PEDIATRICS Vol. 106 No. 5 November 2000, pp. 965-972

Safety and Immunogenicity of Heptavalent Pneumococcal Vaccine Conjugated to CRM₁₉₇ Among Infants With Sickle Cell Disease

Katherine L. O'Brien, MD, MPH^*, , Andrea J. Swift, Jerry A. Winkelstein, MD, Mathuram Santosham, MD, MPH^*, , Beth Stover, RN, Ruth Luddy, MD^§, Joseph E. Gootenberg, MD, Jeffrey T. Nold, DO^¶, Allen Eskenazi, MD^#, Sally J. Snader, RN, BSN, Howard M. Lederman, MD, PhD, and for the Pneumococcal Conjugate Vaccine Study Group^**

From the ^* Johns Hopkins University School of Hygiene and Public Health, Baltimore, Maryland;  Johns Hopkins University School of Medicine, Department of Pediatrics, Baltimore, Maryland; ^§ Sinai Hospital, Baltimore, Maryland;  Georgetown University School of Medicine, Washington, DC; ^¶ Pediatric Center of Annapolis, Annapolis, Maryland; and the ^# University of Maryland School of Medicine, Department of Pediatrics, Baltimore, Maryland.

	ABSTRACT

Top Abstract Methods Results Discussion References

Objectives.  To determine the immunogenicity and safety of heptavalent pneumococcal polysaccharide vaccine (serotypes 4, 6B, 9V, 14, 18C, 19F, and 23F) conjugated to CRM₁₉₇ (7-valent conjugate pneumococcal vaccine [7VPnC]) among infants with sickle cell disease (SCD) and a comparison group of infants without SCD (non-SCD).

Design.  Two cohorts of infants were enrolled and received open-label doses of 7VPnC vaccine; infants enrolled before 2 months of age received 7VPnC vaccine at 2, 4, and 6 months of age followed by 23-valent pneumococcal polysaccharide vaccine (PS-23) at 24 months of age for those infants with SCD (schedule A), and infants enrolled between 2 and 12 months of age received 7VPnC at 12 months of age followed by PS-23 at 24 months of age for infants with SCD (schedule B). Safety data were collected for 3 days after each dose of vaccine. Antibody concentrations were measured to each of the 7VPnC serotypes by enzyme-linked immunosorbent assay before each vaccine dose and 1 month after the last 7VPnC dose and the PS-23 vaccine dose.

Results.  Forty-five infants (34 SCD and 11 non-SCD) were vaccinated according to schedule A and 16 infants (13 SCD and 3 non-SCD) according to schedule B. The 7VPnC vaccine was highly immunogenic for all serotypes among infants with and without SCD who received 3 doses of vaccine according to schedule A: depending on serotype, 89% to 100% achieved antibody concentrations above .15 µg/mL and 56% to 100% achieved antibody concentrations above 1.0 µg/mL. Among infants immunized according to schedule B, a single dose of 7VPnC vaccine resulted in antibody concentrations above .15 µg/mL in 53% to 92% by serotype and above 1.0 µg/mL in 31% to 71% by serotype. A single dose of PS-23 resulted in dramatic increases in the antibody concentrations to all serotypes regardless of 1- or 3-dose priming. There was no difference in the reactogenicity of the 7VPnC vaccine between those with and without SCD. There were no serious reactions to the 7VPnC or PS-23 vaccines, even among those with high antibody concentrations before immunization.

Conclusions.  Infants with SCD respond to 7VPnC vaccine with antibody concentrations that are at least as high as infants without SCD. Infants immunized with 7VPnC vaccine at 2, 4, and 6 months of age developed antibody concentrations in the same range as those achieved among infants without SCD enrolled in a large trial that demonstrated vaccine efficacy against invasive disease. Significant rises were seen in antibody concentrations to all 7VPnC serotypes after the PS-23 booster in children receiving schedule A or B.  Key words:  pneumococcus, sickle cell disease, conjugate vaccine, Prevnar.

