THE JOURNAL OF INFECTIOUS DISEASES. VOL. 136, SUPPLEMENT. AUGUST 1977 © 1977 by the University of Chicago. All rights reserved.
Interim Report of a Controlled Field Trial of Immunization with Capsular Polysaccharides of H aemophilus influenzae Type band Group C Neisseria meningitidis in Mecklenburg County, North Carolina (March 1974-March 1976)
Approximately 16,000 children, from two months to five years of age, were vaccinated with the capsular polysaccharide of either Haemophilus in{luenzae type b or group C Neisseria meningitidis. Immunizations were carried out in a double-masked, randomized manner; the doses of immunogens used were 10 p.g of H. in{luenzae type b polysaccharide and 25 p.g of the group C N. meningitidis polysaccharide. Immunogenicity of the two vaccines was measured in single, random specimens of blood taken from vaccinees of all ages at various intervals after immunization. A positive effect on formation of serum antibody was observed in children of all ages vaccinated with N. meningitidis polysaccharide, but increased levels of serum antibody to H. influenzae type b were observed only in recipients of that vaccine who were three years of age or older. No untoward reactions to either vaccine were noted, and both vaccines retained their original molecular size after storage for three years. Too few cases of disease have been studied for a definitive assessment of vaccine efficacy; however, a slightly protective effect against meningitis was observed for the H. influenzae type b vaccine in infants up to one year of age.
is approximately 10,000 cases [1, 2]. Although antimicrobial therapy has reduced the mortality rate from almost 100% to 3%-10%, morbidity, which includes fixed central nervous system deficits in "cured" patients, is about 30%-50% [3]. Furthermore, the rapid worldwide emergence of resistance to ampicillin, as well as scattered reports of resistance to tetracycline and chloramphenicol among strains of H.i.b. isolated from patients with meningitis, stresses the importance of prevention rather than cure as the most efficient control measure [4, 5]. There is much evidence that serum antibodies to capsular polysaccharide confer immunity to diseases caused by H.i.b. [6-9]. Injection of the purified capsular polysaccharide of H.i.b. into adults and children older than two years has induced a rapid and long-lived serum antibody response in most recipients [10, II). However, this response was slower and of shorter duration in infants, the individuals with the highest attack rate [10-12]. The H.i.b. polysaccharide vaccines did not induce adverse reactions, and contin-
The prevention of diseases due to Haemophilus influenzae type b (H.i.b.) is of unquestionable value. Studies done in several communities indicate that the annual incidence in the United States of meningitis, the most frequent but not the only serious disease caused by this organism,
This study was initially supported by grant no. NOI-HD2294 COA no. 10 from the National Institute of Child Health and Human Development and later by contract no. 223-75-1203 from the Bureau of Biologics, Food and Drug Administration. The authors wish to thank all the physicians, public health officials, and the community of Mecklenburg County for their continuing support and cooperation in this program. We acknowledge Ms. Evelyn Whitesides and Ms. Elizabeth George for technical assistance, supervision of patient records, specimen recovery, and vaccine distribution; Ms. Karen Fetterly for data processing; and Mr. Ernest Hailey for statistical programming. Dr. Emil C. Gotschlich performed the group C Neisseria meningitidis antibody assays. Please address requests for reprints to Dr. J. C. Parke, Jr., Department of Pediatrics, Charlotte Memorial Hospital, P.O. Box 2554, Charlotte, North Carolina 28234.
S51
Downloaded from http://jid.oxfordjournals.org/ at University of Bath Library & Learning Centre on July 11, 2015
From the Charlotte Memorial Hospital and Medical Center) Charlotte) North Carolina, and the Bureau oj Biologics, Food and Drug Administration, and the Biometry Branch) National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland
J. C. Parke, Jr., Rachel Schneerson, John B. Robbins, and James J. Schlesselman
S52
Materials and Methods
Mecklenburg County has approximately 360,000 residents and includes Charlotte (population, 241,000), the largest city in North Carolina. The demographic features, surveillance system, and incidence of bacterial meningitis have been described [1, 13]. The vaccines were administered in a random, double-masked fashion by pediatricians in private practice, in the Outpatient Department of Charlotte "Memorial Hospital, and in public health clinics, and were available to all children two months to five years old whose parents gave informed consent.' The identity of the number-coded vials was kept by the manufacturer and revealed only in cases of disease and for evaluation of studies of antibody to H.i.b. and "Men C in the 431 randomly selected vaccinees, who donated one blood sam pIe each at various times after immunization. Vaccines were produced by Merck Sharp and Dohme, Rahway, N. J. The one-dose vials contained buffered saline solutions of Men C (50 iLg/ml) and H.i.b. (20 iLg/ml) polysaccharides and were stored at -10 C. Each child was given an sc injection of 0.5 ml, and the parent was requested to record the temperature 4 hr later, observe the local injection site, and report any unusual behavior to the physician. Serologic identification of isolates, measurement of antibodies to capsular polysaccharides by radioimmunoassay, and measurement of capsular antigens by countercurrent immunoelec-
The authors followed the guidelines for obtaining informed consent as described in National Institutes of Health protocol no. 73 CH 20 and in Bureau of Biologics protocol 1
no. 75-6.
at.
