S. pneumoniae is an important cause of morbidity and mortality in persons of all ages worldwide. The organism causes invasive infections, such as bacteremia and meningitis, as well as pneumonia and upper respiratory tract infections including otitis media and sinusitis. In children older than 1 month, S. pneumoniae is the most common cause of invasive disease. 1 Data from community-based studies performed between 1986 and 1995, indicate that the overall annual incidence of invasive pneumococcal disease in the United States (US) is an estimated 10 to 30 cases per 100,000 persons, with the highest risk in children aged less than or equal to 2 years of age (140 to 160 cases per 100,000 persons). 2,3 Children in group child care have an increased risk for invasive pneumococcal disease. 4,5 Immunocompromised individuals with neutropenia, asplenia, sickle cell disease, disorders of complement and humoral immunity, human immunodeficiency virus (HIV) infections or chronic underlying disease are also at increased risk for invasive pneumococcal disease. 5 S. pneumoniae is the most common cause of bacterial meningitis in the US. 1 The annual incidence of pneumococcal meningitis in children between 1 to 23 months of age is approximately 7 cases per 100,000 persons. 1 Pneumococcal meningitis in childhood has been associated with 8% mortality and may result in neurological sequelae (25%) and hearing loss (32%) in survivors. 6
Acute otitis media (AOM) is a common childhood disease, with more than 60% of children experiencing an episode by one year of age, and more than 90% of children experiencing an episode by age 5. Prior to the US introduction of Prevnar® in the year 2000, approximately 24.5 million ambulatory care visits and 490,000 procedures for myringotomy with tube placement were attributed to otitis media annually. 7,8 The peak incidence of AOM is 6 to 18 months of age. 9 Otitis media is less common, but occurs, in older children. In a 1990 surveillance by the Centers for Disease Control and Prevention (CDC), otitis media was the most common principal illness diagnosis in children 2-10 years of age. 10 Complications of AOM include persistent middle ear effusion, chronic otitis media, transient hearing loss, or speech delays and, if left untreated, may lead to more serious diseases such as mastoiditis and meningitis. S. pneumoniae is an important cause of AOM. It is the bacterial pathogen most commonly isolated from middle ear fluid, identified in 20% to 40% of middle ear fluid cultures in AOM. 11,12 Pneumococcal otitis media is associated with higher rates of fever, and is less likely to resolve spontaneously than AOM due to either nontypeable H. influenzae or M. catarrhalis . 13,14 Prior to the introduction of Prevnar®, the seven serotypes contained in the vaccine accounted for approximately 60% of AOM due to S. pneumoniae (12%-24% of all AOM). 15
The exact contribution of S. pneumoniae to childhood pneumonia is unknown, as it is often not possible to identify the causative organisms. In studies of children less than 5 years of age with community-acquired pneumonia, where diagnosis was attempted using serological methods, antigen testing, or culture data, 30% of cases were classified as bacterial pneumonia, and 70% of these (21% of total community-acquired pneumonia) were found to be due to S. pneumoniae . 16
In the past decade the proportion of S. pneumoniae isolates resistant to antibiotics has been on the rise in the US and worldwide. In a multi-center US surveillance study, the prevalence of penicillin and cephalosporin-nonsusceptible (intermediate or high level resistance) invasive disease isolates from children was 21% (range <5% to 38% among centers), and 9.3% (range 0%-18%), respectively. Over the 3-year surveillance period (1993-1996), there was a 50% increase in penicillin-nonsusceptible S. pneumoniae (PNSP) strains and a three-fold rise in cephalosporin-nonsusceptible strains. 5 Although generally less common than PNSP, pneumococci resistant to macrolides and trimethoprim-sulfamethoxazole have also been observed. Day care attendance, a history of ear infection, and a recent history of antibiotic exposure, have also been associated with invasive infections with PNSP in children 2 months to 59 months of age. 4,5 There has been no difference in mortality associated with PNSP strains. 5,6 However, the American Academy of Pediatrics (AAP) revised the antibiotic treatment guidelines in 1997 in response to the increased prevalence of antibiotic-resistant pneumococci. 17
