparently having little clinical significance. This level of antibody was previously associated with protection against clinically significant HB infection in a number of efficacy trials, and has been endorsed by national and international vaccine advisory groups as indicating a protective response to vaccine.^14-19 The height of an individual's antibody response after vaccination (or the GMT of a population of vaccinees) does correlate with persistence of antibody over time. However, multiple studies have shown that healthy participants with postvaccination anti-HBsAb titers 10 mIU/mL retained vaccine induced immunologic memory for HBsAg over periods of 5 to 12 years.²⁶ Furthermore, this immunologic memory protects against clinically significant HB infection when antibody drops below 10 mIU/mL, or even below the minimum detectable level.²⁰

This study showed that adolescents retained robust immunologic memory for HBsAg 2 years after being immunized with only two 5-µg doses. Just before booster vaccination, only 75.0% of the 2-dose recipients and 84.1% of the 3-dose recipients still retained 10 mIU/mL of anti-HBsAb. A week after booster vaccination, all participants had an anti-HBsAb titer 10 mIU/mL, and the GMT for both the 2-dose and the 3-dose groups rose 66-fold over the prebooster baseline.

	CONCLUSION

Top Abstract Methods Results Discussion Conclusion References

This study supports that a 2-dose regimen of Recombivax HB can be as immunogenic in healthy adolescents as the recommended 3-dose series. Overall, regimens of two 10-µg or two 5-µg doses were similar to each other and to the standard 3-dose regimen with regard to the proportion of vaccinees developing a protective anti-HBsAb titer after the final dose. However, given the observed effect that age had on antibody response, in this study, the 2-dose regimen would seem best suited for younger adolescents. Among the 2-dose regimens tested, a larger proportion of vaccinees developed a protective level of antibody after a single 10-µg dose than after a single 5-µg dose. Increased immunogenicity of the 10-µg dose helps compensate for the direct relationship between vaccination interval and the magnitude of the antibody response, making a shorter interval 2-dose regimen (eg, 0 and 4 months) a more attractive possibility for use in school-based clinics. Finally, results from the follow-up study suggest that a 2-dose regimen can induce a high level of immunity to HB infection with immunologic memory that persists for a long period of time.

	ACKNOWLEDGMENTS

This study was funded by Merck & Co, Inc.

We thank Pam Saloom, RN, and Sharon Piercy, RN, for providing invaluable support in the execution of this study.

	FOOTNOTES

William M. Cassidy, MD, has received honorarium, research grants, and participated in clinical research with Merck & Co, Inc Pharmaceuticals.

Karen Williams, MD, has received research grants and participated in clinical research with Merck & Co, Inc Pharmaceuticals.

Barbara Watson, MD, has received honorarium and research grants and has participated in clinical research with Merck & Co, Inc Pharmaceuticals.

Virginia A. Ioli, BA, and Steven Bird, MS, are both employed by Merck & Co, Inc Pharmaceuticals.

David J. West, PhD, was formerly employed by Merck & Co, Inc Pharmaceuticals. He is now a consultant to them. He owns stock in the company.

^a On September 24, 1999, the Food and Drug Administration approved the regimen of 2 doses of 10 µg given to 11- to 15-year-olds at 0 and 4 to 6 months.

Received for publication Apr 27, 2000; accepted Aug 14, 2000.

Address correspondence to William M. Cassidy, MD, LSU Unit-Earl K, Long Medical Center, 5825 Airline Hwy, Baton Rouge, LA 70805. E-mail: wcassi{at}lsuhsc.edu

	ABBREVIATIONS

HB, Hepatitis B; LSUHSC, Louisiana State University Health Science Center; IRB, institutional review board; HBsAg, hepatitis B surface antigen; GMT, geometric mean antibody titer; WHI, weight-height index; anti-HBsAb, anti-hepatitis B surface antibody.