Streptococcus pneumoniae (pneumococcus) is a significant cause of morbidity and mortality among infants and children with sickle cell disease (SCD), accounting for as many as 25% of all deaths.^1-7 Without prophylactic measures, the rate of invasive pneumococcal disease among children with SCD is 30 to 100 times greater than that of otherwise healthy children. Children under the age of 2 years are at highest risk with attack rates of 6.5 to 27.3 per 100 child years in the prepenicillin prophylaxis era. Prevention strategies which include administration of 23-valent polysaccharide vaccine (PS-23) at 2 years of age and penicillin prophylaxis reduce the risk, but invasive pneumococcal disease rates among SCD children younger than 5 years of age continue to be ~10-fold higher than rates among similarly aged children without SCD.⁸

A variety of defects in host defense have been identified in children with SCD. These include both functional/anatomic asplenia⁹ and decreased serum opsonizing activity,^10,11 each of which contributes to their increased susceptibility to systemic pneumococcal infections.¹² Both experimental and clinical evidence suggest that the defects in host defense against the pneumococcus seen in children with SCD can be overcome by active or passive immunization to the pneumococcus.^11,13,14 Immunization of children with SCD over the age of 2 years with pneumococcal capsular polysaccharide induces the formation of specific anticapsular antibodies and protects them against some systemic pneumococcal infections.¹⁴ The effectiveness of currently licensed pneumococcal polysaccharide vaccines is limited by poor immunogenicity for important serotypes, particularly among children younger than 2 years of age.

Investigational and recently licensed pneumococcal vaccines that covalently conjugate serotype-specific polysaccharides or oligosaccharides to protein carriers are immunogenic among young infants without SCD and among older children and adults with SCD.^15,16 Recently one of these vaccines has been shown to be efficacious in preventing invasive pneumococcal disease among infants less than 2 years of age and has just been licensed under the trade name Prevnar (Wyeth Lerderle Vaccines [WLV], West Henrietta, NY); February 17, 2000) by the US Food and Drug Administration for use among children through 9 years of age.¹⁷ To date no studies have evaluated the immunologic response of infants with SCD to this 7-valent conjugate pneumococcal vaccine. Therefore, we evaluated the antibody response of infants with and without SCD to this 7-valent pneumococcal protein conjugate vaccine to predict the likelihood that the vaccine will be effective at protecting these most vulnerable infants against the morbidity and mortality of invasive pneumococcal disease.

	METHODS

Top Abstract Methods Results Discussion References

Subjects

All infants born in the state of Maryland are tested at birth for hemoglobinopathies. We recruited infants diagnosed to have hemoglobin SS, SC, and S- thalassemia to participate in the immunogenicity trial during visits to their hematologists and through the Maryland State Health Department. Hemoglobin status was confirmed by hemoglobin electrophoresis at each medical center or at the Maryland State Health Department. Race and age-matched children without SCD were recruited from primary care practices in Maryland. The parents of all participating infants provided written informed consent for participation in the study.

Safety Data

Parents were asked to record the temperature, demeanor, and signs or symptoms of any side effects on the day of and for the 3 days after vaccination on a log sheet provided to them. The log sheet was then mailed to the investigators. If a log sheet was not received, active telephone follow-up was conducted. All hospitalizations during the course of the study were documented.

Vaccines

A 7-valent pneumococcal vaccine (7VPnC, WLV) containing polysaccharides (serotypes 4, 6B, 9V, 14, 19F, and 23F) and oligosaccharide (serotype 18C) covalently linked to a cross-reacting diphtheria variant (CRM₁₉₇) protein carrier was used to immunize infants. Each 0.5-mL dose contains 2 µg of saccharide of serotypes 4, 9V, 14, 19F, and 23F and 4 µg of serotype 6B, ~20 µg of CRM₁₉₇ protein, and ~0.5 mg of aluminum phosphate. We immunized infants with SCD and without SCD according to 1 of 2 schedules, A or B, depending on their age at enrollment. Schedule A consisted of 3 doses of the 7VPnC vaccine at 2 (range: 6-12 weeks of age), 4 (range: 12-20 weeks of age), and 6 (range: 20-28 weeks of age) months of age. The minimum interval between doses was 4 weeks and the maximum interval was 12 weeks. Schedule B consisted of a single dose of 7VPnC at 12 months of age (range: 11-15 months of age). This schedule was chosen over a 2-dose 7VPnC regimen because of its simplicity and because there were no existing data to guide the choice of a 1-dose or 2-dose regimen in this age group. Infants with SCD in both schedules A and B also received the 23-valent pneumococcal polysaccharide vaccine PS-23 (Merck, West Point, PA or WLV) at 24 months of age (range: 22-26 months of age). All infants received other routine vaccinations at the discretion of their primary care practitioner.