Approximately 90 serotypes of S. pneumoniae have been identified based on antigenic differences in their capsular polysaccharides. The distribution of serotypes responsible for disease differ with age and geographic location. 18
Serotypes 4, 6B, 9V, 14, 18C, 19F, and 23F have been responsible for approximately 80% of invasive pneumococcal disease in children <6 years of age in the US. 15 These 7 serotypes also accounted for 74% of PNSP and 100% of pneumococci with high level penicillin resistance isolated from children <6 years with invasive disease during a 1993-1994 surveillance by the CDC. 19
Results of Clinical Evaluations
Efficacy Against Invasive Disease
Efficacy was assessed in a randomized, double-blinded clinical trial in a multiethnic population at Northern California Kaiser Permanente (NCKP) from October 1995 through August 20, 1998, in which 37,816 infants were randomized to receive either Prevnar® or a control vaccine (an investigational meningococcal group C conjugate vaccine [MnCC]) at 2, 4, 6, and 12-15 months of age. Prevnar® was administered to 18,906 children and the control vaccine to 18,910 children. Routinely recommended vaccines were also administered which changed during the trial to reflect changing AAP and Advisory Committee on Immunization Practices (ACIP) recommendations. A planned interim analysis was performed upon accrual of 17 cases of invasive disease due to vaccine-type S. pneumoniae (August 1998). Ancillary endpoints for evaluation of efficacy against pneumococcal disease were also assessed in this trial.
Invasive disease was defined as isolation and identification of S. pneumoniae from normally sterile body sites in children presenting with an acute illness consistent with pneumococcal disease. Weekly surveillance of listings of cultures from the NCKP Regional Microbiology database was conducted to assure ascertainment of all cases. The primary endpoint was efficacy against invasive pneumococcal disease due to vaccine serotypes. The per protocol analysis of the primary endpoint included cases which occurred ≥14 days after the third dose. The intent-to-treat (ITT) analysis included all cases of invasive pneumococcal disease due to vaccine serotypes in children who received at least one dose of vaccine. Secondary analyses of efficacy against all invasive pneumococcal disease, regardless of serotype, were also performed according to these same per protocol and ITT definitions. Results of these analyses are presented in Table 1.
TABLE 1 Efficacy of Prevnar® Against Invasive Disease Due to S. pneumoniae in Cases Accrued From October 15, 1995 Through August 20, 1998 20,21
| ||Prevnar® |
Number of Cases
Number of Cases
|Efficacy ||95% CI |
| Vaccine serotypes || || || || |
| Per protocol || 0 || 17 || 100% || 75.4, 100 |
| Intent-to-treat || 0 || 22 || 100% || 81.7, 100 |
|All pneumococcal serotypes || || || || |
| Per protocol ||2 ||20 ||90.0% ||58.3, 98.9 |
| Intent-to-treat ||3 ||27† ||88.9% ||63.8, 97.9 |
|* Investigational meningococcal group C conjugate vaccine (MnCC). |
† Includes one case in an immunocompromised subject.
All 22 cases of invasive disease due to vaccine serotype strains in the ITT population were bacteremic. In addition, the following diagnoses were also reported: meningitis (2), pneumonia (2), and cellulitis (1).
Data accumulated through an extended follow-up period to April 20, 1999, resulted in a similar efficacy estimate (Per protocol: 1 case in Pneumococcal 7-valent Conjugate Vaccine (Diphtheria CRM197 Protein), Prevnar® group, 39 cases in control group; ITT: 3 cases in Prevnar® group, 49 cases in the control group). 21
Efficacy Against Otitis Media
The efficacy of Prevnar® against otitis media was assessed in two clinical trials: a trial in Finnish infants at the National Public Health Institute and the invasive disease efficacy trial in US infants at Northern California Kaiser Permanente (NCKP).
The trial in Finland was a randomized, double-blind trial in which 1,662 infants were equally randomized to receive either Prevnar® or a control vaccine (Hepatitis B vaccine [Hep B]) at 2, 4, 6, and 12-15 months of age. All infants received a Diphtheria Tetanus Pertussis Vaccine - Haemophilus influenzae type b vaccine (DTP-Hib) combination vaccine concurrently at 2, 4, and 6 months of age, and Inactivated Poliovirus Vaccine (IPV) concurrently at 12 months of age. Parents of study participants were asked to bring their children to the study clinics if the child had respiratory infections or symptoms suggesting acute otitis media (AOM). If AOM was diagnosed, tympanocentesis was performed, and the middle ear fluid was cultured. If S. pneumoniae was isolated, serotyping was performed.
