Safety and immunogenicity of a tetravalent group B ... - Science Direct

Elsevier 0264-410X(95)00147-6

ELSEVIER

Vaccine, Vol. 14, No. 5, pp. 44&450, 1996 Copyright 0 1996 Elsevier Science Ltd. All rights reserved Printed in Great Britain 026&410X/96 $15+0.00

Safety and immunogenicity of a tetravalent group B streptococcal polysaccharide vaccine in healthy adults Karen L. Kotloff*$ Ali Fattom’f, Lisa Basham?_, Abbas Hawwarit, Scott Harkonenf and Robert Edelman” Proposed strategies for prevention of neonatal group B streptococcal (GBS) infection have included active immunization of pregnant women and passive immunization of high-risk infants with hyperimmune GBS globulin derived from vaccinated plasma donors. To explore the feasibility of a program for generating hyperimmune GBS globulin, we evaluated the safety and immunogenicity of a candidate multivalent GBS vaccine containingpurtfiedpolysaccharide from types Ia, Ib, II, and III among subjects most likely to develop an immune response following vaccination, i.e. those with pre-existing antibody to GBS. Thirty volunteers prescreened for serum antibody to type III GBS were immunized with a single subcutaneous injection of vaccine containing either 10, 25, or 50 ,ug of each polysaccharide type (Group 1). An additional ten volunteers prescreenedfor antibody to type Ia were vaccinated with the 50 yg dose (Group 2). Vaccination was generally well tolerated with minor reactions occurring in 27% of subjects. Using a quantitative enzyme-linked immunosorbent assay (ELISA). the seroconversion rates (>fourfold rise) and geometric mean antibody concentration (GMC in ,ug IgG ml’) 6 weeks after vaccination in Group 1 to type Ia, II, and III were 33% (GMC 5.2), 17% (GMC 3.6), and 70% (GMC 43.4), respectively. Quantitative titers were not available for type Ib, but a fourfold rise in ELISA units was seen in 13% of subjects. In Group 2, seroconversion rates to type Ia and III were 90% (GMC 73.4) and 40% (GMC 22.2), respectively. No signtficant dose-response eflect was detected. Combined analysis of Groups 1 and 2 demonstrated that subjects with prevaccination antibody concentrations >2 ,ug IgG ml-’ had sign$cantly higher type-specific antibody concentrations following vaccination compared with subjects possessing lower levels of antibody before immunization. We conclude that our tetravalent GBS polysaccharide vaccine is safe but only modestly immunogenic in healthy seropositive adults. More potent vaccines will be required for public health use. Copyright 0 1996 Elsevier Science Ltd. Keywords: Group

B Streptococcus:

polysaccharide;

vaccine;

immunization;

Group B streptococci (Streptococcus agalactiae; GBS) are an important cause of invasive infections in neonates, pregnant women, and in adults with underlying diseases or advancing age’. Considering the significant morbidity and mortality that results each year from an estimated 15000 infections’, prevention of GBS is a public health priority”. *Division of Infectious Diseases and Tropical Pediatrics, Department of Pediatrics, and Division of Geographic Medicine, Department of Medicine, Center for Vaccine Development, University of Maryland School of Medicine, Baltimore, MD 21201, USA. TWalter W. Karakawa Laboratory of Microbial Pathogenesis and Vaccine Development, Univax Biologics, Inc., Rockville, MD 20852, USA. fTo whom correspondence should be addressed at: University of Maryland School of Medicine, Center for Vaccine Development, 685 West Baltimore, HSF480, Baltimore, MD 21201, USA. (Received 29 March 1995; accepted 11 July 1995

446

Vaccine

1996 Volume

14 Number

5

antibody

Four serotypes (Ia, Ib, II, and III) of GBS, based on capsular polysaccharide antigens, cause the majority (95%) of clinical illnesses4. The observation that maternal anticapsular antibody is associated with protection from GBS sepsis in neonates has stimulated efforts to develop vaccines that contain the capsular types most commonly associated with disease5. In addition to preventing infection, a potential therapeutic application for a vaccine is to generate hyperimmune GBS globulin from healthy plasma donors for use in high-risk or septic infant8. To date, clinical trials detailing the use of purified GBS capsular polysaccharide vaccines have been limited to monovalent vaccines composed of serotypes Ia, II or 1117-’ ’ . These vaccines are most immunogenic in subjects with pre-existing anticapsular antibody against the homologous serotype’ ‘. Herein we report the safety and immunogenicity of a tetravalent capsular polysaccharide