Sera Collection

We collected a venous serum specimen from each subject. For infants receiving vaccine according to schedule A, serum samples were collected before each 7VPnC vaccine dose, 1 month after the last dose of 7VPnC vaccine, at 12 months of age, 24 months of age, and 1 month after the PS-23 vaccine for those who received it. For infants receiving vaccine according to schedule B, serum samples were collected before and after the 7VPnC vaccine dose, at 24 months of age, and 1 month after the PS-23 vaccine for those who received it. Blood specimens were collected and stored on ice while being transported to the laboratory where sera were separated and frozen at 70°C within 2 hours of collection.

IgG Antibody Quantitation

Type-specific antipneumococcal antibody levels were measured at Johns Hopkins University using an enzyme-linked immunosorbent assay, modified from the method of Koskela.¹⁸ The enzyme-linked immunosorbent assay (ELISA) methodology followed that used at WLV. Before initiation of the study, sera were run in duplicate at Johns Hopkins University and WLV to ensure that the assay was standardized. All sera were premixed at a 1:50 dilution with 2.0 µg/mL of pneumococcal C-polysaccharide to minimize nonspecific binding. For each serotype, eight 2.5-fold dilutions were prepared, and each dilution was added to duplicate wells of a microtiter-plated precoated with the type-specific polysaccharide. In parallel, dilutions of the US Food and Drug Administration pneumococcal reference serum lot 89-SF¹⁹ were added to microtiter wells. Plates were incubated overnight, washed, and then developed after addition of goat antihuman immunoglobulin G alkaline phosphatase-labeled antibody. Comparison of patient and reference sera were performed by using a 4-parameter logistic-log function to form standard curves.²⁰

Statistical Analyses

We evaluated the response to vaccines by calculating geometric mean concentrations (GMCs) of serum antibodies and their 95% confidence intervals by serotype and time after vaccination in each of the vaccine groups (Excel 97, Microsoft Corp, Redmond, WA). We compared the response to vaccine within and between groups using paired and nonpaired t tests, using the logs of the antibody concentrations (Stata, Version 6.0, Stata Corp, College Station, TX). We conducted all analyses using a 2-sided P value and a type I error () of <.05. We compared the reaction to vaccination at each dose of vaccine and between groups using the ² test or Fisher's exact test as appropriate (EpiInfo, Version 6.03, Centers for Disease Control and Prevention, Atlanta, GA).

Ethical Approval

The study was approved by the institutional review boards of the Johns Hopkins Medical Institutions, the Maryland State Department of Health, the University of Maryland Medical School, and Sinai Hospital.

	RESULTS

Top Abstract Methods Results Discussion References

Study Subjects

Subjects were enrolled between January 1995 and July 1997. Forty-six 2-month-old infants were enrolled in schedule A, the 3-dose primary series group (35 SCD and 11 non-SCD), and sixteen 12-month-old children were enrolled in schedule B, the 1-dose primary series group (13 SCD and 3 non-SCD). All children were born at term except a single child in the schedule B SCD group who was born at 33.5 weeks' gestation. One of the 2-month-old SCD subjects disenrolled from the study after the first vaccine dose because retesting revealed the child did not have SCD. The distribution of infants who remained in the study by vaccine group, sex, race, and hemoglobin are shown in Table 1. The number of vaccine doses administered at each period and the mean age at each vaccination also are shown in Table 1. Thirty-three infants with SCD completed a 3-dose primary series and 27 of these infants also received the PS-23 vaccine. One 2-month-old child with SCD received only 1 dose of 7VPnC vaccine before being lost to follow-up and is not included in the analysis of antibody response.

Seventeen children with SCD had 51 hospitalizations between the time of enrollment and 25 months of age. Only 1 child without SCD was hospitalized. Reasons for hospitalization were pain crises/dactylitis (7), fever/rule-out sepsis (30), respiratory syncytial virus bronchiolitis (4), central line infection (2), pneumonia/bronchitis (2), gastroenteritis (1), rule-out osteomyelitis (1), splenic sequestration (1), transfusion reaction (1), priapism (1), and surgery (1). No study subject had pneumococcus isolated from a normally sterile site.

Loss to Follow-Up

Individuals who did not complete the vaccination series or blood drawing schedule had a variety of reasons for noncompletion of the study including: moved out of state or country (4), lost to follow-up with no response to telephone or written reminders (4), received PS-23 off schedule because of emergent splenectomy (2), and parent returned to full-time schooling and was unable to bring the child to study clinic (1).

Antibody Responses

The serotype-specific GMCs and their 95% confidence intervals for selected time points among infants who received schedule A are shown in Table 2 and for those who received schedule B are shown in Table 3.