	REFERENCES

Top Abstract Methods Results Discussion Conclusion References

1. Alter MJ, Hadler SC, Margolis HS, The changing epidemiology of hepatitis B in the United States: need for alternative vaccination strategies. JAMA 1990; 263:1218-1222 [Abstract/Free Full Text]
2. Centers for Disease Control. Immunization of adolescents. Recommendations of the Advisory Committee on Immunization Practices, the American Academy of Pediatrics, the American Academy of Family Practice, and the American Medical Association. MMWR Morb Mortal Wkly Rep. 1996;45(RR-13):1-16
3. Society for Adolescent Medicine Position statement on hepatitis B immunization. J Adolesc Health 1995; 17:256 [Medline]
4. Hollinger FB, Adam E, Heiberg D, Melnick JL. Response to hepatitis B vaccine in a young adult population. In: Szmuness W, Alter HJ, Maynard JE, eds. Viral Hepatitis: 1981 International Symposium. Philadelphia, PA: Franklin Institute Press; 1982:451-466
5. Bofinger E, Bofinger M Equivalence with respect to a control: stepwise tests. J Royal Stat Soc B 1995; 57:731-733
6. Weber DJ, Rutala WA, Samsa GP, Santimaw JE, Lemon SM Obesity as a predictor of poor antibody response to hepatitis B plasma vaccine. JAMA 1985; 254:3187-3189 [Abstract/Free Full Text]
7. National Center for Health Statistics. Pattern of Ambulatory Care in Pediatrics: The National Ambulatory Medical Care Survey, United States, January 1980-December 1981. Hyattsville, MD: US Department of Health and Human Services, Public Health Service, Centers for Disease Control; 1983. DHHS Publication No. (PHS) 84-1736
8. Wong VK, Woodruff C, Shapiro R Compliance of hepatitis B vaccination in patients presenting to a teenage clinic. Pediatr Infect Dis J 1994; 13:936 [Medline]
9. Boyer-Chuanroong L, Woodruff BA, Unti LM, Sumida YU Immunizations from ground zero: lessons learned in urban middle schools. J Sch Health 1997; 67:269-272 [Medline]
10. Centers for Disease Control Hepatitis B vaccination of adolescentsCalifornia, Louisiana, and Oregon, 1992-1994. MMWR Morb Mortal Wkly Rep. 1995; 43:605-609
11. Cassidy WM, Mahoney FJ A hepatitis B vaccination program targeting adolescents. J Adolesc Health 1995; 17:244-247 [CrossRef][Medline]
12. Dobson SR, Scheifele DW, Bell A Evaluation of a universal hepatitis B vaccination program for adolescents in British Columbia [abstract]. Pediatr Res 1993; 33:167A
13. Brookman RR, Koff RS, Schaffner W, Critical issues surrounding hepatitis B vaccination for adolescents: a roundtable. J Adolesc Health 1995; 17:208-233 [CrossRef][Medline]
14. Szmuness W, Stevens CE, Zang EA, A controlled clinical trial of the efficacy of the hepatitis B vaccine (Heptavax B): a final report. Hepatology 1981; 1:377-385 [Medline]
15. Francis DP, Hadler SC, Thompson SE, The prevention of hepatitis B with vaccine. Report of the Centers for Disease Control multi-center efficacy trial among homosexual men. Ann Intern Med 1982; 97:362-368
16. Centers for Disease Control. Recommendation of the Immunization Practices Advisory Committee (ACIP). Recommendations for protection against viral hepatitis. MMWR Morb Mortal Wkly Rep. 1985;34:313-324,329-335
17. Centers for Disease Control. Recommendations of the Immunization Practices Advisory Committee. Update on hepatitis B prevention. MMWR Morb Mortal Wkly Rep. 1987;36:353-360,366
18. International Group Immunisation against hepatitis B. Lancet 1988; 1:875-876 [Medline]
19. World Health Organization. Informal Consultation on Quadrivalent Diphtheria-Tetanus-Pertussis-Hepatitis Vaccine. Final Report. Geneva, Switzerland: World Health Organization; 1992
20. West DJ, Calandra GB Vaccine induced immunologic memory for hepatitis B surface antigen: implications for policy on booster vaccination. Vaccine 1996; 14:1019-1027 [CrossRef][Medline]