1064
CONCISE COMMUNICATIONS
Response to Hepatitis A Vaccination in Human Immunodeficiency Virus–
Infected and –Uninfected Homosexual Men
Graham A. Neilsen,* Neil J. Bodsworth,
and Norbert Watts *
Sydney Sexual Health Centre, Sydney Hospital, Sydney, and Taylor
Square Private Clinic, Darlinghurst, Department of Clinical Chemistry,
Prince of Wales Hospital, Randwick, Australia
The influence of human immunodeficiency virus (HIV) infection and vaccination schedule on the
immunogenicity of a hepatitis A vaccine was examined. Ninety HIV-infected homosexual men
received two vaccinations with hepatitis A vaccine (each 2 mL of 720 ELISA units/mL) either 1 or
6 months apart; 44 HIV-uninfected men received vaccine at study entry and at 6 months. Anti –
hepatitis A virus (HAV) titer after vaccination was measured in 83 HIV-positive and 39 HIVnegative men. Seroconversion (anti-HAV antibody §20 IU/L) after two vaccinations occurred more
frequently in HIV-negative men (100% vs. 88.2%; P Å .03). Anti-HAV titer after two vaccinations
was also significantly greater in HIV-negative men (1086 vs. 101 IU/L; P Å .0001). HIV-positive
men who responded to vaccination had significantly more CD4 lymphocytes (mean, 540/mL) at
baseline than those who did not (280/mL; P Å .033). Vaccine schedule did not affect response.
Vaccination of susceptible patients against HAV should be recommended early in HIV infection
using the shorter course to encourage compliance.
In the past 6 years, significant epidemics of hepatitis A have
occurred in the homosexual communities of New York, San
Francisco, Toronto, Melbourne, Sydney, and elsewhere [1 – 3].
For homosexual men, the risk of infection with hepatitis A
virus (HAV) is increased by higher numbers of sex partners
and by certain sex practices that facilitate the ingestion of feces.
These practices, such as oroanal contact, are often adopted to
reduce the risk of acquiring the human immunodeficiency virus
(HIV) but may be particularly efficient for the transmission of
HAV. In 1991, a safe and effective whole virion vaccine for
hepatitis A became available [4]. This vaccine induces protective levels of anti – HAV antibody in virtually 100% of immunocompetent adults when given according to the recommended
schedule of 720 ELISA units (ELU) at 0, 1, and 6 months [4].
However, many other vaccines, including those against influenza [5], tetanus [5, 6], pneumococcal infection [5, 6], and
hepatitis B [7, 8], have been shown to be less effective in HIVinfected persons, thereby potentially limiting their utility for
Received 31 December 1996; revised 13 May 1997.
Presented in part: XI International Conference on AIDS, Vancouver, Canada,
July 1996 (abstract We.A.3044).
This study was conducted with the approval of the Ethics Review Committee
of the Eastern Sydney Area Health Service and in accordance with the Declaration of Helsinki on Human Experimentation. Informed written consent was
obtained from all participants.
Reprints or correspondence: Dr. Neil J. Bodsworth, Sydney Sexual Health
Centre, Sydney Hospital, PO Box 1614, Sydney, NSW, 2001 Australia.
* Current affiliations: Indonesia HIV/AIDS and STD Prevention and Care
Project, Denpasar, Bali, Indonesia (G.A.N.); Department of Paediatric Endocrinology, Royal Alexandria Hospital for Children, Westmead, Australia (N.W.).
The Journal of Infectious Diseases 1997;176:1064–7
q 1997 by The University of Chicago. All rights reserved.
0022–1899/97/7604–0029$02.00
homosexual men who are also at increased risk of HIV. Among
homosexual men with acute hepatitis A during the Sydney
epidemic of 1991, 27% were coinfected with HIV [3].
In order to examine the influence of HIV infection, HIVrelated immune depression, and vaccine schedule on the immunogenicity of an inactivated hepatitis A vaccine, a double dose
(2 mL of 720 ELU/mL) was administered to 90 HIV-positive
and 44 HIV-negative homosexual men. The study was conducted between August 1993 and June 1995.