AOM was defined as a visually abnormal tympanic membrane suggesting effusion in the middle ear cavity, concomitantly with at least one of the following symptoms of acute infection: fever, ear ache, irritability, diarrhea, vomiting, acute otorrhea not caused by external otitis, or other symptoms of respiratory infection. A new visit or "episode" was defined as a visit with a study physician at which time a diagnosis of AOM was made and at least 30 days had elapsed since any previous visit for otitis media. The primary endpoint was efficacy against AOM episodes caused by vaccine serotypes in the per protocol population.
In the NCKP invasive disease efficacy trial, the effectiveness of Prevnar® in reducing the incidence of otitis media was assessed from the beginning of the trial in October 1995 through April 1998. During this time, 34,146 infants were randomized to receive either Prevnar® (N=17,070), or the control, an investigational meningococcal group C conjugate vaccine (N=17,076), at 2, 4, 6, and 12-15 months of age.
Physician visits for otitis media were identified by physician coding of outpatient encounter forms. Because visits may have included both acute and follow-up care, a new visit or "episode" was defined as a visit that was at least 21 days following a previous visit for otitis media (at least 42 days, if the visit appointment was made > 3 days in advance). Data on placement of ear tubes were collected from automated databases. No routine tympanocentesis was performed, and no standard definition of otitis media was used by study physicians. The primary otitis media endpoint was efficacy against all otitis media episodes in the per protocol population.
Table 2 presents the per protocol and intent-to-treat results of key otitis media analyses for both studies. The per protocol analyses include otitis media episodes that occurred ≥14 days after the third dose. The intent-to-treat analyses include all otitis media episodes in children who received at least one dose of vaccine.
TABLE 2 Efficacy of Prevnar® Against Otitis Media in the Finnish and NCKP Trials 20,21,22,23
| ||Per Protocol ||Intent-to-Treat |
| ||Vaccine |
|Finnish Trial ||N=1632 ||N=1662 |
| AOM due to Vaccine Serotypes || 57% || 44, 67 || 54% || 41, 64 |
| All culture-confirmed pneumococcal |
AOM regardless of serotype
|34% ||21, 45 ||32% ||19, 42 |
| NCKP Trial || N=23,746 || N=34,146 |
| All Otitis Media Episodes regardless of etiology † || 7% || 4, 10 || 6% || 4, 9 |
|* All vaccine efficacy estimates in the table are statistically significant. |
† The vaccine efficacy against all AOM episodes in the Finnish trial, while not reaching statistical significance, was 6% (95% CI: -4, 16) in the per protocol population and 4% (95% CI: -7, 14) in the intent-to-treat population.
The vaccine efficacy against AOM episodes due to vaccine-related serotypes (6A, 9N, 18B, 19A, 23A), also assessed in the Finnish trial, was 51% (95% CI: 27, 67) in the per protocol population and 44% (95% CI: 20, 62) in the intent-to-treat population. The vaccine efficacy against AOM episodes caused by serotypes unrelated to the vaccine was -33% (95% CI: -80, 1) in the per protocol population and -39% (95% CI: -86, -3) in the intent-to-treat population, indicating that children who received Prevnar® appear to be at increased risk of otitis media due to pneumococcal serotypes not represented in the vaccine, compared to children who received the control vaccine. However, vaccination with Prevnar® reduced pneumococcal otitis media episodes overall.
Several other otitis media endpoints were also assessed in the two trials. Recurrent AOM, defined as 3 episodes in 6 months or 4 episodes in 12 months, was reduced by 9% in both the per protocol and intent-to-treat populations (95% CI: 3, 15 in per protocol and 95% CI: 4, 14 in intent-to-treat) in the NCKP trial. This observation was supported by a similar trend, although not statistically significant, seen in the Finnish trial. The NCKP trial also demonstrated a 20% reduction (95% CI: 2, 35) in the placement of tympanostomy tubes in the per protocol population and a 21% reduction (95% CI: 4, 34) in the intent-to-treat population.