trophoresis were done according to methods described previously [10, 14, 15]. Results
Vaccines. Both vaccines contained low levels of endotoxin as measured by the rabbit pyrogen and limulus lysate assays (table 1). Immediate adverse reactions, such as fever, local erythema, swelling, and soreness, were not reported. The molecular size of both polysaccharides estimated by K d values of gel filtration profiles on Sepharose 4B did not change during the three years of storage at -10 C [16]. Attack rate. Approximately 16,000 infants and children were immunized with either of the two vaccines. The age-specific attack rates for disease due to H.i.b., including meningitis and septicemia, among the vaccinees are reported in table 2. Using a conditional likelihood and analysis based on the total of 19 cases of meningitis, we evaluated the overall effectiveness of the H.i.b. vaccine to be 40%. It is unlikely (P 0.13) that the vaccine reduces the attack rate by > 70% or that it increases the attack rate by > 10%. These wide limits of possible activity of H.i.b. reflect the limited number of disease cases to date. The most likely value for the overall effectiveness of the
=
Table 1. Physicochemical characteristics of polysaccharide vaccines used in Mecklenburg County. North Carolina, March 1974-March 1976.
Assay Molecular size (Kd) July 6, 1973 March 15, 1976 Protein (%, wt/wt) Phosphorus (%, wt/wt) Nucleic acid (%, wt/wt) Pentose (%, wt/wt) Sialic acid (%, wt/wt) Endotoxin Rabbit pyrogen test (,ug)t Limulus lysate test (,ug}t
Haemophilus influenzae type b
Group C Neisseria meningitidis
0.30 0.30 2.0 4.1 0.8 29.0 ND
0.34 0.34 0.3 ND* 0.1 ND 79.8
20.0 5.0
50.0 50.0
NOTE. The polysaccharide vaccine lot numbers were H. influenzae MSD lot no. 675,964-00B01 and group C N. meningitidis MSD lot no. 552jC-D038. *ND = not detected. tFinal container dose not yielding fever or coagulation.
Downloaded from http://jid.oxfordjournals.org/ at University of Bath Library & Learning Centre on July 11, 2015
ued observation of the immunized infants suggested an increased level of antibody to H.i.b. as compared with levels in age-matched controls [12]. Accordingly, a double-masked field trial with H.i.b. polysaccharide as the test vaccine and group C Neisseria meningitidis (l'vIen C) capsular polysaccharide as the control vaccine was undertaken to assess the safety and efficacy of these vaccines in an area with a high attack rate of bacterial meningitis and with reliable surveillance.
Parke et
553
Vaccine Field Trial
Table 2. Age-specific attack rates of Haemophilus influenzae type b diseases in Mecklenburg County field trial participants, March 1974-March 1976. Vaccine group Group C Neisseria meningitidis
H. influenzae type b Disease, year of life
Subtotals Meningi tis and septicemia First Second Third Fourth Fifth Totals
No. of cases
2,067 2,571 1,766 1,436 1,183
4 2 1 0 0
9,023
7
2,067 2,571 1,766 1,436 1,183
6 2 1 1 0
9,023
10
vaccine for prevention of both meningi tis and septicemia is 30%. Restricting the analysis to the first year of life, we found that the most likely value for vaccine effectiveness against meningitis was 60%, with a range of probable effectiveness of from zero to 80% (P == 0.13). For prevention of meningitis plus septicemia, the effectiveness during the first year of life was 40%, with a possible range of effectiveness between -20% and 70% (P = 0.14). Only one case of meningitis due to Men C occurred in a recipient of H.i.b. vaccine; this low incidence precluded any evaluation of efficacy for this vaccine. Levels of antibody. The median levels of antibody to H.i.b. in the 431 participants are listed in table 3. The antibody data were related to age at vaccination and at sampling. There were no differences in the overall levels of antibody to H.i.b. between the two vaccine groups for persons up to three years of age. There was a suggestion that levels of antibody to H.i.b. were higher in the homologous vaccine group after the third year of life. With use of an arbitrary limit of 1.0 fJ-g of antibody Iml, it was seen that the fraction of individuals with low levels of antibody was greater among those who received the IVIen C vaccine than among H.i.b. vaccinees: 29 of 39 vs. 14 of 40, respectively, at four years of
Attack rate (per 1,000)
Personyears
No.of cases
Attack rate (per 1,000)
1.94 0.78 0.57 0.00 0.00
2,027 2,473 1,744 1,472 1,234
9 2
4.44 0.81 0.00 0.68 0.00
8,950
12
2,027 2,473 1,744 1,472 1,234
10 3 0 2
8.950
15
2.90 0.78 0.57 0.70 0.00
a
1 0
4.93 1.21 0.00 1.36 0.00
a
°
age (P < 0. 1) and 15 of 36 vs. nine of !$5, res pectively, at five years (P = 0.14). Median levels of antibody in the same 431 trial participants after administration of Men C Table 3. Median levels of serum antibody to Haemophilus influenzae type b (H.i.b.) in children vaccinated with H.i.b. or group C Neisseria meningitidis (Men C) vaccine. Months after vaccination
Year of life when tested First
Second
Third
Fourth
Fifth
H.i.b. vaccinees
0-3 4-6 7-9 10-12
0.18* 0.17 0.13 0.05
0.22 0.23 0.10 0.13
2.2 0.32 0.16 0.20
7.6 1.1 3.8 0.96
6.7 10.5
0-12
0.175
0.195
0.23
1.45
7.4
0-3 4-6 7-9 10-12
0.19 0.15 0.18 0.04
0.43 0.13 0.10 0.21
0.53 0.34 0.13 0.38
0.91 0.35 0.18 0.32
0.93 1.57 7.65 1.15
0-12
0.150
0.160
0.275
0.310
1.35
NDt
1.8
Mene vaccinees
*Antibody levels are expressed as J,Lg of antibody Iml of serum. tND == not determined (some samples were too high in concentration to be read, and dilutions for analysis had not been made at the' time of this symposium).