Schedule A (SCD) Compared with prevaccination sera, there were significant increases in serum antibody concentrations after the first, second, and third doses of 7VPnC for all serotypes. These antibody concentrations remained significantly different from those at 2 months of age through 12 months of age for all serotypes, except type 14. By 24 months of age the serum antibody concentrations had fallen significantly from those achieved by the primary 3-dose series for all serotypes but were higher than the preimmunization concentrations. As expected there were large booster responses to all serotypes 1 month after administration of the PS-23 dose at 24 months of age. Antibody concentrations to all 7 serotypes in a single subject were several-fold higher than were those among all other infants at 4 months of age. Analysis of the GMC without these data points did not result in a significantly different GMC.

Schedule A (SCD Verus Non-SCD) At 2 months of age, the serum antibody concentrations of the infants without SCD were significantly higher than were those of the children with SCD for all serotypes except types 14 and 18C. However, after the third dose of 7VPnC, the concentrations were similar between groups for all serotypes except 18C and 19F, which were higher among those with SCD. Finally, at 12 months of age and at 24 months of age declines in antibody concentrations occurred among both those with and without SCD for all serotypes. These declines were similar between groups for all serotypes except type 14 at 12 months of age.

Schedule B (SCD) Among infants with SCD immunized at 12 months of age there was a significant increase in antibody concentration to all serotypes after a single dose of 7VPnC vaccine. These antibodies declined significantly by 24 months of age for serotypes 9V, 14, and 18C only. A subsequent dose of PS-23 at 24 months of age resulted, as expected, in a booster response to all serotypes.

Schedule B (SCD Versus Non-SCD) There were no significant differences in the antibody concentrations for any serotypes at any of the 3 time points (12, 13, and 24 months of age) between those with and without SCD.

Schedule A Versus Schedule B (SCD) As expected, the prevaccination serum antibody concentrations were higher at 2 months of age than at 12 months of age for all serotypes; however, these differences were only statistically significant for serotypes 4, 6B, and 23F. Schedule A infants with SCD produced a greater antibody level to their 3 dose primary series than did schedule B infants with SCD to their 1-dose primary series for serotypes 6B, 14, 19F, and 23F. At 12 months of age, the infants with SCD who had received 3 doses of 7VPnC vaccine had significantly higher antibody concentrations for all serotypes than did those infants with SCD who had not yet received any doses of pneumococcal vaccine. By 24 months of age, there were no significant differences in antibody concentrations between the groups except infants immunized with schedule A had higher antibody concentrations to 6B than those immunized with schedule B. After the booster dose of PS-23, the SCD infants who had received a single dose of 7VPnC vaccine had greater antibody concentrations to serotypes 4, 9V, and 14 than did those infants who had received the 3-dose series.

Absolute Height of Antibody Concentration The proportion of subjects in schedules A and B who achieved an antibody concentration .15 µg/mL and 1.0 µg/mL after their primary series of 7VPnC vaccine is illustrated in Fig 1. One month after the 3-dose primary series, the proportion of infants who achieved an antibody concentration .15 µg/mL or 1.0 µg/mL ranged from 89% to 100% and from 56% to 100%, respectively, by serotype. Similarly, among children who received a single dose of 7VPnC vaccine at 12 months of age the proportion who achieved an antibody concentration .15 µg/mL or 1.0 µg/mL 1 month after the vaccine ranged from 53% to 92% and from 31% to 71%, respectively, by serotype. There were no significant differences in the proportion that achieved these concentrations by sickle cell status. By 24 months of age, the proportion with antibody concentrations .15 µg/mL or 1.0 µg/mL had fallen to 80% to 100% and 16% to 80%, respectively, by serotype; there were no differences in the proportions by sickle cell status or by schedule of 7VPnC vaccine. These data are illustrated in Fig 2. After the PS-23 vaccine, 32 of 33 immunized children achieved antibody concentrations 1.0 µg/mL for all serotypes.

View larger version (47K): [in this window] [in a new window]   Fig. 1.   Proportion of sickle cell subjects with a defined serotype-specific antibody response 1 month after the primary 7VPnC vaccine series. Black bars (schedule A, response  .15 µg/mL), diagonal bars (schedule B, response  .15 µg/mL), white bars (schedule A, response  1.0 µg/mL), dotted bars (schedule B, response  1.0 µg/mL).

View larger version (42K): [in this window] [in a new window]   Fig. 2.   Proportion of sickle cell subjects with a defined serotype-specific antibody response before the PS-23 booster vaccine at 24 months of age. Black bars (schedule A, response  .15 µg/mL), diagonal bars (schedule B, response  .15 µg/mL), white bars (schedule A, response  1.0 µg/mL), dotted bars (schedule B, response  1.0 µg/mL).