Data from the NCKP trial accumulated through an extended follow-up period to April 20, 1999, in which a total of 37,866 children were included (18,925 in Prevnar® group and 18,941 in MnCC control group), resulted in similar otitis media efficacy estimates for all endpoints. 24
Subjects from a subset of selected study sites in the NCKP efficacy study were approached for participation in the immunogenicity portion of the study on a volunteer basis. Immune responses following three or four doses of Prevnar® or the control vaccine were evaluated in children who received either concurrent Diphtheria and Tetanus Toxoids and Pertussis Vaccine Adsorbed and Haemophilus b Conjugate Vaccine (Diphtheria CRM197 Protein Conjugate), (DTP-HbOC), or Diphtheria and Tetanus Toxoids and Acellular Pertussis Vaccine Adsorbed (DTaP), and Haemophilus b Conjugate Vaccine (Diphtheria CRM197 Protein Conjugate), (HbOC) vaccines at 2, 4, and 6 months of age. The use of Hepatitis B (Hep B), Oral Polio Vaccine (OPV), Inactivated Polio Vaccine (IPV), Measles-Mumps-Rubella (MMR), and Varicella vaccines were permitted according to the AAP and ACIP recommendations.
Table 3 presents the geometric mean concentrations (GMC) of pneumococcal antibodies following the third and fourth doses of Prevnar® or the control vaccine when administered concurrently with DTP-HbOC vaccine in the efficacy study.
TABLE 3 Geometric Mean Concentrations (Î¼g/mL) of Pneumococcal Antibodies Following the Third and Fourth Doses of Prevnar® or Control* When Administered Concurrently With DTP-HbOC in the Efficacy Study 20,21
|Serotype ||Post dose 3 GMC† |
(95% CI for Prevnar®)
|Post dose 4 GMC‡ |
(95% CI for Prevnar®)
| ||Prevnar® § ||Control* ||Prevnar® § ||Control* |
| ||N=88 ||N=92 ||N=68 ||N=61 |
|4 ||1.46 |
|0.03 ||2.38 |
|6B ||4.70 |
|0.08 ||14.45 |
|9V ||1.99 |
|0.05 ||3.51 |
|14 ||4.60 |
|0.05 ||6.52 |
|18C ||2.16 |
|0.04 ||3.43 |
|19F ||1.39 |
|0.09 ||2.07 |
|23F ||1.85 |
|0.05 ||3.82 |
|* Control was investigational meningococcal group C conjugate vaccine (MnCC). |
† Mean age of Prevnar® group was 7.8 months and of control group was 7.7 months.
N is slightly less for some serotypes in each group.
‡ Mean age of Prevnar® group was 14.2 months and of control group was 14.4 months.
N is slightly less for some serotypes in each group.
§ p<0.001 when Prevnar® compared to control for each serotype using a Wilcoxon's test.
In another randomized study (Manufacturing Bridging Study, 118-16), immune responses were evaluated following three doses of Pneumococcal 7-valent Conjugate Vaccine (Diphtheria CRM197 Protein), Prevnar® administered concomitantly with DTaP and HbOC vaccines at 2, 4, and 6 months of age, IPV at 2 and 4 months of age, and Hep B at 2 and 6 months of age. The control group received concomitant vaccines only. Table 4 presents the immune responses to pneumococcal polysaccharides observed in both this study and in the subset of subjects from the efficacy study that received concomitant DTaP and HbOC vaccines.