Tetravalent group B streptococcal polysaccharide vaccine in healthy adult volunteers. To address the feasibility of using this vaccine for generating GBS hyperimmune globulin, subjects most likely to produce high titer antibody were selected, i.e. those with preexisting anticapsular antibody to GBS.

MATERIALS AND METHODS

vaccine: K.L. Kotloff et al.

was used as a reference and assigned 100 ELISA Units (EU). All subjects identified during screening as having 2 1 pg ml-’ of antibody to either type Ia or III by RABA also had 2 1 ,ug ml-’ of type-specific antibody by ELISA. An immune response was defined as a fourfold rise in antibody concentration (for types Ia, II, and III) or EU (for type Ib) 6 weeks after vaccination.

Subjects A total of 176 college student volunteers age 1840 years attending the University of Maryland at College Park were screened to identify 30 healthy subjects, usin? previously described procedures”, with 2 1 ,ug mlantibody against type III GBS (Group 1). An additional 77 community volunteers living in Baltimore were screened to identify ten subjects with pre-existing antibody to type Ia (Group 2). To identify eligible subjects, sera were screened for the presence of specific antibody using radioactive antigen binding (RABA or Farr assay, kindly performed by Dr Carol Baker) and the results were expressed as pug ml-’ 13. Informed consent was obtained according to the guidelines of the Department of Health and Human Services and the University of Maryland Institutional Review Board.

Statistical methods Proportions were compared using Fisher’s exact or chi-square test. When multiple comparisons were made between groups to assess the relative immunogenicity of the vaccine antigens, the Bonferroni approximation was applied. To assess the immune response to vaccination among groups of subjects, the antibody concentrations were log transformed to better approximate normality and the geometric mean concentrations (GMC) were compared using r-tests. Analysis of variance was used to determine factors, such as baseline antibody concentration and vaccine dose, that may be significantly associated with postvaccination levels. These statistical tests were interpreted in a two-tailed fashion to estimate p values.

Vaccination and clinical assessment The vaccine was extracted and purified by Dr. Dennis Kasper from types Ia, Ib, II, and III GBS strains using previously described methods7.9. Eligible subjects belonging to Group 1 were randomly assigned to receive a single subcutaneous injection of tetravalent GBS polysaccharide vaccine at one of three ascending dose levels: 40 ,ug (10 ,ug of each serotype), 100 ,ug (25 pug of each serotype), or 200 ,ug (50 pug of each serotype). The ten subjects belonging to Group 2 were given a single subcutaneous injection of vaccine at the highest (200 pg) dose. Subjects were observed at the study site for 20 min after vaccination for evidence of hypersensitivity reactions. They recorded their pulse and temperature every 8 h for 48 h. At 24 and 48 h after vaccination, each subject returned to the study site to be examined and questioned about local and systemic reactions and to be tested for proteinuria. All signs and symptoms were graded for severity according to predetermined criteria and recorded on standardized forms. A complete blood count with differential and serum transaminase levels were measured 48 h after vaccination. The serum antibody response was measured in blood specimens that were collected prior to vaccination, and 6 weeks, 6 months and 12 months following vaccination.