Antibody Kinetics We evaluated the serotype-specific antibody response after each dose of the 3-dose primary series of 7VPnC vaccine. Although the antibody response varied by serotype, 3 general patterns of response could be distinguished. There was a vigorous antibody response to the first dose of vaccine for serotype 4; this antibody response did not increase dramatically with the subsequent doses of the primary series. For all other serotypes, the antibody response to the first dose of vaccine was marginal, and progressive increases were seen with each subsequent dose in the primary series. This pattern was most clearly seen with serotypes 6B and 23F. These patterns of antibody response were similar for those with and without SCD. Representative antibody kinetic graphs are shown in Fig 3.

View larger version (8K): [in this window] [in a new window]   Fig. 3.   Kinetics of antibody response to 7VPnC vaccine among infants with SCD immunized according to schedule A (serotype 4, solid line with squares; serotype 23F, dashed line with circles).

Vaccine Safety Overall 84% of diary cards were returned after doses of vaccine. The proportion of subjects reporting any local reaction (redness, pain, or lump) was not different between those with and without SCD for doses in the 3-dose primary series, schedule A (overall primary series 52% vs 57%, respectively). Among infants with SCD who received the PS-23 vaccine, there was a slight increase in reactogenicity among those who had received a 3-dose primary series, compared with a 1-dose primary series, but it was not statistically significant (58% vs 50%). Overall 15% (range: 4%-43%) reported a moderate local reaction (redness or lump larger than the size of a quarter, or decreased movement of limb). The proportion of subjects who reported a fever (rectal temperature >38°C or >100.4°F) was 9.5% (range: 0%-33%). Infants with SCD were no more likely than were those without SCD to report fever (7% vs 19%, Fisher's exact P = .12). One child experienced a single episode of high fever (>103.3°F) after a single dose of 7VPnC vaccine administered at 12 months of age. The rate of nonspecific reactions (increased fussiness, sleeping, or decreased appetite) was 56% and was similar between those infants with and without SCD (54% vs 65%; P = .30). There were no trends for increasing reactogenicity with each subsequent primary dose among those with or without SCD (² for linear trend, P = .79).

	DISCUSSION

Top Abstract Methods Results Discussion References

The design of this study allows for comparisons of immune response to 7VPnC vaccine between infants with and without SCD. It also permits evaluation of the response of children with SCD to a PS-23 booster dose at 24 months of age after either a 3-dose primary series of 7VPnC beginning at 2 months of age or a single dose of 7VPnC at 12 months of age.

The 7VPnC vaccine was generally safe and immunogenic among infants with SCD. The frequency of local reactions was similar to those identified among a large group of children without SCD.¹⁶ Because some studies have raised the concern that more serious reactions can result from repeated vaccination with pneumococcal polysaccharide vaccine, we wanted to determine the reactogenicity of PS-23 after a primary series of pneumococcal conjugate vaccine. There were no serious local reactions in response to the PS-23 booster dose, and the observed minor and moderate reactions to the booster were no more frequent than those seen with the primary series of 7VPnC vaccine.

Infants with SCD who were immunized with a 3-dose series of 7VPnC vaccine have significant rises in their antibody concentrations against all serotypes in response to this primary series. These responses are comparable to those among infants without SCD for most of the serotypes. The decline in these concentrations by 24 months of age is also comparable between those with and without SCD. Similarly, among infants who are immunized with a single dose of 7VPnC at 12 months of age, there were no significant differences in antibody level between those with and without SCD. There is no evidence from this study that infants with SCD have an altered immune response to the 7VPnC vaccine as measured by antibody concentrations. The opsonic activity of the antibodies against 6B and 14 from the infants with SCD immunized with schedules A and B have been evaluated and were found to correlate with antibody concentration.²¹ Whether these antibody concentrations are produced in a sufficiently high concentration and with sufficient functional activity to protect these infants from invasive pneumococcal disease is yet to be determined.

Our analysis of the effect of preexisting antibody concentrations on response to the vaccine was limited by the number of subjects with elevated concentrations at the 2-month visit. For serotype 14 we did demonstrate that high maternal antibody concentrations may indeed dampen the immune response to the vaccine, although the GMCs in both groups were above an absolute value of 1.0 µg/mL for virtually all time points.