TABLE 4 Geometric Mean Concentrations (Î¼g/mL) of Pneumococcal Antibodies Following the Third Dose of Prevnar® or Control* When Administered Concurrently With DTaP and HbOC in the Efficacy Study† and Manufacturing Bridging Study 20,21,25
| ||Efficacy Study ||Manufacturing Bridging Study |
|Serotype ||Post dose 3 GMC‡ |
(95% CI for Prevnar®)
|Post dose 3 GMC§ |
(95% CI for Prevnar®)
| ||Prevnar® ll ||Control* ||Prevnar® ll ||Control* |
| ||N=32 ||N=32 ||N=159 ||N=83 |
|4 ||1.47 |
|0.02 ||2.03 |
|6B ||2.18 |
|0.06 ||2.97 |
|9V ||1.52 |
|0.04 ||1.18 |
|14 ||5.05 |
|0.04 ||4.64 |
|18C ||2.24 |
|0.04 ||1.96 |
|19F ||1.54 |
|0.10 ||1.91 |
|23F ||1.48 |
|0.05 ||1.71 |
|* Control in efficacy study was investigational meningococcal group C conjugate vaccine (MnCC) and in Manufacturing Bridging Study was concomitant vaccines only. |
† Sufficient data are not available to reliably assess GMCs following 4 doses of Prevnar® when administered with DTaP in the NCKP efficacy study.
‡ Mean age of the Prevnar® group was 7.4 months and of the control group was 7.6 months. N is slightly less for some serotypes in each group.
§ Mean age of the Prevnar® group and the control group was 7.2 months.
ll p<0.001 when Prevnar® compared to control for each serotype using a Wilcoxon's test in the efficacy study and two-sample t-test in the Manufacturing Bridging Study.
In all studies in which the immune responses to Prevnar® were contrasted to control, a significant antibody response was seen to all vaccine serotypes following three or four doses, although geometric mean concentrations of antibody varied among serotypes. 20,21,23, 25,26,27, 28,29,30 The minimum serum antibody concentration necessary for protection against invasive pneumococcal disease or against pneumococcal otitis media has not been determined for any serotype. Prevnar® induces functional antibodies to all vaccine serotypes, as measured by opsonophagocytosis following three doses. 30
Previously Unvaccinated Older Infants and Children
To determine an appropriate schedule for children 7 months of age or older at the time of the first immunization with Prevnar®, 483 children in 4 ancillary studies received Prevnar® at various schedules and were evaluated for immunogenicity. GMCs attained using the various schedules among older infants and children were comparable to immune responses of children, who received concomitant DTaP, in the NCKP efficacy study (118-8) after 3 doses for most serotypes, as shown in Table 5. These data support the schedule for previously unvaccinated older infants and children who are beyond the age of the infant schedule. For usage in older infants and children, see DOSAGE AND ADMINISTRATION .
TABLE 5 Geometric Mean Concentrations (Î¼g/mL) of Pneumococcal Antibodies Following Immunization of Children From 7 Months Through 9 Years of Age With Prevnar® 31
|Age group, |
|Study ||Sample |
|4 ||6B ||9V ||14 ||18C ||19F ||23F |
mo. 3 doses
|118-12 ||22 || 2.34 || 3.66 || 2.11 || 9.33 || 2.31 || 1.60 || 2.50 |
| ||118-16 ||39 || 3.60 || 4.63 || 2.04 || 5.48 || 1.98 || 2.15 || 1.93 |
mo. 2 doses
|118-15* ||82-84† || 3.91 || 4.67 || 1.94 || 6.92 || 2.25 || 3.78 || 3.29 |
| ||118-18 ||33 || 7.02 || 4.25 || 3.26 || 6.31 || 3.60 || 3.29 || 2.92 |
mo. 2 doses
|118-15* ||52-54† || 3.36 || 4.92 || 1.80 || 6.69 || 2.65 || 3.17 || 2.71 |
| ||118-18 ||45 || 6.85 || 3.71 || 3.86 || 6.48 || 3.42 || 3.86 || 2.75 |
mo. 1 dose
|118-18 ||53 || 5.34 || 2.90 || 3.43 ||1.88 || 3.03 || 4.07 || 1.56 |
mo. 1 dose
|118-18 ||52 || 6.27 || 6.40 || 4.62 || 5.95 || 4.08 || 6.37 || 2.95 |
yrs. 1 dose
|118-18 ||101 || 6.92 || 20.84 || 7.49 || 19.32 || 6.72 || 12.51 || 11.57 |
|31-32† ||1.47 ||2.18 ||1.52 ||5.05 ||2.24 ||1.54 ||1.48 |
|Bold = GMC not inferior to 118-8, DTaP post dose 3 (one-sided lower limit of the 95% CI of GMC ratio ≥0.50). |
* Study in Navajo and Apache populations.
† Numbers vary with serotype.