Downloaded from http://jid.oxfordjournals.org/ at University of Bath Library & Learning Centre on July 11, 2015
Meningitis First Second Third Fourth Fifth
Personyears
554
Parke et al.
Table 4. Median levels of antibody to group C Neisseria meningitidis (Men C) in children vaccinated with Men C or Haemophilus influenzae type b (H.i.b.) vaccine.
First
MenC vaccinees 0-3 4-6 7-9 10-12
Year of life when tested Second
Third
Fourth
Fifth
0.13* 0.00 0.00 0.00
1.5 0.13 0.11 0.15
0.47 0.62 0.42 0.16
1.03 0.76 0.25 0.25
NDt 2.15 0.93 0.60
0.00
0.13
0.37
0.63
0.66
0-3 4-6 7-9 10-12
0.00 0.00 0.00 0.00
0.06 0.00 0.00 0.00
0.00 0.00 0.12 0.10
0.13 0.00 0.05 0.00
0.22 0.00 0.12 0.19
0-12
0.00
0.00
0.00
0.00
0.12
0-12 H.i.b. vaccinees
*Antibody levels are expressed as J,Lg of antibody jml of serum. tND = not determined (see footnote to table 3).
vaccine are reported in table 4. At each year of life, the levels of antibody in the recipients of the Men C vaccine were elevated compared with levels in the group that received the H.i.b. vaccine (P< 0.01; Wilcoxon signed rank test for censored data) [17]. Although the reported median antibody levels in both groups equaled 0.00 in the first year of life, the fraction with no detectable antibody in the group that received
Discussion
Levels of antibody to Men C were higher in the homologous vaccine group than in the H.i.b. vaccinees during each of the first five years of life. Natural antibodies to Men C were infrequent in the first year of life (these antibodies were found in 9% of H.i.b. vaccinees up to one year of age)
Table 5. Clinical and laboratory data on patients with meningitis due to Haemophilus influenzae type b among vaccine field trial participants, March 1974- December 1975. Vaccine administered Group C Neisseria meningitidis
H. influenzae type b
Criteria *
Median
Initial temperature Fever duration WBC (X 10 3/mm 3 ) CSF protein (mg/IOO ml) CSF WBC (X 10 3/mm 3 ) Serum Ag (ngjml) CSF Ag (ngjml) Urine Ag (ng/ml) Illness Severe Moderate Mild
100.7 F 4 days 5.6 176.5 5.4 2 16 0
*WBC
Range 100 F-102.4 F 1-7 days 5.6-11.8 87-305 5.0-8.0 0-8 0-512 0-8
= white blood cells; CSF = cerebrospinal fluid; Ag = antigen.
No. of patients 4 4 4 4 4 4 4 4 0 3
1
Median I03.0F 5 days 8.9 155 1.3
Range 101.6F-I04.4F 2-13 days 4.3-15.4 33-579 4.0-11.6
32
0-128
24 4
0-256
0-32
No. of patients 11 11 11 11 11 7 10 9 3 5 2
Downloaded from http://jid.oxfordjournals.org/ at University of Bath Library & Learning Centre on July 11, 2015
Months after vaccination
H.i.b. vaccine was 59 of 64 as contrasted to 38 of 64 in the "Men C vaccinees. Characteristics of disease. Clinical criteria of fever and severity of disease, routine blood and cerebrospinal fluid analyses, and measurement of the capsular antigen of H.i.b. in the two groups were compared for identification of any effect of immunization on the severity of H.i.b. disease. The small number of patients in the group vaccinated with H.i.b. who were available for this report does not permit a definitive evaluation (table 5). There was no marked difference between the two vaccine groups, although there were no severe cases of meningi tis among the four recipients of H.i.b. vaccine as contrasted with three severe cases among 11 vaccinees given Men C polysaccharide. There was also a suggestion that the H.i.b. vaccinees may have had lower levels of H.i.b. antigen in their sera. Data obtained from 18 patients in the nonimmunized group are not shown. There were no obvious differences between these patients and the group vaccinated with Men C polysaccharide.