The antibody concentrations for all serotypes at 12 months of age after a 3-dose primary series were higher than for those at 12 months of age among infants who had not yet received any vaccines. Thus, although antibody concentrations certainly decline over time after a primary 3-dose series, they do not decline to levels that exist among unimmunized children of the same age.

The response to a single dose of vaccine did not vary remarkably by age at which the dose was administered. Two-month-old infants with SCD had a response to a single dose of vaccine that was equivalent to that of a single dose administered to 12-month-old infants with SCD for most serotypes. The equivalence of responses suggests that there is no immune dysfunction in the first year of life significant enough to impair the response to vaccination. This is not entirely unexpected because the spleen, which becomes progressively dysfunctional and ultimately autoinfarcts, is unlikely to play a significant role in the immune response to an intramuscular vaccine injection, which is absorbed from muscle and circulates to the regional lymph nodes. Importantly, the response to a single dose of vaccine at 12 months of age was lower for 4 of the 7 serotypes than the response to a primary 3-dose series begun at 2 months of age. This suggests that there is little antibody response to environmental exposure during the first year of life in infants receiving penicillin prophylaxis and that 1-year-olds may benefit from a 2- or 3-dose primary series of immunization.

The booster dose of PS-23 resulted in dramatic increases in antibody concentrations for all serotypes regardless of which primary series had been administered. In fact for some serotypes the response was greater among those infants who had received a single 7VPnC dose at 12 months of age than among those infants who had received a 3-dose primary series beginning at 2 months of age.

One other published study has evaluated the use of 7VPnC among individuals with SCD between the ages of 4 and 30 years.¹⁵ In that study, 11 individuals with SCD received 2 doses of 7VPnC and a dose of PS-23 each separated by 2 months. A comparison group of 12 individuals with SCD were immunized with a single dose of PS-23. The GMCs of antibody after the PS-23 vaccine were significantly higher among the group primed with 7VPnC only for serotypes 14 and 19F. Of particular note, the GMCs we achieved after the PS-23 booster among children in either schedule A or schedule B were 2- to 12-fold higher (except for schedule A serotype 14) than those achieved by the children and adults primed according to the schedule above. Several factors may contribute to the differences in response to the PS-23 booster dose in these 2 studies. First, it is likely that an interval of >2 months between the priming series and the PS-23 vaccine dose is required to achieve maximum boosting. Second, infants may respond with greater intensity and immune diversity to the 7VPnC vaccine, compared with older children and adults. Finally, the ELISA assays were conducted in separate laboratories, which were not cross-standardized; therefore, some of these differences may be a laboratory phenomena.

We observed no cases of invasive pneumococcal disease among our study participants. Based on a total of 87.3 person years of observation and rates of invasive pneumococcal disease among infants with SCD of 2.3/100 person years, we would have expected to see 2 cases of disease. Although this sample size is too small to draw conclusions about the efficacy of the vaccine among SCD infants, this observation is encouraging.

This study demonstrates that the 7VPnC vaccine is equally immunogenic among infants with and without SCD. There are data demonstrating that infants with SCD have strong immune responses to Haemophilus influenzae type b (Hib) protein conjugate vaccines.^22-24 The serotype-specific antibody concentrations that will protect normal children from invasive and respiratory tract pneumococcal disease are not firmly established. Although an antibody concentration of .15 µg/mL has not been shown to be protective against pneumococcal infections, we have selected it as a midrange value of antibody concentration shown to protect mice against invasive pneumococcal disease in an in vivo assay. It is also the concentration of antibody that was achieved by virtually 100% of subjects for all serotypes in a large immunogenicity trial of 7VPnC vaccine among normal infants.¹⁶ Finally, it is the value shown in some reports to provide short-term protection against Hib disease. Clearly, Hib conjugate vaccination and disease are different entities than pneumococcal disease and vaccination, but this provides a starting point for evaluation and comparisons of antibody concentrations achieved following vaccination. Of great importance is the demonstration that a pneumococcal protein conjugate vaccine induces a boostable immune response among infants with SCD just as among infants without SCD. Although absolute antibody concentrations induced among children with SCD may not achieve protective levels after 1 or even 2 doses for some serotypes, the T cell-dependent response indicates that they have the ability to rapidly respond to subsequent challenges with pneumococcal polysaccharide exposures.

The antibody concentrations required to prevent disease among infants with SCD may be higher than those required among infants without SCD because of the immune system defects found among those with SCD.^3,9-11,25-27 Until the antibody concentrations, opsonophagocytic activity, or other measures of the immune response required for protection from disease are established, we can only evaluate pneumococcal vaccines among infants with SCD by demonstrating that they are equally responsive to these products compared with infants without SCD and infer that they are likely to be protective.