Methods
Study population. Subjects were homosexual men with known
HIV antibody status attending one of two sexually transmitted
disease clinics in Sydney. Excluded were men who tested positive
for anti-HAV antibody at screening, had received human immunoglobulin in the previous 9 months, or were receiving immunosuppressive medication.
Vaccine. The vaccine used was a double dose of the licensed
inactivated whole virion hepatitis A vaccine (Havrix; SmithKline
Beecham Biologicals, Rixensart, Belgium). Two 1-mL vials each
containing 720 ELU were mixed, and the contents were administered by the same injection into the deltoid muscle.
Study procedure. HIV-positive volunteers were randomly assigned to receive two vaccinations, either 1 month or 6 months
apart. All HIV-negative men received the longer treatment schedule, with vaccination at study entry and 6 months later (0, 6–
month course). Prior to vaccination, clinical assessment was done
and blood was drawn for anti-HAV serology, liver function, HIV
serology, and CD4 cell count (HIV-positive men only). Followup HIV testing was performed on initially negative subjects.
Subjects receiving vaccination at study entry and 1 month later
(0, 1–month course) were assessed at months 1, 2, and 6 for anti-
JID 1997;176 (October)
Concise Communications
1065
Table 1. Baseline demographic, clinical, and immunologic characteristics by HIV status and vaccine
schedule.
Vaccine schedule*
HIV-negative
patients
0, 6 months
n
Age (years)
Weight (kg)
HIV stage
Asymptomatic (%)
ARC (%)
AIDS (%)
CD4 cells
Mean (/mL)
Range (/mL)
CD4 %
CD8 cells (/mL)
Lymphocytes (/mL)
46
32.0
73.0
—
—
—
—
—
—
—
—
HIV-positive patients
0, 1 months
0, 6 months
Both
48
33.3
72.5
42
33.3
74.0
90
33.3
73.2
75
19
6
62
26
12
69
22
9
569
28 – 1500
23.9
1356
2348
454
20 – 1230
20.5
1451
2245
515
20 – 1500
22.4
1397
2302
NOTE. Data are means. ARC Å AIDS-related complex (non-AIDS CDC group IV disease).
* 0, 1 Å at study entry and 1 month later; 0, 6 Å at study entry and 6 months later.
HAV titer, liver function, and tolerance of the vaccine. Subjects
receiving the 0, 6–month course were similarly assessed at months
1, 6, and 7.
Laboratory. Titer of anti-HAV antibody following vaccination
was determined using a commercial competitive enzyme immunoassay (Enzymun-Test Anti-HAV; Boehringer Mannheim, Mannheim, Germany). Quantitation of this colorimetric assay is
achieved by comparing optical density to cutoff ratios against standard curves. Previous studies of passive immunization with serum
immune globulin have suggested an anti-HAV titer of §20 IU/L
to be protective against hepatitis A in most cases [9].
Statistics. The proportion of subjects developing protective antibody were compared using x2 or, where inappropriate, Fisher’s
exact method. Antibody titers were compared using the MannWhitney U test. Calculation of geometric mean titer (GMT) included subjects with no antibody response. Other continuous variables were compared using a 2-tailed t test.
Results
Study subjects. We recruited 136 homosexual men; 90
were HIV-positive and 46 were HIV-negative. Of these, 134
(90 HIV-positive) received vaccine at least once, and 121 (82
HIV-positive) received vaccine twice. Anti-HAV titer after
vaccination was obtained for 122 participants (83 HIV-positive
and 39 HIV-negative). HIV-negative men were similar in age
and weight to their HIV-positive counterparts. There were no
significant differences among HIV-positive men on different
vaccine schedules with respect to age, weight, clinical status,
or CD4 cell count (table 1).
Anti-HAV response. Among HIV-negative men, seroconversion occurred 1 month after the first and second vaccinations
in 90.2% and 100% of cases, respectively. Among HIV-positive men, seroconversion occurred in 77.9% and 88.2%, respectively (table 2).
HIV-positive men who seroconverted to anti-HAV at any
time during the study had a significantly higher mean CD4 cell
count (540/mL) at baseline than those who did not seroconvert
(280/mL; P Å .033). Only 9 (64%) of 14 HIV-positive subjects
with CD4 cell counts £200/mL responded. Baseline mean CD4
cell percents were also higher in those who responded (23.2%
vs. 12.2%; P Å .019). Otherwise, men who did and did not
respond were similar with respect to age, weight, and study
site (data not shown).