RESULTS

Measurement of type-specific antibody with an avidin-biotin ELISA The type-specific immune response to vaccination was determined before immunization and again 6 weeks following immunization by measuring antibody concentrations, expressed as ,ug IgG ml’, using an avidinbiotin enzyme-linked immunosorbent assay (ELISA), as described in detail in a companion paper14. Reference sera for types Ia, II, and III were quantified using purified IgG standards. Standards were not available to develop a quantitative assay for type Ib. Therefore, a high titered serum (optical density 1.3 at a 1:20 dilution)

Clinical response to vaccination Eleven subjects (27%) reported a reaction following vaccination: six subjects (15%) experienced local reactions (pain, tenderness, ~3 cm of erythema, or < 1 cm of induration) and six subjects (15%) reported systemic symptoms (nausea, malaise, myalgia, or headache). Eight subjects who reported a reaction rated their symptom as mild, and three (one with local pain, one with headache, and one with malaise) rated the severity as moderate. Reactions were not correlated with either vaccine dose or with pre-existing antibody concentration. No clinically significant laboratory abnormalities were observed. Immune response following vaccination among 30 volunteers prescreened for type III anticapsular antibody (Group 1) A significant dose effect was not found even after controlling for pre-existing antibody level. Therefore, the immune response in Group 1 was assessed by combining the ten volunteers in each of the three vaccine dose levels. A fourfold rise in antibody against Ia, Ib, II, and III was detected 6 weeks after vaccination in 33, 13, 17, and 70% of subjects, respectively (Table I). Taken together, five subjects (17%) showed no response, 13 (45%) exhibited a response to only one antigen, 8 (28%) to two antigens and 3 (10%) to three antigens. No subject responded to all four antigens. Following vaccination, a significant rise in GMC was detected for all three antibody types tested (Ia, II, and III). Immune response following vaccination among volunteers prescreened for type Ia anticapsular antibody (Group 2) Group 2 subjects were evaluated for an immune response to types Ia and III (Table 2). Vaccination elicited a fourfold rise in the concentration to at least

Vaccine 1996 Volume 14 Number 5

447

Tetravalenf

group 8 streptococcal

polysaccharicfe

vaccine:

Table 1 Immune response to tetravalent GBS vaccinea among 30 subjects prescreened for antibody to type III GBS

GBS type la lb

Number (%) of subjects with fourfold rise in type-specific antibody” 10 (33%) 4 (13%) 5 (17%) 21 (70%)

111

GMC of type-specific antibody (range) Prevaccination

Postvaccination

1.6 (l-8) ND 1.6 (O-30) 6.2 (l-39.5)

5.2 (1-260)b

K. L. Kotloff et al. Table 3 Immune response to tetravalent polysaccharide vaccine according to baseline level of type-specific antibody

GBS type

Parameter

Antibody response according to baseline antibody concentration 52 pg ml-’ >2 pug ml-’

la

No. fourfold rise/total GMC prevaccination GMC postvaccination No. fourfold rise/total GMC prevaccination GMC oostvaccination No. fourfold rise/total GMC prevaccination GMC postvaccination

7/26 (27%) 1.2 (l-2) 4.0 (l-82)” 2/19 (11%) 1.2 (O-2) 1.9 (O-7)= 9115 (60%) 1.6 (l-2) 13.0 (2-l 74)a

II !.: (0-72)b 43.4 (2-496)b III

GM&geometric mean concentration of antibody (range), expressed as pg IgG ml-‘; ND=not done. “Subjects with at least 1 ,ug IgG ml-’ of antibody were randomized to receive a vaccine dose of either 40, 100 or 200 pg. bwO.OO1, comparing GMC before and after vaccination for each serotype. “Expressed in ELISA units for type lb, and in IgG pg ml-’ for types la, II, and Ill

Table 2 Immune response to tetravalent GBS vaccine= among ten subjects prescreened for antibody to type la GBS

GBS type

Number (%) of subjects with fourfold rise in type-specific antibody

la Ill

9 (90%) 4 (40%)

12/l 4 (86%) 6.2 (3-11) 62.7 (3-806)a,6 3/10 (30%) 5.3 (3-30) 12.1 (3-72Fb 16125 (64%) 11.2 (3-84) 66.4 (5-496)a,b

GMC=geometric mean concentration of antibody (range), expressed as ,ug IgG ml-‘. Subjects received a vaccine dose of either 40, 100 or 200 ,ug. ‘~0.05, as compared with prevaccination GMC for each serotype. *p2 fig IgG ml-‘.