Vaccine Field Trial
References
1. Parke, J. C., Jr., Schneerson, R., Robbins, J. B. The attack rate, racial distribution, age incidence and
case fatality rate of Haemophilus intluenzae type b meningitis. J. Pediatr. 81:765-769, 1972. 2. Fraser, D. W., Geil, G. C., Feldman, R. Bacterial meningitis in Bernalillo County, New Mexico: a comparison with three other American populations. Am. J. Epidemiol. 100:29-34, 1974. 3. Sell, H. W., Merrill, R. E., Doyne, E. 0., Zimsky, E. P. Long-term sequellae of Hemophilus inilueruae meningitis. Pediatrics 49:206-211, 1972. 4. Elwell, L. P., DeGraaf, J., Seibert, D., Falkow, S. Plasmid-linked ampicillin resistance in Haemophilus in[luenzae type b. Infec. Immun. 12:404-410, 1975. 5. Hansmann, D. Haemophilus iniluenzae type b resistant to tetracycline. Lancet 2:893, 1975. 6. Pittman, M. The actions of type-specific Haemophilus injluenzae antiserum. J. Exp. Med, 58:683-700, 1933. 7. Alexander, H. E., Heidelberger, M., Leidy, G. The curative element in type b H. iniluenzae rabbit serum. Yale J. BioI. Med. 16:425-434, 1944. 8. Schneerson, R., Rodrigues, L. P., Parke, J. C., Jr., Robbins, J. B. Immunity to diseases caused by Hemophilus injiuenzae type b. II. Specificity and some biologic characteristics of "natural" infection and immunization-induced antibodies to the capsular polysaccharide of H. intlueniae type b. J. Immunol. 107:103-110,1971. 9. Anderson, J. P., Johnston, R. B., Smith, D. H. Human serum activities against Haemophilus iniluenzae type b. J. Clin. Invest. 51:31-38, 1972. 10. Robbins, J. B., Parke, J. C., Jr., Schneerson , R., Whisnant, J. W., Robbins, J. B. Quantitative measurement of "natural" and immunization-induced Haemophilus in[luenzae type b capsular antibodies. Pediatr. Res. 7:103-110,1973. II. Smith, D. H., Peter, G., Ingram, D. L., Harding, L., Anderson, P. Responses of children immunized with the capsular polysaccharides of Hemophilus iniluenzae type b. Pediatrics 52:637-644, 1973. 12. Robbins, J. B. Acquistion of "natural" and immunization-induced immunity to Haemophilus injluenzae type b diseases. In D. Schlessinger [ed.], Microbiology -1975. American Society for Microbiology, Washington, D.C., 1975, p. 400-405. 13. Parke, J. C., Jr., Schneerson, R., Robbins, J. B. In S. H. W. Sell [ed.]. Haemophilus iniluenxae. Vanderbilt University Press, Nashville, Tenn., 1973, p. 251-260. 14. Gotschlich, E. C. A simplication of the radioactive antigen-binding test by a double label technique. 1Imrnunol, 107:910-911, 1971. 15. Prince, A. M., Burke, K. Counterimmunoosmoelectrophoresis for the detection of Australia antigen in human sera. Science 169:593-594, 1970. 16. Gotschlich, E. C. Proposal for collaborative studies for the laboratory evaluation of polysaccharide meningococcal vaccine. Bull. W.H.O. BD/CSM/74.l1, 1974. 17. Halperin, M. Extension of the Wilcoxon-Mann Whitney test for samples censored at the same fixed point J. Am. Stat. Assoc. 55: 125-138, 1960. 18. Gold, R., Lepow, M. L., Coldschneider, I., Draper, T. L., Gotschlich, E. C. Clinical evaluation of group
Downloaded from http://jid.oxfordjournals.org/ at University of Bath Library & Learning Centre on July 11, 2015
and developed slowly (incidence in five-year-old H.i.b. vaccinees, 54~o)' However, ~[en C seemed to be immunogenic when used for immunization of persons of all ages [18]. The molecular weight of polysaccharides has been shown to be directly related to their immunogenicity [19, 20]. Thus, the K d value of 0.34 would seem to be a minimal value for future vaccine trials of Men C polysaccharide in infants. No difference in antibodies to H.i.b. between the two vaccine groups was detected during the first three years of life. In contrast to the infrequent presence of natural antibodies to Men C in infants, natural antibodies to H.i.b. were observed in 83% of infants in the first year of life, and this figure rose to 100~o for the fifth year (in lVren C vaccinees). Some lVren C vaceinees had high levels (20 ,ugjml) at age three to five years. These findings are consistent with the observation that organisms that cross-react with H.i.b. are more common than those that crossreact with Men C [21]. There was a higher percentage of four- to six-year-old vaccinees with lower levels of antibody to H.i.b. in the group given ?\len C vaccine than there was in the same age group that received H.i.b. vaccine. A slightly lower rate of meningitis was observed in the H.i.b. vaccinees. Based on our current data, the most likely value for vaccine effectiveness against meningitis during the first year of life is 60% (probable range, zero to 80~o) and is 40% against meningitis and septicemia (probable range, -10% to 70%). However, it is unlikely that the vaccine is >80~o effective against meningitis during the first five years of life, since the conditional probability of observing an outcome of seven cases among the H.i.b. group and 12 cases among the l\;fcnC vaccinees would be P = 0.03 [17]. This study indicated that a more effective H.i.b. vaccine must be prepared for a high degree of protection against serious diseases caused by this organism. The determinations of molecular size and endotoxin content, the antibody data, and the absence of adverse reactions to vaccines used in this study provide a basis for evaluation of new vaccine preparations.
855
S56
A and group C meningococcal polysaccharide vaccines in infants. J. Clin. Invest. 56:1536-1547, 1975. 19. Kabat, E. A., Bezer, A. E. The effect of variation in molecular weight on the antigenicity of dextran in man. Arch. Biochem. Biophys. 78:306-314, 1958. 20. Howard, J. G., Zola, H., Christie, G. R., Courtney, B. M. Studies on immunological paralysis. V. The influence of molecular weight on the immunogenicity, tolerogenicity and antibody-neutralizing activity of type III pneumococcal polysaccharide. Immunology 21:535-546, 1971.
Parke et al.
21. Robbins, J. B., Schneerson, R., Liu, T., Schiffer, M. S. Schiffman, G., Myerowitz, R. L., McCracken, G. H., 0rskov, I., 0rskov, F. Cross-reacting bacterial antigens and immunity to disease caused by encapsulated bacteria. In E. Neter and F. Milgram [ed.], The immune system and infectious diseases. Four International Convocation on Immunology, 1974, Buffalo, N. Y. Karger, Basel, 1975, p. 218-241.