The Advisory Committee on Immunization Practices of the Centers for Disease Control and Prevention has recommended Prevnar (WLV) for use among all children with SCD up to 5 years of age, although immunogenicity data are very limited among those older than 12 months of age. There are currently no recommendations for its use among older children and adults with SCD. Postlicensure studies to evaluate the many unanswered questions are needed. These studies should include, among others, assessment of the immune response and safety of 7VPnC vaccine among children who have already received PS-23 vaccine, surveillance for invasive disease among children with SCD to provide estimates of vaccine efficacy, and studies of antibody persistence and memory to determine the optimum timing and intervals for PS-23 vaccine booster doses among children with SCD. Other areas of research that are needed include further evaluation of various vaccine regimens of 7VPnC and PS-23 among older children and adults with SCD.

	ACKNOWLEDGMENTS

Financial support for this study was provided by the Thrasher Fund, Wyeth Lederle Vaccines, and the Thomas Wilson Sanatorium.

We thank Dr S. Panny, Kellie Hall, and Sonya Ross of the Maryland State Health Department for their assistance in recruiting patients for the study.

	FOOTNOTES

^** Other participants in the Pneumococcal Conjugate Vaccine Study Group were Dr George J. Dover, Dr Archie Golden (Johns Hopkins University, Baltimore, MD); Dr Elizabeth H. Lasley (Pediatric Center of Annapolis, Annapolis, MD); Joan Marasciulo (Sinai Hospital, Baltimore, MD); Kristen Sawyer (University of Maryland Medical Center, Baltimore, MD); Dr Allen Kimura, Dr Frank Malinoski, Michelle Mineo-Kuhn, and Catherine Tuffey (Wyeth-Lederle Vaccines, Pearl River, NJ).

Received for publication Mar 20, 2000; accepted Mar 23, 2000.

Address correspondence to Katherine L. O'Brien, MD, MPH, Johns Hopkins University School of Hygiene and Public Health, 621 N Washington St, Baltimore, MD, 21205. E-mail: klobrien{at}jhsph.edu

	ABBREVIATIONS

SCD, sickle cell disease; PS-23, 23-valent polysaccharide pneumococcal vaccine; 7VPnC, 7-valent conjugate pneumococcal vaccine; WLV, Wyeth Lederle Vaccines; ELISA, enzyme-linked immunosorbent assay; GMC, geometric mean concentration; Hib, Haemophilus influenzae type b.

