Weactogenicity and immunogenicity of inactivated hepatitis A vaccines Max Just* and Rosemarie

Berger

Two inactived hepatitis A virus (HA V) candidate vaccine strain were tested, derivedfrom strains CLF and HA41 75. Neither vaccine increased liver enzymes levels and reactogenicity was similar to that observed with other alum-adsorbed products. Antibody responses were dose-dependent and protection against HAV can be presumed to last for at least three years. All persons receiving 720 ELISA units (E1.U) of the CLF vaccine seroconverted after one dose. For the HM175 vaccine, anti-HA Vpersisted until month 12 after injections at months 0 and 1, suggesting that the third dose of vaccine could be given at any time from month 6 to 12. A double dose of HMI 75 vaccine (1440 E1.U) given as a single bolus resulted in 100% seroconversion by day 14 with a geometric mean anti-HAV level of 121 mIU/ml. This implies that rapid protection can be induced using large doses of inactivated HA V vaccine should time constraints dictate such an approach. Keywords: Hepatitis

A virus; HM 175 strain; reactogenicity;

travellers

INTRODUCTION On 3 1 December 1991, Switzerland became the first country in the world to license the hepatitis A vaccine produced by SmithKline Beecham Biologicals. This decision solved a lot of problems for Swiss travellers who were getting ready for their next vacation season. To assist the licensing authorities in their deliberations about this new vaccine, a number of studies were designed: testing the safety, immunogenicity and kinetics of antibody persistence using three different vaccine doses; testing the consistency of two vaccine lots employing different vaccination schedules; and evaluating a single dose schedule using a double vaccine dose.

MATERIALS

AND METHODS

Vaccines used

In the first study, the candidate vaccine consisted of the CLF virus strain’ lot D1600 (72 ELISA units or E1.U) and lot 87E22/D1600A (360 or 720 E1.U). In the second and third studies, the candidate vaccine was manufactured from the HM175 strain, prepared as described previously2. Vaccine was always injected intramuscularly into the deltoid muscle. Study participants

Volunteers participating in these studies were medical students of Basle University or staff members of the Childrens Hospital, most of whom were 20-30 years of Childrens Hospital, University of Basle, Romergasse 8, Postfach CH 4005, Basle, Switzerland. ‘To whom correspondence should be addressed

SllO

Vaccine, Vol. 10, Suppl. 1, 1992

age. There was no significant difference in the mean ages of the groups or the percentage of males and females. Inclusion criteria included good health, as determined by history taking and physical examination at entry; absence of clinical signs of acute disease at entry; abstinence from eating shellfish during the study and no travel to underdeveloped countries; and participation in a contraceptive programme. Exclusion criteria included a history of liver disease; chronic alcohol consumption; hepatomegaly; and the presence of antibodies to hepatitis A virus (anti-HAV). Written consent was obtained from all volunteers and the studies were conducted in accordance with the provision of the Declaration of Helsinki. The clinical protocols were approved by the Ethical Review Committee of Basle University. Clinical examination

of volunteers

On the day of vaccination (3 and 8 h postvaccination) and once a day for three consecutive days, all subjects were required to record any systemic and local reactions graded as to severity. Biochemical

and serological testing

Liver enzymes [alkaline phosphatase, aspartate aminotransferase (AST), alanine aminotransferase (ALT), yglutamyltranspeptidase (yGTP), lactate dehydrogenase (LDH) and serum bilirubin] were measured using standard laboratory techniques. Sera were examined for antibodies to hepatitis B s and c antigens (anti-HBs and anti-HBc) and hepatitis B serum antigen (HBsAg) using enzyme-immunoassay techniques (AUSAB, COREIA and AUSZYME, respectively). Total antibodies against HAV were measured by RIA 0264-410x/92/100s110-04 @ 1992 Butterworth-Heinemann Ltd

Reactogenicity Table 1

Group characteristics

for vaccination

and immunogenicity

with the CLF inactivated

HAV vaccine Group No.

1+2 4+5 6+7 3

No of Volunteers

Vaccination

40 40 40 10

0.1.2,12

schedule (months)

031.6 0,2,6 0,1,2,12

Vaccine dose (EIU)