Downloaded from http://jid.oxfordjournals.org/ at University of Bath Library & Learning Centre on July 11, 2015
Interim Report of a Controlled Field Trial of Immunization with Capsular Polysaccharides of H aemophilus influenzae Type band Group C Neisseria meningitidis in Mecklenburg County, North Carolina (March 1974-March 1976)
Approximately 16,000 children, from two months to five years of age, were vaccinated with the capsular polysaccharide of either Haemophilus in{luenzae type b or group C Neisseria meningitidis. Immunizations were carried out in a double-masked, randomized manner; the doses of immunogens used were 10 p.g of H. in{luenzae type b polysaccharide and 25 p.g of the group C N. meningitidis polysaccharide. Immunogenicity of the two vaccines was measured in single, random specimens of blood taken from vaccinees of all ages at various intervals after immunization. A positive effect on formation of serum antibody was observed in children of all ages vaccinated with N. meningitidis polysaccharide, but increased levels of serum antibody to H. influenzae type b were observed only in recipients of that vaccine who were three years of age or older. No untoward reactions to either vaccine were noted, and both vaccines retained their original molecular size after storage for three years. Too few cases of disease have been studied for a definitive assessment of vaccine efficacy; however, a slightly protective effect against meningitis was observed for the H. influenzae type b vaccine in infants up to one year of age.
is approximately 10,000 cases [1, 2]. Although antimicrobial therapy has reduced the mortality rate from almost 100% to 3%-10%, morbidity, which includes fixed central nervous system deficits in "cured" patients, is about 30%-50% [3]. Furthermore, the rapid worldwide emergence of resistance to ampicillin, as well as scattered reports of resistance to tetracycline and chloramphenicol among strains of H.i.b. isolated from patients with meningitis, stresses the importance of prevention rather than cure as the most efficient control measure [4, 5]. There is much evidence that serum antibodies to capsular polysaccharide confer immunity to diseases caused by H.i.b. [6-9]. Injection of the purified capsular polysaccharide of H.i.b. into adults and children older than two years has induced a rapid and long-lived serum antibody response in most recipients [10, II). However, this response was slower and of shorter duration in infants, the individuals with the highest attack rate [10-12]. The H.i.b. polysaccharide vaccines did not induce adverse reactions, and contin-
The prevention of diseases due to Haemophilus influenzae type b (H.i.b.) is of unquestionable value. Studies done in several communities indicate that the annual incidence in the United States of meningitis, the most frequent but not the only serious disease caused by this organism,
This study was initially supported by grant no. NOI-HD2294 COA no. 10 from the National Institute of Child Health and Human Development and later by contract no. 223-75-1203 from the Bureau of Biologics, Food and Drug Administration. The authors wish to thank all the physicians, public health officials, and the community of Mecklenburg County for their continuing support and cooperation in this program. We acknowledge Ms. Evelyn Whitesides and Ms. Elizabeth George for technical assistance, supervision of patient records, specimen recovery, and vaccine distribution; Ms. Karen Fetterly for data processing; and Mr. Ernest Hailey for statistical programming. Dr. Emil C. Gotschlich performed the group C Neisseria meningitidis antibody assays. Please address requests for reprints to Dr. J. C. Parke, Jr., Department of Pediatrics, Charlotte Memorial Hospital, P.O. Box 2554, Charlotte, North Carolina 28234.
S51
Downloaded from http://jid.oxfordjournals.org/ at University of Bath Library & Learning Centre on July 11, 2015
From the Charlotte Memorial Hospital and Medical Center) Charlotte) North Carolina, and the Bureau oj Biologics, Food and Drug Administration, and the Biometry Branch) National Institute of Child Health and Human Development, National Institutes of Health, Bethesda, Maryland
J. C. Parke, Jr., Rachel Schneerson, John B. Robbins, and James J. Schlesselman
S52
Materials and Methods
Mecklenburg County has approximately 360,000 residents and includes Charlotte (population, 241,000), the largest city in North Carolina. The demographic features, surveillance system, and incidence of bacterial meningitis have been described [1, 13]. The vaccines were administered in a random, double-masked fashion by pediatricians in private practice, in the Outpatient Department of Charlotte "Memorial Hospital, and in public health clinics, and were available to all children two months to five years old whose parents gave informed consent.' The identity of the number-coded vials was kept by the manufacturer and revealed only in cases of disease and for evaluation of studies of antibody to H.i.b. and "Men C in the 431 randomly selected vaccinees, who donated one blood sam pIe each at various times after immunization. Vaccines were produced by Merck Sharp and Dohme, Rahway, N. J. The one-dose vials contained buffered saline solutions of Men C (50 iLg/ml) and H.i.b. (20 iLg/ml) polysaccharides and were stored at -10 C. Each child was given an sc injection of 0.5 ml, and the parent was requested to record the temperature 4 hr later, observe the local injection site, and report any unusual behavior to the physician. Serologic identification of isolates, measurement of antibodies to capsular polysaccharides by radioimmunoassay, and measurement of capsular antigens by countercurrent immunoelec-
The authors followed the guidelines for obtaining informed consent as described in National Institutes of Health protocol no. 73 CH 20 and in Bureau of Biologics protocol 1
no. 75-6.
at.
trophoresis were done according to methods described previously [10, 14, 15]. Results
Vaccines. Both vaccines contained low levels of endotoxin as measured by the rabbit pyrogen and limulus lysate assays (table 1). Immediate adverse reactions, such as fever, local erythema, swelling, and soreness, were not reported. The molecular size of both polysaccharides estimated by K d values of gel filtration profiles on Sepharose 4B did not change during the three years of storage at -10 C [16]. Attack rate. Approximately 16,000 infants and children were immunized with either of the two vaccines. The age-specific attack rates for disease due to H.i.b., including meningitis and septicemia, among the vaccinees are reported in table 2. Using a conditional likelihood and analysis based on the total of 19 cases of meningitis, we evaluated the overall effectiveness of the H.i.b. vaccine to be 40%. It is unlikely (P 0.13) that the vaccine reduces the attack rate by > 70% or that it increases the attack rate by > 10%. These wide limits of possible activity of H.i.b. reflect the limited number of disease cases to date. The most likely value for the overall effectiveness of the
=
Table 1. Physicochemical characteristics of polysaccharide vaccines used in Mecklenburg County. North Carolina, March 1974-March 1976.