	REFERENCES

Top Abstract Methods Results Discussion References

1. Gaston MH, Verter JI, Woods G, Prophylaxis with oral penicillin in children with sickle cell anemia. N Engl J Med 1986; 314:1593-1599 [Abstract]
2. Wong W-Y, Powars DR, Chan L, Hiti A, Johnson C, Overturf GD Polysaccharide encapsulated bacterial infection in sickle cell anemia: a thirty year epidemiologic experience. Am J Hematol 1992; 39:176-182 [Medline]
3. Wong W-Y, Overturf GD, Powers DR Infection caused by Streptococcus pneumoniae in children with sickle cell disease: epidemiology, immunologic mechanisms, prophylaxis and vaccination. Clin Infect Dis 1992; 14:1124-1136 [Medline]
4. Leikin SL, Gallagher D, Kinney TR, Sloane D, Klug P, Rida W Mortality in children and adolescents with sickle cell disease. Pediatrics 1989; 84:500-508 [Abstract/Free Full Text]
5. Centers for Disease Control and Prevention Mortality among children with sickle cell disease identified by newborn screening during 1990-1994California, Illinois, and New York. MMWR Morb Mortal Wkly Rep 1998; 47:169-172 [Medline]
6. Davis H, Schoendorf KC, Gerone PJ, Moore RM National trends in mortality of children with sickle cell disease, 1968 through 1992. Am J Public Health 1997; 87:1317-1322 [Abstract/Free Full Text]
7. Pegelow CH, Armstrong RD, Light S, Toledano SR, Davis J Experience with the use of prophylactic penicillin in children with sickle cell anemia. J Pediatr 1991; 118:736-738 [CrossRef][Medline]
8. Adamkiewicz TV, Buchanan GR, Facklam RR, Iyer RB, O'Brien KL, Pegelow C. Streptococcus pneumoniae infections in children with sickle cell disease. Presented at the First International Symposium on Pneumococci and Pneumococcal Disease; June 13-17, 1998; Elsinore, Denmark
9. Pearson HA, Spencer RP, Cornelius EA Functional asplenia in sickle-cell anemia. N Engl J Med 1969; 281:923-926
10. Winkelstein JA, Drachman RH Deficiency of pneumococcal serum opsonizing activity in sickle-cell disease. N Engl J Med 1968; 279:459-466
11. Johnston RB, Newman SL, Struth AG An abnormality of the alternate pathway of complement activation in sickle-cell disease. N Engl J Med 1973; 288:803-808
12. Winkelstein JA Pneumococcal infections in sickle cell disease. J Pediatr 1977; 91:521-522 [CrossRef][Medline]
13. Levine ML Increased susceptibility of splenectomized rats to infection with Diplococcus pneumoniae. J Infect Dis 1972; 126:507-513 [Medline]
14. Ammann AJ, Addiego J, Wara DW, Lubin B, Smith WB, Mentzer WC Polyvalent pneumococcal-polysaccharide immunization of patients with sickle-cell anemia and patients with splenectomy. N Engl J Med 1977; 297:897-900 [Abstract]
15. Vernacchio L, Neufeld EJ, MacDonald K, Combined schedule of 7-valent pneumococcal conjugate vaccine followed by 23-valent pneumococcal vaccine in children and young adults with sickle cell disease. J Pediatr 1998; 133:275-278 [CrossRef][Medline]
16. Rennels MB, Edwards KM, Keyserling H, Safety and immunogenicity of heptavalent pneumococcal vaccine conjugated to CRM197 in United States infants. Pediatrics 1998; 101:604-611 [Abstract/Free Full Text]
17. Black S, Shinefield H, Ray P, et al. Efficacy of heptavalent conjugate pneumococcal vaccine (Wyeth-Lederle) in 37,000 infants and children: results of the Northern California Kaiser Permanente efficacy trial. Presented at the 38th Interscience Conference on Antimicrobial Agents and Chemotherapy, American Society of Microbiology; September 24-27, 1998; San Diego, CA
18. Koskela M Serum antibodies to pneumococcal C polysaccharide in children: response to acute pneumococcal otitis media or to vaccination. Pediatr Infect Dis J 1987; 6:519-526 [Medline]
19. Quartaert SA, Kirch CS, Wiedl LJ, Assignment of weight-based antibody units to a human antipneumococcal standard reference serum, lot 89-S. Clin Diagn Lab Immunol 1995; 2:590-597 [Abstract]
20. Plikaytis BD, Carlone GM, Maslanka SE, Gheesling LL, Holder PF. Program ELISA User's Manual. Atlanta, GA: Centers for Disease Control and Prevention; 1993;118:917-919
21. Nowak-Wegrzyn A, Winkelstein JA, Swift AJ, Lederman HM, and the Pneumococcal Conjugate Vaccine Study Group Serum opsonic activity in infants with sickle cell disease immunized with pneumococcal polysaccharide protein-conjugate vaccine. Clin Diag Lab Immunol 2000; 7:788-793 [Abstract/Free Full Text]
22. Gigliotti F, Feldman S, Wang WC, Day SW, Brunson G Serologic follow-up of children with sickle cell disease immunized with a Haemophilus influenzae type b conjugate vaccine during early infancy. J Pediatr 1991; 118:917-919 [CrossRef][Medline]
23. Gigliotti F, Feldman S, Wang WC, Day SW, Brunson G Immunization of young infants with sickle cell disease with a Haemophilus influenzae type b saccharide-diphtheria CRM197 protein conjugate vaccine. J Pediatr 1989; 114:1006-1010 [CrossRef][Medline]
24. Frank AL, Labotka RJ, Rao S, Haemophilus influenzae type b immunization of children with sickle cell disease. Pediatrics 1988; 82:571-575 [Abstract/Free Full Text]
25. Bjornson AB, Lobel JS Direct evidence that decreased serum opsonization of Streptococcus pneumoniae via the alternative pathway in sickle cell disease is related to antibody deficiency. J Clin Invest 1987; 79:388-398
26. Bjornson AB, Lobel JS Lack of a requirement for the Fc region of IgG in restoring pneumococcal opsonization via the alternative complement pathway in sickle cell disease. J Infect Dis 1986; 154:760-769 [Medline]
27. De Ceulaer K, Forbes M, Maude GH, Pagliucca A, Sergeant GR Complement and immunoglobulin levels in early childhood in homozygous sickle cell disease. J Clin Lab Immunol 1986; 21:37-41 [Medline]