Timing of first dose

72 or 360 72 or 360 72 or 366 720

May May July June

1966 WE!8 1966 1968

(HAVAB, Abbott) following the manufacturer’s recommendations. Results were expressed as percentage inhibition of the binding of a radiolabelled HAV-specific antibody preparation to antigen adsorbed on polystyrene beads. According to the manufacturer, a sample is considered positive if the inhibition is > 50%. Because it is possible that subjects with a percentage inhibition lower than 50% also may be protected, we analysed the data using 20% as the cut-off value. An ELISA inhibition assay also was used. In this assay, inactivated HAV and twofold dilutions of serum from vaccinees were incubated in the presence of purified human anti-HAV immunoglobulin G (IgG) adsorbed on to plates. After washing the plates, added purified IgG anti-HAV, coupled to peroxidase, was used as the detector system. Serum samples unreactive at dilutions below 1: 12.5 were considered negative. Titres were expressed as mIU/ml using the reference line method by comparing the results with a standardized immunoglobulin preparation obtained from the WHO. Neutralizing antibodies were assayed in MRC-5 cells using the radioimmunofocusing inhibition test (RIFIT)-7. Titres were expressed as the reciprocal of the highest serum dilution neutralizing 50% of the HAV foci. Neutralizing antibodies were tested against the CLF and HM 175 strains. Study design The first human study was initiated in the spring of 1988. At first, low vaccine doses, 72 and 360 EI.U, were given to 120 volunteers. After the immunogenicity and safety data had been evaluated for the first subjects, an additional group of 10 subjects was selected to receive a higher dose of vaccine at a concentration of 720 E1.U. It should be noted that the dose chosen for the commercial vaccine contains 720 EI.U of HAV antigen. Thus, the doses used in the initial study were relatively low. For these analyses, seroconversion rates were based on an anti-HAV response of 20 mIU/ml or greater, which is a level believed to be protective based on passive immunization data4. Vaccination Several

schedules vaccination

of inactivated

(1-2 days). No events of a serious nature were reported by any subject. The frequency of solicited symptoms did not change appreciably with successive vaccine doses and no difference was observed in the incidence of symptoms between the vaccine dosage groups (Tubfes 2 and 3). The most common general symptom was headache (6.3%). Other general symptoms occurred with an incidence of < 2.4%. Of subjects, z 27% reported soreness at the injection site and 7.6% noted induration. Redness and swelling were each reported in 3.7% of injections. The percentage of subjects reporting local reactions only were similar with successive doses. Overall, 60.5% of the volunteers had no local symptoms to report. Antibody response In all groups, seroconversion rates and the mean concentration of anti-HAV increased with successive vaccine doses (Tables 4 and 5). After the first vaccine dose, the 72 E1.U dose level induced significantly lower seroconversion rates than the 360 E1.U dose vaccine. The geometric mean concentrations of anti-HAV also were significantly lower. After two doses given at monthly intervals, seroconversion rates with the 360 E1.U dose were > 90%. In contrast, only 35-58% of subjects who received the 72 EI.U dose vaccine had seroconverted after two doses. All subjects receiving two doses, given two months apart, seroconverted irrespective of dose level. Geometric mean concentrations, however, were significantly lower in the 72 E1.U group than in the 360 EI.U group. Higher antibody levels were obtained four weeks after receiving three doses of vaccine at 0,1,6 or 0,2,6 months compared with a 0,1,2 month schedule. Whether this difference would be as great if the latter group had been evaluated at the seventh month is unclear. All subjects given three 360 E1.U doses seroconverted, whereas from 85-100% of subjects seroconverted following three 72 E1.U doses. One month after the three Table 2

were evaluated

Dose Dose Dose Dose

1 2 3 4

Overall

with general

reactions”

72 EI.U

360 EIU

720 ELU

4152 2153 II52 o/17

4152 (8) 2l52 (4) 2l51 (4) 1116 (6)

119 (11) 019 (0) 019 (0) 218 (25)

5

9

(8) (4) (2) (0)

4

Reactogenicity

RESULTS Dose Dose Dose Dose

1 2 3 4

Overall

LDH) and

of the CLF vaccine

Number (%) of volunteers receiving

0.

Vaccination did not induce medically significant elevations of enzyme levels. Symptoms were generally mild to moderate in severity and if severe were only short-lasting

of the CLF vaccine

* Liver enzymes (alkaline phosphatase, AST, ALT, yGTP, serum bilirubin were always within normal values

(Table

Safety and reactogenicity

Reactogenicity

Number (%) of volunteers receiving

Table 3

schedules

hepatitis A vaccines: M. Just and R. Berger

with local reactions

72 EI.U

360 EI.U

720 EI.U

19/52 (37) 18153 (34) 11/52 (21) 6117 (35)

15152 (29) 12/52 (23) 14/51 (27) 2/16 (12)

319 (33) 219 (22) 319 (33) 310 (37)

31

25

31

Vaccine, Vol. 10,

Suppl.