448

GBS


To achieve the goal of developing hyperimmune GBS immunoglobulin for use as therapy of neonatal infection, plasma donors must be identified who can develop high titered multivalent responses to vaccination. This requirement avoids dilution effects that may result when plasma containing high titers of only a single serotype are pooled. Although prescreening consistently identified subjects who were likely to have a vigorous homotypic immune response, only a minority of subjects (7%) responded to three or more of the four antigens. The paucity of subjects with polyvalent responses suggests that it would be difficult to identify appropriate plasma donors for an immunostimulation program to generate high-titered, broadly reactive GBS immunoglobulin using the tetravalent polysaccharide vaccine. Maternal immunization with a multivalent GBS vaccine is purported to be a promising strategy for prevention of GBS infection in the newborn. It is estimated that nearly 60% of neonatal GBS infections could be prevented if 80% of pregnant women were vaccinated and protective levels of antibody were achieved in 80% of newborns born on or after 34 weeks gestation”. Our study provides an opportunity to evaluate the feasibility of this approach using a tetravalent purified polysaccharide vaccine. Because 80-90% of pregnant women are not immune to GBS15. we specifically examined the antibody response to vaccination among subjects lacking pre-existing GBS antibody ( 12 yg ml-‘), and found that response rates were modest. A fourfold rise in antibody to type Ia, II, and III antigens was achieved in 27, 11, and 60% of subjects, respectively. Furthermore, the postvaccination antibody concentrations against type Ia, II, and III antigens was below 4 pg ml-’ in 65, 79, and 33% of subjects, respectively, suggesting that vaccination would not elicit sufficient maternal antibody to induce transplacental immunity in many neonates.

Tetravalent group B streptococcal polysaccharide Higher doses of polysaccharide antigen (up to 200 pug) did not enhance immunogenicity, although we cannot exclude the possibility that a dose-response may have been detected with larger sample sizes. These data support the contention that better immunogens are needed to induce an adequate rate of protection. There are limitations in our predictions regarding the ability of the tetravalent vaccine to elicit protective immunity based on the quantitative ELISA used in this study. Historically, the level of type-specific antibody presumed to confer protection was derived largely from observations that mothers whose infants develop invasive type III GBS infection are significantly more likely to have levels of type-specific antibody ~2 ,ug ml-‘, as determined by RABA, than women who were genitally colonized with type III GBS but whose babies remained healthy16. Little is known about protective levels of antibody to other types’7-‘9. Although there was significant correlation between the concentration of type III antibody measured by the ELISA and the RABA of Baker et al.‘“, there are quantitative and qualitative differences in the antibody measured by these two methods that may detract from the validity of the comparison. For one, while IgG is the principal antibody isotype elicited by vaccination, RABA measures total antibody concentration whereas our ELISA is specific for IgG. Second, avidity or affinity properties of the specific antibodies measured may affect the quantity of bound antibody detected by ELISA. Taken together, these considerations may explain, at least in part, why the mean antibody concentration measured by RABA was 2-3-fold higher than that measured by ELISA, and suggest that use of ELISA in our study may have underestimated the ability of the vaccine to induce a protective level of antibody. It is also notable that the proportion of subjects who mounted an immune response to tetravalent vaccine in our study is generally lower than the response to monovalent vaccine reported by Kasper, Baker and colleagues’.’ ’ . An important reason for this discrepancy is the difference in definition of response among investigators. Kasper and Baker defined response as > 1 pug ml-’ increase in antibody level four weeks after immunization. We chose a more stringent definition of a fourfold rise in antibody concentration (for types Ia, II, and III) or EU (for type Ib) 6 weeks after vaccination. When Kasper and Baker’s criterion are applied, the number of subjects in our study who are classified as responders increases substantially, and significant differences are no longer seen in the proportion of subjects in the two studies who develop an immune response to serotypes Ia and III (Table 4). However, the response rate to type II vaccine among subjects in our study was significantly lower than the rate following administration of monovalent GBS II vaccine reported by Kasper and Baker approximately 10 years earlier. Aside from differences in methodology, possible reasons for the discrepancy include variations in immunogenicity of the type II polysaccharide antigen contained in the two vaccines, interference in the antigenicity of type II polysaccharide by the other GBS serotypes contained in our multivalent preparation, and genetic differences in immune responsiveness among the study samples. Despite the superior immunogenicity in humans of monovalent type II GBS in the RABA, the protective efficacy of the tetravalent vaccine against challenge with virulent GBS type II in