Assay Molecular size (Kd) July 6, 1973 March 15, 1976 Protein (%, wt/wt) Phosphorus (%, wt/wt) Nucleic acid (%, wt/wt) Pentose (%, wt/wt) Sialic acid (%, wt/wt) Endotoxin Rabbit pyrogen test (,ug)t Limulus lysate test (,ug}t
Haemophilus influenzae type b
Group C Neisseria meningitidis
0.30 0.30 2.0 4.1 0.8 29.0 ND
0.34 0.34 0.3 ND* 0.1 ND 79.8
20.0 5.0
50.0 50.0
NOTE. The polysaccharide vaccine lot numbers were H. influenzae MSD lot no. 675,964-00B01 and group C N. meningitidis MSD lot no. 552jC-D038. *ND = not detected. tFinal container dose not yielding fever or coagulation.
Downloaded from http://jid.oxfordjournals.org/ at University of Bath Library & Learning Centre on July 11, 2015
ued observation of the immunized infants suggested an increased level of antibody to H.i.b. as compared with levels in age-matched controls [12]. Accordingly, a double-masked field trial with H.i.b. polysaccharide as the test vaccine and group C Neisseria meningitidis (l'vIen C) capsular polysaccharide as the control vaccine was undertaken to assess the safety and efficacy of these vaccines in an area with a high attack rate of bacterial meningitis and with reliable surveillance.
Parke et
553
Vaccine Field Trial
Table 2. Age-specific attack rates of Haemophilus influenzae type b diseases in Mecklenburg County field trial participants, March 1974-March 1976. Vaccine group Group C Neisseria meningitidis
H. influenzae type b Disease, year of life
Subtotals Meningi tis and septicemia First Second Third Fourth Fifth Totals
No. of cases
2,067 2,571 1,766 1,436 1,183
4 2 1 0 0
9,023
7
2,067 2,571 1,766 1,436 1,183
6 2 1 1 0
9,023
10
vaccine for prevention of both meningi tis and septicemia is 30%. Restricting the analysis to the first year of life, we found that the most likely value for vaccine effectiveness against meningitis was 60%, with a range of probable effectiveness of from zero to 80% (P == 0.13). For prevention of meningitis plus septicemia, the effectiveness during the first year of life was 40%, with a possible range of effectiveness between -20% and 70% (P = 0.14). Only one case of meningitis due to Men C occurred in a recipient of H.i.b. vaccine; this low incidence precluded any evaluation of efficacy for this vaccine. Levels of antibody. The median levels of antibody to H.i.b. in the 431 participants are listed in table 3. The antibody data were related to age at vaccination and at sampling. There were no differences in the overall levels of antibody to H.i.b. between the two vaccine groups for persons up to three years of age. There was a suggestion that levels of antibody to H.i.b. were higher in the homologous vaccine group after the third year of life. With use of an arbitrary limit of 1.0 fJ-g of antibody Iml, it was seen that the fraction of individuals with low levels of antibody was greater among those who received the IVIen C vaccine than among H.i.b. vaccinees: 29 of 39 vs. 14 of 40, respectively, at four years of
Attack rate (per 1,000)
Personyears
No.of cases
Attack rate (per 1,000)
1.94 0.78 0.57 0.00 0.00
2,027 2,473 1,744 1,472 1,234
9 2
4.44 0.81 0.00 0.68 0.00
8,950
12
2,027 2,473 1,744 1,472 1,234
10 3 0 2
8.950
15
2.90 0.78 0.57 0.70 0.00
a
1 0
4.93 1.21 0.00 1.36 0.00
a
°
age (P < 0. 1) and 15 of 36 vs. nine of !$5, res pectively, at five years (P = 0.14). Median levels of antibody in the same 431 trial participants after administration of Men C Table 3. Median levels of serum antibody to Haemophilus influenzae type b (H.i.b.) in children vaccinated with H.i.b. or group C Neisseria meningitidis (Men C) vaccine. Months after vaccination
Year of life when tested First
Second
Third
Fourth
Fifth
H.i.b. vaccinees
0-3 4-6 7-9 10-12
0.18* 0.17 0.13 0.05
0.22 0.23 0.10 0.13
2.2 0.32 0.16 0.20
7.6 1.1 3.8 0.96
6.7 10.5
0-12
0.175
0.195
0.23
1.45
7.4
0-3 4-6 7-9 10-12
0.19 0.15 0.18 0.04
0.43 0.13 0.10 0.21
0.53 0.34 0.13 0.38
0.91 0.35 0.18 0.32
0.93 1.57 7.65 1.15
0-12
0.150
0.160
0.275
0.310
1.35
NDt
1.8
Mene vaccinees
*Antibody levels are expressed as J,Lg of antibody Iml of serum. tND == not determined (some samples were too high in concentration to be read, and dilutions for analysis had not been made at the' time of this symposium).