1,

1992

Sill

Reactogenicity and immunogenicity of inactivated hepatitis A vaccines. M. Just and R. Berger Table 4 Seroconversion rates for different dose levels and vaccination schedules of the CLF vaccine

Dose

Schedule (months)

Antibody response (%) 4 weeks after dose

First

Second

Third

72 EI.U

0/1/2 0/1/6 0/2/6

11 10 71

56 35 100

89 85 100

360 EI.U

0/1/2 0/1/6 0/2/6

58 85 85

95 90 100

100 100 100

0/1/2

100

100

100

720 EI.U

Table 5 Geometric mean anti-HAV concentrations four weeks after the third dose of the CLF HAV vaccine comparing different dosages and vaccination schedules Vaccination schedule (months)

0/1/2 0/1/6 0/2/6

Anti-HAV (mlU/ml) with dose of

72 EI.U

380 EI.U

720 EI.U

408 699 1212

1134 2242 2893

1287 ND ND

Levels of anti-HAV were evaluated one month after the third dose ND, Not determined

360 EI.U doses, 19 of the 20 sera tested had neutralizing antibodies against both viral strains. A fourth dose given at month 12 to subjects on a 0,1,2 month schedule boosted antibody levels approximately tenfold. Similar dose-related results were observed when results from the RIA test were substituted. As far as the nine analysable subjects given the 720 E1.U dose are concerned, only one dose was needed to elicit detectable antibodies in all vaccinees. Antibody concentration after three years Of the 117 volunteers vaccinated in 1988, 78 (67%) were available for retesting of their hepatitis A antibody level (Table 6). All 40 volunteers vaccinated with three or four doses of 360 or 720 El.U, resulting in peak geometric mean HAV antibody levels of 1300 5100 m I U / ml, depending on the vaccination schedule, still had a level of circulating anti-HAV antibodies that exceeded 20 mIU/ml, a level which is reached following passive immunization with immune globulin. However, as might be expected, not all subjects of the group receiving onetenth of the commercial licensed dose, i.e., 72 El.U, still had sufficient circulating antibody to be protected, although their initial peak geometric antibody level was 150-1100 m I U / m l in the different vaccination schedule groups. Conclusions from the first study The hepatitis A candidate vaccine (CLF) did not induce an increase in the liver enzyme values or in serum bilirubin and was well tolerated. Reactogenicity was similar to that observed for other alum-adsorbed vaccines. Symptoms reported were generally mild or moderate in severity and, if severe, were short-lasting.

Sl12

Vaccine, Vol. 10, Suppl. 1, 1992

Antibody responses were clearly dose-dependent. All subjects given three 360 E1.U doses seroconverted. After vaccination with three or four doses of a vaccine containing at least 360 El.U, protection against a hepatitis A infection can be presumed to last for at least three years. All patients receiving 720 EI.U of this vaccine seroconverted after one dose. Evaluation of two different production lots and the injection of a third dose at month 2, 6 or 12 The candidate hepatitis A vaccine for this evaluation was prepared from the H M I 7 5 strain and contained 720 E1.U per injection. Details of the vaccination schedule can be found in Table 7. O f the 168 subjects enrolled in the study, 147 (88%) returned for their final blood sampling. There was no significant elevation in the liver enzyme levels and no reports of any serious adverse reactions. Reported symptoms were, with a few exceptions (headache), only local ones and were generally mild to moderate in severity. All subjects returning for blood sampling had detectable anti-HAV antibody titres at months 12-14. No differences in the geometric mean antibody levels were found between groups receiving different lots of the H M I 7 5 vaccine using the same vaccination schedule. The geometric mean concentrations were 1261 and 881 m I U / m l for the groups with the schedule 0, 1, 2 and 2260 and 2081 m I U / m l for the groups with the schedule 0, 1, 6. At month 14 and later, all but one of the groups tested (group 2) had geometric mean antibody levels of 200 m I U / m l or higher. All subjects in group 5 receiving vaccine at 0, 1 and 12 months still had detectable antibody titres at month 12, that is 11 months after the initial two doses of vaccine. Following the third dose, the anti-HAV level rose from 463 m I U / m l at month 12 to 9051 in subjects returning at month 13, and 4153 and 9682 m I U / m l in subjects returning at months 14/15 and 16/17, respectively. In summary, no significant differences could be shown between the geometric mean anti-HAV levels between the two vaccine lots tested using the same vaccination schedule. Antibody persistence at month 12 following injections at months 0 and I is such that the third dose of Table 6 Persistence of anti-HAV three years after completion of a primary series of injections with the CLF vaccine

No. of subjects

Dose (El.U)

Still Geometric mean seropositive (%) concentration (mlU/ml)

38 37 3

72 380 720

79 100 100

481 1338 766

Table I Vaccination schedule for the evaluation production lots of HAV vaccine (HM175 strain)

of two different

Vaccination group

Schedule (months)

Group Group Group Group Group

0/1/2 0/1/6 0/1/2 0/1/6 0/1/12

1: lot VHA002A4 2: same as group 1 3: lot VHA005A4 4: same as group 3 5: same as group 3