Table 4 Type-specific immune responsea to tetravalent GBS vaccine among Maryland volunteers using a quantitative avidin-biotin enzyme-linked immunosorbent assay (ELISA) compared with the previously reported response to monovalent vaccine measured by radioactive binding assay (RABA)

Vaccine serotype la II Ill

Immune status before vaccination kg ml-‘)*

No. (%) subjects with >l ,ug ml-’ rise: Baker et a/. Current report (RABA)b (ELISA)

23 >3 2 52 >2

27167 24/24 21/24 15/15 27145 18/l 9

(40) (100) (88) (100) (60) (95)

p value

11128 (39) 12112 (100) 6/19 (32) 6/10 (60)

NS NS l pg ml-’ rise in antibody concentration following vaccination were derived from reference 11. %eference 11.

the mouse model was significantly lower than the efficacy against challenge with GBS type III (16 vs 33%, p=o.02)20, suggesting that type II antibody measurements by RABA may not predict in viva activity. The fact that these different assays yield different antibody determinations does not render either invalid, particularly because neither has yet been correlated with protective immunity against type II infection in humans. The tetravalent vaccine was well-tolerated by volunteers in a wide dose range (40-200 pg) even among subjects with pre-existing immunity. A minority of subjects (27%) developed local or systemic reactions, but these were generally mild, and in the absence of a placebo control group, one cannot determine with certainty how many of these symptoms were attributable to the vaccine. Our findings are comparable to experience with over 300 recipients of monovalent vaccines, including pregnant women, which indicates that GBS polysaccharide vaccines are clinically safe’ ‘. Although efficacy data in humans are unavailable, there is indirect evidence to suggest that GBS infection can be prevented by immunization. Vaccine-induced anticapsular antibodies induce opsonophagocytic activity in vitro and provide protective immunity in animal studies’. Moreover, such antibodies are predominantly IgG and capable of efficient transfer across the placenta where they may provide protection to newborns against GBS sepsis”. Our trial suggests that strategies are warranted to increase the immunogenicity of GBS polysaccharides, such as to couple the polysaccharide or a derivative oligosaccharide to a carrier protein, to produce greater concentrations of high affinity IgG antibody.

ACKNOWLEDGEMENTS We thank Don Haines, Sylvia O’Donnell, and the nursing staff at the Center for Vaccine Development for help in recruitment and assessment of volunteers, the faculty and staff at the University of Maryland Student Health Center for support, Garvin Bixler and Xiuru Li for technical assistance, Dr Steven Wasserman for statistical advise, Dr Carol Baker for performing the RABA and Dr Dennis Kasper for providing the vaccine used in these studies.