Downloaded from http://jid.oxfordjournals.org/ at University of Bath Library & Learning Centre on July 11, 2015
Meningitis First Second Third Fourth Fifth
Personyears
554
Parke et al.
Table 4. Median levels of antibody to group C Neisseria meningitidis (Men C) in children vaccinated with Men C or Haemophilus influenzae type b (H.i.b.) vaccine.
First
MenC vaccinees 0-3 4-6 7-9 10-12
Year of life when tested Second
Third
Fourth
Fifth
0.13* 0.00 0.00 0.00
1.5 0.13 0.11 0.15
0.47 0.62 0.42 0.16
1.03 0.76 0.25 0.25
NDt 2.15 0.93 0.60
0.00
0.13
0.37
0.63
0.66
0-3 4-6 7-9 10-12
0.00 0.00 0.00 0.00
0.06 0.00 0.00 0.00
0.00 0.00 0.12 0.10
0.13 0.00 0.05 0.00
0.22 0.00 0.12 0.19
0-12
0.00
0.00
0.00
0.00
0.12
0-12 H.i.b. vaccinees
*Antibody levels are expressed as J,Lg of antibody jml of serum. tND = not determined (see footnote to table 3).
vaccine are reported in table 4. At each year of life, the levels of antibody in the recipients of the Men C vaccine were elevated compared with levels in the group that received the H.i.b. vaccine (P< 0.01; Wilcoxon signed rank test for censored data) [17]. Although the reported median antibody levels in both groups equaled 0.00 in the first year of life, the fraction with no detectable antibody in the group that received
Discussion
Levels of antibody to Men C were higher in the homologous vaccine group than in the H.i.b. vaccinees during each of the first five years of life. Natural antibodies to Men C were infrequent in the first year of life (these antibodies were found in 9% of H.i.b. vaccinees up to one year of age)
Table 5. Clinical and laboratory data on patients with meningitis due to Haemophilus influenzae type b among vaccine field trial participants, March 1974- December 1975. Vaccine administered Group C Neisseria meningitidis
H. influenzae type b
Criteria *
Median
Initial temperature Fever duration WBC (X 10 3/mm 3 ) CSF protein (mg/IOO ml) CSF WBC (X 10 3/mm 3 ) Serum Ag (ngjml) CSF Ag (ngjml) Urine Ag (ng/ml) Illness Severe Moderate Mild
100.7 F 4 days 5.6 176.5 5.4 2 16 0
*WBC
Range 100 F-102.4 F 1-7 days 5.6-11.8 87-305 5.0-8.0 0-8 0-512 0-8
= white blood cells; CSF = cerebrospinal fluid; Ag = antigen.
No. of patients 4 4 4 4 4 4 4 4 0 3
1
Median I03.0F 5 days 8.9 155 1.3
Range 101.6F-I04.4F 2-13 days 4.3-15.4 33-579 4.0-11.6
32
0-128
24 4
0-256
0-32
No. of patients 11 11 11 11 11 7 10 9 3 5 2
Downloaded from http://jid.oxfordjournals.org/ at University of Bath Library & Learning Centre on July 11, 2015
Months after vaccination
H.i.b. vaccine was 59 of 64 as contrasted to 38 of 64 in the "Men C vaccinees. Characteristics of disease. Clinical criteria of fever and severity of disease, routine blood and cerebrospinal fluid analyses, and measurement of the capsular antigen of H.i.b. in the two groups were compared for identification of any effect of immunization on the severity of H.i.b. disease. The small number of patients in the group vaccinated with H.i.b. who were available for this report does not permit a definitive evaluation (table 5). There was no marked difference between the two vaccine groups, although there were no severe cases of meningi tis among the four recipients of H.i.b. vaccine as contrasted with three severe cases among 11 vaccinees given Men C polysaccharide. There was also a suggestion that the H.i.b. vaccinees may have had lower levels of H.i.b. antigen in their sera. Data obtained from 18 patients in the nonimmunized group are not shown. There were no obvious differences between these patients and the group vaccinated with Men C polysaccharide.