Reactogenicity and immunogenicity of inactivated hepatitis A vaccines: M. Just and R. Berger

inactivated HAV vaccine could be administered at any time from month 6 to month 12. Single dose schedule using a double dose of vaccine Health-care professionals who are counselling travellers recognize that most will not seek medical advice one or even two months before leaving their home country for an endemic area. The recommended schedule and the response pattern of two doses given at least one month apart makes it difficult to schedule their hepatitis A vaccinations. Therefore, the safety and immunogenicity of a double dose of vaccine (1440 El.U) either as a single bolus injected in one arm or divided and injected simultaneously in both arms were evaluated. Because the precise level of protection after vaccination is not known, a booster dose of 1440 E1.U was given to all study participants at month 6. Blood samples were taken on days 0, 7, 14 and 28, before the booster dose at month 6, and one month later. The candidate hepatitis A vaccine was prepared from the HM175 strain and was injected intramuscularly into the deltoid muscle. The vaccine was injected as a double dose (1440 El.U) in the left upperarm (group 1) or as a single dose (720 El.U) in each arm (group 2). Sixty subjects were initially enrolled in this study; only one subject was unavailable for evaluation at day 28. No serious adverse events were reported by any of the vaccinees during the study. No reaction was considered to be incapacitating. Liver enzyme activities stayed within the normal range for all subjects. The proportion of volunteers in each group who reported local reactions was essentially identical (66%). Thus, the injection of 2 ml vaccine into one arm (group 1) did not result in a higher frequency or an increase in severity of local reactions than when the vaccine dose was split and injected in two different sites (group 2). The seroconversion rates and the geometric mean anti-HAV concentrations (GMC) are given for each group in Table 8. At day 7, none of the vaccinees had responded. However, by day 14 all of the subjects in group 1 and 25 of 29 (86%) of those in group 2 had seroconverted. All subjects who returned at day 28 had anti-HAV antibodies. The G M C s of the two groups were not significantly different at either day 14 or day 28 when individual titres ranged between 22 and 2421 m I U / m l for subjects in group 1 and between 35 and-839 mIU/ml for group 2. The G M C s increased approximately three fold between day 14 and day 28 and was equal to 351 and 466 m I U / m l for groups I and 2, respectively, ranging from 83 to 3389 m I U / m l for both groups. Anti-HAV levels attained several days after receiving a booster dose at month 6 are significantly

Table 8 Immunogenicity of a double dose (1440 EU) of the inactivated hepatitis A vaccine (HM-175) given as a single injection in one arm (group 1) or divided into two doses and given in both arms simultaneously Group Time point tested

No. of subjects

Seroconversion GMC (mlU/ml) rate,

Prevaccination Day 7 Day 14 Day 28

27 27 27 26

0 0 100% 100%

0 0 121 351

Prevaccination Day 7 Day 14 Day 28

29 29 29 29

0 0 86% 100%

0 0 170 466

aThe seroconversion rate is defined as the acquisition of an anti-HAV antibody level of >_ 20 mlU/ml following vaccination in a previously seronegative subject. Differences between the seroconversion rates and the GMC between groups at each time point are not significant (p > 0.05 by ;(2 analysis and t test, respectively) GMC, geometric mean anti-HAV concentration

increased ( G M C of 208 M I U / m l before injection to 2907 seven days after injection). From these results, higher doses of the inactivated HM175 hepatitis A vaccine appears to be safe and welltolerated. Specific antibodies against HAV ( > 20 m I U / ml) were observed in > 92% of the subjects within two weeks of vaccine administration. All volunteers retained and increased their levels of anti-HAV at one month and after six months. A double-dose booster after six months showed a strong response 7 days and one month later (data not shown). Because all volunteers developed antiHAV by one month and the G M C was > 10-fold higher than that of a group of volunteers administered immune globulin 4 and because the G M C was not significantly lower than levels attained after two doses of vaccine administered one month apart, this would imply that when two doses of vaccine cannot be administered because of time constraints, a double dose of vaccine may provide adequate antibody levels for protection. REFERENCES 1 Siegl, G., de Chastonay, J. and Kronauer, G. Propagation and assay of hepatitis A virus in vitro. J. ViroL Methods 1984, 9, 53-67 2 Andre, F.E., Hepburn, A. and D'Hondt, E. Inactivated candidate vaccines for hepatitis A. Prog. Med. ViroL 1990, 17, 72-95 3 Lemon, S.M. and Binn, L.N. Serum neutralizing antibody response to hepatitis A virus. J. Infect. Dis. 1983, 148, 1933-1939 4 Ambrosch, F., Wiedermann, G., Andre, F.E., D'Hondt, E., Delem, A. and Safary, A. Comparison of HAV antibodies induced by vaccination, passive immunization, and natural infection. In: Viral Hepatitis and Liver Disease (Eds Blaine Hollinger, F.B., Lemon, S.M. and Margolis, H.S.) Williams & Wilkins, Baltimore, 1991, pp. 98-100

Vaccine, Vol. 10, Suppl. 1, 1992

$113