The GMT of anti-HAV antibody was also greater in HIVnegative men than in HIV-positive men (table 2). HIV-positive
men with §200 CD4 cells/mL at baseline had higher GMTs
(130 IU/L) after two vaccinations than subjects with õ200
CD4 cells/mL (20 IU/L; P Å .0001). Vaccine schedule did not
significantly affect response. Seroconversion was observed in
93.3% of HIV-infected men assigned the shorter 0, 1 – month
dose compared with 81.8% assigned the longer 0, 6 – month
schedule.
Safety. The vaccine was generally well tolerated. Local
soreness was the most common side effect, reported by 10%
of HIV-positive and 9% of HIV-negative men. Mild systemic
symptoms such as headache, rash, nausea, lightheadedness, and
myalgia were reported by a further 33% of HIV-positive and
15% of HIV-negative men.
Discussion
In this study, vaccination with two double doses of vaccine
(720 ELU) resulted in a protective serologic response in 88% of
1066
Concise Communications
JID 1997;176 (October)
Table 2. Anti-HAV response following vaccination with hepatitis A vaccine (1440 ELU): influence
of HIV status and vaccination schedule.
Vaccine schedule*
HIV-negative
patients
No. responding following
1 dose
2 doses
GMT (IU/L) following
1 dose
2 doses
HIV-positive patients
P†
0, 6 months
0, 1 months
0, 6 months
Both
37/41
(90.2)
37/37
(100)
36/45
(80.0)
41/44
(93.2)
31/41
(75.6)
26/32
(81.3)
67/86
(77.9)
67/76
(88.2)
.092‡
103.8
1086.2
56.7
112.1
37.0
100.7
46.8
107.1
.0026x
.0001x
.029§
NOTE. % are of subjects responding (anti-HAV §20 IU/L) over no. with postvaccination serology (%). GMT
Å geometric mean titer.
* 0, 1 Å at study entry and 1 month later; 0, 6 Å at study entry and 6 months later.
†
HIV-negative vs. all HIV-positive subjects by ‡x2 test, §Fisher’s exact method, xMann-Whitney rank test.
HIV-infected homosexual men. These results reflect somewhat
greater success than was reported for the single dose in HIVinfected hemophiliacs (76%) [10] and homosexual men (77%)
[11] and far greater success than the 30% – 50% response rates
usually reported for hepatitis B vaccination in HIV-infected
adults [7, 8]. Although this study did not directly compare the
720 and 1440 ELU doses, these findings suggest a greater
immunogenicity with the larger dose in HIV-infected adults
than has been seen in immunocompetent populations [12, 13].
Published data assessing the association of CD4 cell count
with the response to vaccination against other pathogens in
HIV-positive study participants have reflected disparate results.
In some studies, vaccination against hepatitis B [7], influenza
[5], tetanus [5, 6], and Streptococcus pneumoniae [5] was more
successful in participants with higher CD4 cell counts, but this
association was not seen in other studies of hepatitis B [8] and
hepatitis A [11] vaccines. In our population of HIV-infected
homosexual men, a high CD4 cell count correlated with both
an increased chance of seroconversion and a higher mean titer
of anti-HAV antibody. A similar association was observed by
Wilde et al. [14] in HIV-infected hemophiliacs but not in the
homosexual men studied by Hess et al. [11]. However, the
small number (n Å 14) of HIV-infected homosexual men in the
latter study would have limited detection of such an association.
Although Wilde et al. recommended that vaccination against
hepatitis A not be undertaken in persons with õ200 CD4 cells/
mL, we found that 64% of men with this degree of immune
depression could still produce protective levels of antibody.
Vaccination should therefore be offered as early as possible to
all HIV-positive homosexual men with a clinical need.
The GMTs of anti-HAV antibody in HIV-infected men after
two doses of vaccine (101 IU/L) was much lower than in their
HIV-negative counterparts (1086 IU/L). This study did not
examine antibody decline; however, if anti-HAV is lost more
quickly with HIV infection (as has been observed following
vaccination against hepatitis B [15]), then the titer of antiHAV might rapidly decline to below protective levels. Further
research on the rate of antibody decline in HIV infection is
needed before any recommendations for booster vaccinations
can be made. Meanwhile, anti-HAV titer should be estimated
regularly for HIV-positive persons following vaccination.