Vaccine Field Trial
References
1. Parke, J. C., Jr., Schneerson, R., Robbins, J. B. The attack rate, racial distribution, age incidence and
case fatality rate of Haemophilus intluenzae type b meningitis. J. Pediatr. 81:765-769, 1972. 2. Fraser, D. W., Geil, G. C., Feldman, R. Bacterial meningitis in Bernalillo County, New Mexico: a comparison with three other American populations. Am. J. Epidemiol. 100:29-34, 1974. 3. Sell, H. W., Merrill, R. E., Doyne, E. 0., Zimsky, E. P. Long-term sequellae of Hemophilus inilueruae meningitis. Pediatrics 49:206-211, 1972. 4. Elwell, L. P., DeGraaf, J., Seibert, D., Falkow, S. Plasmid-linked ampicillin resistance in Haemophilus in[luenzae type b. Infec. Immun. 12:404-410, 1975. 5. Hansmann, D. Haemophilus iniluenzae type b resistant to tetracycline. Lancet 2:893, 1975. 6. Pittman, M. The actions of type-specific Haemophilus injluenzae antiserum. J. Exp. Med, 58:683-700, 1933. 7. Alexander, H. E., Heidelberger, M., Leidy, G. The curative element in type b H. iniluenzae rabbit serum. Yale J. BioI. Med. 16:425-434, 1944. 8. Schneerson, R., Rodrigues, L. P., Parke, J. C., Jr., Robbins, J. B. Immunity to diseases caused by Hemophilus injiuenzae type b. II. Specificity and some biologic characteristics of "natural" infection and immunization-induced antibodies to the capsular polysaccharide of H. intlueniae type b. J. Immunol. 107:103-110,1971. 9. Anderson, J. P., Johnston, R. B., Smith, D. H. Human serum activities against Haemophilus iniluenzae type b. J. Clin. Invest. 51:31-38, 1972. 10. Robbins, J. B., Parke, J. C., Jr., Schneerson , R., Whisnant, J. W., Robbins, J. B. Quantitative measurement of "natural" and immunization-induced Haemophilus in[luenzae type b capsular antibodies. Pediatr. Res. 7:103-110,1973. II. Smith, D. H., Peter, G., Ingram, D. L., Harding, L., Anderson, P. Responses of children immunized with the capsular polysaccharides of Hemophilus iniluenzae type b. Pediatrics 52:637-644, 1973. 12. Robbins, J. B. Acquistion of "natural" and immunization-induced immunity to Haemophilus injluenzae type b diseases. In D. Schlessinger [ed.], Microbiology -1975. American Society for Microbiology, Washington, D.C., 1975, p. 400-405. 13. Parke, J. C., Jr., Schneerson, R., Robbins, J. B. In S. H. W. Sell [ed.]. Haemophilus iniluenxae. Vanderbilt University Press, Nashville, Tenn., 1973, p. 251-260. 14. Gotschlich, E. C. A simplication of the radioactive antigen-binding test by a double label technique. 1Imrnunol, 107:910-911, 1971. 15. Prince, A. M., Burke, K. Counterimmunoosmoelectrophoresis for the detection of Australia antigen in human sera. Science 169:593-594, 1970. 16. Gotschlich, E. C. Proposal for collaborative studies for the laboratory evaluation of polysaccharide meningococcal vaccine. Bull. W.H.O. BD/CSM/74.l1, 1974. 17. Halperin, M. Extension of the Wilcoxon-Mann Whitney test for samples censored at the same fixed point J. Am. Stat. Assoc. 55: 125-138, 1960. 18. Gold, R., Lepow, M. L., Coldschneider, I., Draper, T. L., Gotschlich, E. C. Clinical evaluation of group
Downloaded from http://jid.oxfordjournals.org/ at University of Bath Library & Learning Centre on July 11, 2015
and developed slowly (incidence in five-year-old H.i.b. vaccinees, 54~o)' However, ~[en C seemed to be immunogenic when used for immunization of persons of all ages [18]. The molecular weight of polysaccharides has been shown to be directly related to their immunogenicity [19, 20]. Thus, the K d value of 0.34 would seem to be a minimal value for future vaccine trials of Men C polysaccharide in infants. No difference in antibodies to H.i.b. between the two vaccine groups was detected during the first three years of life. In contrast to the infrequent presence of natural antibodies to Men C in infants, natural antibodies to H.i.b. were observed in 83% of infants in the first year of life, and this figure rose to 100~o for the fifth year (in lVren C vaccinees). Some lVren C vaceinees had high levels (20 ,ugjml) at age three to five years. These findings are consistent with the observation that organisms that cross-react with H.i.b. are more common than those that crossreact with Men C [21]. There was a higher percentage of four- to six-year-old vaccinees with lower levels of antibody to H.i.b. in the group given ?\len C vaccine than there was in the same age group that received H.i.b. vaccine. A slightly lower rate of meningitis was observed in the H.i.b. vaccinees. Based on our current data, the most likely value for vaccine effectiveness against meningitis during the first year of life is 60% (probable range, zero to 80~o) and is 40% against meningitis and septicemia (probable range, -10% to 70%). However, it is unlikely that the vaccine is >80~o effective against meningitis during the first five years of life, since the conditional probability of observing an outcome of seven cases among the H.i.b. group and 12 cases among the l\;fcnC vaccinees would be P = 0.03 [17]. This study indicated that a more effective H.i.b. vaccine must be prepared for a high degree of protection against serious diseases caused by this organism. The determinations of molecular size and endotoxin content, the antibody data, and the absence of adverse reactions to vaccines used in this study provide a basis for evaluation of new vaccine preparations.
855
S56
A and group C meningococcal polysaccharide vaccines in infants. J. Clin. Invest. 56:1536-1547, 1975. 19. Kabat, E. A., Bezer, A. E. The effect of variation in molecular weight on the antigenicity of dextran in man. Arch. Biochem. Biophys. 78:306-314, 1958. 20. Howard, J. G., Zola, H., Christie, G. R., Courtney, B. M. Studies on immunological paralysis. V. The influence of molecular weight on the immunogenicity, tolerogenicity and antibody-neutralizing activity of type III pneumococcal polysaccharide. Immunology 21:535-546, 1971.
Parke et al.
21. Robbins, J. B., Schneerson, R., Liu, T., Schiffer, M. S. Schiffman, G., Myerowitz, R. L., McCracken, G. H., 0rskov, I., 0rskov, F. Cross-reacting bacterial antigens and immunity to disease caused by encapsulated bacteria. In E. Neter and F. Milgram [ed.], The immune system and infectious diseases. Four International Convocation on Immunology, 1974, Buffalo, N. Y. Karger, Basel, 1975, p. 218-241.
Downloaded from http://jid.oxfordjournals.org/ at University of Bath Library & Learning Centre on July 11, 2015