Vaccine

1996 Volume

14 Number

5

449

Tetravalent group l3 streptococcal polysaccharide


REFERENCES 1

2

3

4

5

6

7

8

9

10

450

Schwartz, B., Schuchat, A., Oxtoby, M.J., Cochi, S.L., Hightower, A. and Broome, C.V. Invasive group B streptococcal disease in adults. A population-based study in metropolitan Atlanta. J. Am. Med. Assoc. 1991, 266, 1112-1114 Centers for Disease Control and Prevention. Group B streptococcal disease in the United States, 1990: Report from a multistate active surveillance system. Morbid. Mortal. WIdy Rep. 1992, 41, 25-32 New Vaccine Development. Establishing Priorities. Volume I. Diseases of Importance in the United States. Institute of Medicine, National Academy Press, Washington, DC, 1985 Dillon, H.C., Khare, S. and Gray, B.M. Group B streptococcal carriage and disease: a 6-year prospective study. J. Ped. 1987, 110, 31-36 Baker, C.J. and Kasper, D.L. Correlation of maternal antibody deficiency with susceptibility to neonatal group B streptococcal infection. N. Engl. J. Med. 1976, 294, 753-756 Gloser, H., Bachmayer, H. and Helm, A. Intravenous immunoglobulin with high activity against group B Streptococcus. Ped. Infect. Dis. J. 5 Suppl., 1986, S176-S179 Baker, C.J., Edwards, M.S. and Kasper, D.L. lmmunogenicity of polysaccharides from type Ill, group B Streptococcus. J. C/in. Invest. 1978, 61, 1107-1110 Eisenstein, T.K., De Cueninck, B.J., Resavy, D., Shockman, G.D., Carey, R.B. and Swenson, R.M. Quantitative determination in human sera of vaccine-induced antibody to typespecific polysaccharides of group B streptococci using an enzyme-linked immunosorbent assay. J. Infect. Dis. 1983, 147, 847-856 Kasper, D.L., Baker, C.J., Galdes, B., Katzenellenbogen, E. and Jennings, H.J. lmmunochemical analysis and immunogenicity of the type II group B streptococcal capsular polysaccharide. J. C/in. Invest. 1983, 7, 260-269 Baker, C.J., French, M.A., Edwards, M.S., Carpenter, R.J., Hays, B.M. and Kasper, D.L. lmmunization of pregnant women with a polysaccharide vaccine of group B Sfrepfococcus. N. Engl. J. Med. 1988, 319, 118&1185


11 12

13

14

15

16

17

18

19

20

Baker, C.J. and Kasper, D.L. Group B streptococcal vaccines. Rev. Infect. Dis. 1985, 7, 458-467 Levine, M.M., Black, R.E., Clements, M.L. ef a/. Evaluation in humans of attenuated Vibrio cholerae El Tor Ogawa Strain Texas Star-SR as a live oral vaccine. Infect. Immun. 1984, 43, 515-522 Baker, C.J., Kasper, D.L., Tager, I.B., Paredes, A., Alpert, S., McCormack, W.M. and Goroff, D. Quantitative determination of antibody to capsular polysaccharide in infection with type Ill strains of group B Streptococcus. J. C/in. Invest. 1977, 59, 810-818 Basham, L.E., Pavliak, V., Li, X., Hawwari, A., Kotloff, K.L., Edelman, R. and Fattom, A. A simple, quantitative, reproducible avidin-biotin ELISA for the evaluation of group B Streptococcus type-specific antibodies in humans. Vaccine 1996, 14, 439-445 Mohle-Boetani, J.C., Schuchat, A., Pikaytis, D.B., Smith, J.D. and Broome, C.V. Comparison of prevention strategies for neonatal group B streptococcal infection. J. Am. Med. Assoc. 1993, 270, 1442-1448 Baker, C.J., Edwards, M.S. and Kasper, D.L. Role of antibody to native type Ill polysaccharide of group B Streptococcus in infant infection. Pediafrics 1981. 68. 544-549 Stewardson-Krieger, P.B., Albrandt, K., Nevin, T., Kretschmer, R.R. and Gotoff, S.P Perinatal immunity to group B &hemolytic Streptococcus type la. J. Infect. Dis. 1977, 136, 649-654 Vogel, L.C., Boyer, K.M., Gadzala, CA. and Gotoff, S.P. Prevalence of type-specific group B streptococcal antibody in pregnant women. J. fed. 1980, 96, 1047-1051 Chnstensen, K.K., Christensen, P., Dahlander, K., Faxelius, G., Jacobson, B. and Svenningsen, N. Quantitation of serum antibodies to surface antigens of group B streptococci types la, lb, and Ill: low antibody levels in mothers of neonatally infected infants. Stand. J. Infect. Dis. 1980, 12, 105-110 Paoletti, L.C., Wessels, M.R., Rodewald, A.K., Shroff, A.A., Jennings, H.J. and Kasper, D.L. Neonatal mouse protection against infection with multiple group B streptococcal (GBS) serotypes by maternal immunization with a tetravalent GBS polysaccharide-tetanus toxoid conjugate vaccine. Infect. lmmun. 1994, 62,32363243