A nonsignificant trend towards a better response was seen
with the shorter course (93% vs. 82%), but this may be due in
part to a greater mean baseline CD4 cell count (569 vs. 454/
mL). Dose-response studies in larger populations are needed
before the 0, 1 – month schedule can be recommended to encourage compliance and to provide earlier protection.
The 1440 ELU dose was very well tolerated in this population, with no novel, severe, or unusual reactions reported. Headache, myalgia, and nausea were all reported more frequently
by HIV-infected persons but probably represent the underlying
disease.
Acknowledgment
Vaccine for this study was provided by SmithKline Beecham
Biologicals.
References
1. Schomer K, Douglas JM, Cohn DL, et al. Hepatitis A among homosexual
men — United States, Canada, and Australia. MMWR Morb Mortal
Wkly Rep 1992; 41:161 – 4.
2. Stewart T. A cluster of cases of hepatitis A in Victoria. Commun Dis
Intell 1991; 15:246.
JID 1997;176 (October)
Concise Communications
3. Ferson MJ, Stokes ML, Bell SM. Hepatitis A: an epidemic in eastern
Sydney and the gay community: epidemiological aspects. Venereology
1992; 5:154.
4. André FE, D’Hondt E, Delem A, Safary A. Clinical efficacy of the safety
and efficacy of an inactivated hepatitis A vaccine: rationale and summary of findings. Vaccine 1992; 10(suppl 1):S160 – 8.
5. Kroon FP, van Dissell JT, de Jong JC, van Furth R. Antibody response
to influenza, tetanus and pneumococcal vaccines in HIV-seropositive
individuals in relation to the number of CD4 lymphocytes. AIDS 1994;
8:469 – 76.
6. Opravil M, Matter FL, Blaser J, Lüthy R. Poor antibody response after
tetanus and pneumococcal vaccination in immunocompromised, HIVinfected patients. Clin Exp Immunol 1991; 84:185 – 9.
7. Keet IPM, van Doornum G, Safary A, Coutinho RA. Insufficient response
to hepatitis B vaccination in HIV-positive homosexual men. AIDS 1992;
6:509 – 10.
8. Wong EKL, Bodsworth NJ, Slade MA, Mulhall BP, Donovan B. Response
to hepatitis B vaccination in a primary care setting: influence of HIV
infection, CD4 lymphocyte count, and vaccination schedules. Int J STD
AIDS 1996; 7:490 – 4.
1067
9. Stapleton JT. Passive immunisation against hepatitis A. Vaccine 1992;
10(suppl 1):S45 – 7.
10. Tilzey AT, Palmer SJ, Harrington C, O’Doherty MJ. Hepatitis A vaccine
responses in HIV-positive persons with haemophilia. Vaccine 1996; 14:
1039 – 41.
11. Hess G, Clemens R, Bienzle U, Schönfeld C, Schunk B, Bock HL. Immunogenicity and safety of an inactivated hepatitis A vaccine in anti-HIV
positive and negative homosexual men. J Med Virol 1995; 46:40 – 2.
12. van Damme P, Matheı̈ C, Thoelen S, Meheus A, Safary A, André FE.
Single dose inactivated hepatitis A vaccine: rationale and clinical
assessment of the safety and immunogenicity. J Med Virol 1994; 44:
435 – 41.
13. Clemens R, Safary A, Hepburn A, Roche C, Stanburg WJ. Clinical experience with an inactivated hepatitis A vaccine. J Infect Dis 1995;
171(suppl 1):S44 – 9.
14. Wilde JT, Rymes N, Skidmore S, Swann M. Hepatitis A immunisation in
HIV-infected haemophilic patients. Haemophilia 1995; 1:196 – 9.
15. Biggar RJ, Goedert JJ, Hoofnagle JH. Accelerated loss of antibody to
hepatitis B surface antigen among immunodeficient homosexual men
infected with HIV [letter]. N Engl J Med 1987; 316:630 – 1.