In a recently published report we demonstrated the safety, immunogenicity and efficacy of the CRM197 pneumococcal conjugate vaccine (PCV) in infants <2 years of age. 1 Low birth weight (LBW; ≤2500 g) and preterm (PT; ≤38 weeks) infants were included in that study, but the data were not separately analyzed. We report here on the safety, immunogenicity and efficacy of PCV in these infants based on the data collected in that trial.
The methods for the evaluation of safety, immunogenicity and efficacy in the Kaiser Permanente efficacy trial have been described previously. 1 Therefore only the additions to the methods relevant to this analysis in LBW and PT infants are described here. In the efficacy trial 37 868 infants were enrolled in a double blind randomized study to receive PCV or meningococcal C conjugate vaccine (MCV) on a 1:1 basis. Infants received either PCV or MCV concomitantly with routine childhood vaccines administered at recommended ages. LBW and PT were not exclusion criteria if the infants were healthy at the time of their visit. Study protocol required patients be vaccinated at outpatient clinics. Therefore LBW and PT infants were not eligible to participate until they were discharged from the hospital. Infants were excluded if they received gamma-globulin-containing product (i.e. Respigam) within 3 months before immunization. One infant (28 weeks, 1430 g) received 1 dose of Respigam 1 week after completing 3 doses of PCV. No fourth dose was given. LBW and PT infants were identified through specific birth hospitalization discharge codes in place within Kaiser Permanente databases that specify gestational age and birth weight. Cases of invasive pneumococcal disease were identified in the subpopulation of LBW and PT infants via the same methodology as in the efficacy trial. 1
To evaluate the risk of invasive pneumococcal disease in LBW and PT infants, the rates of invasive disease in LBW and PT infants who had received the MCV were compared with the rates of invasive disease in infants weighing >2500 g (NBW) and with term infants (FT), respectively, who had also received the control vaccine.
Vaccine efficacy against invasive pneumococcal disease was evaluated with the binomial test of the null hypothesis that the vaccine had no efficacy for the seven serotypes. Exact binomial confidence intervals were calculated by the Klopper-Pearson method. 2
Serum antibody responses (IgG) to the seven pneumococcal vaccine serotypes were determined by enzyme-linked immunosorbent assay. 3 Serum samples were obtained before the first dose and 1 month after the third dose.
To evaluate safety, rates of hospitalization and emergency department (ED) visits within 30 days postvaccination were assessed for infants receiving at least three doses of vaccine by 12 months of age comparing LBW and PT infants with NBW and FT infants. In addition rates of events in LBW and PT infants receiving PCV were also compared with those of LBW and PT infants who received MCV. Similar analyses were conducted for local and systemic reactions observed within 48 h of receipt of PCV as assessed through telephone interviews of the parents of a subset of the total cohort of these infants. For these analyses statistical significance was assessed using a hierarchical linear model adjusting for nonindependence of repeat responses to multiple doses in the same patients.
In the group of 37 868 infants there were 1756 LBW and 4340 PT subjects (Table 1). There was no difference in mean age of administration of vaccine for LBW and PT infants compared with NBW and FT infants in the study cohort (Table 2). The relative risk (RR) for pneumococcal invasive disease in these groups is shown in Table 3. The ages at invasive pneumococcal disease and diagnoses seen were: infants between 1000 and 1500 g at birth, bacteremia at 8 months; infants 1501 to <2500 g at birth, septicemia at 25 months, pneumonia at 30 months, sepsis at 21 months, bacteremia at 19 months, bacteremia at 32 months; infants <32 weeks gestational age, bacteremia at 8 months, septicemia at 25 months; infants 32 to <36 weeks, pneumonia at 19 months, infants 36 to <38 weeks, pneumonia at 4 months, pneumonia at 24 months, pneumonia at 30 months, meningitis at 4 months, bacteremia at 7 months, bacteremia at 19 months. For LBW infants the RR was 2.6 (P = 0.03); for PT infants the RR was 1.6 (P = 0.06). Vaccine efficacy for both groups was 100% (Table 4).
Post-Dose 3 immunogenicity for LBW and PT infants compared with NBW and FT infants is shown in Table 5. There was no significant difference in geometric mean titer between LBW and NBW infants for any of the vaccine serotypes. However, PT infants had significantly higher serum values than FT infants for serotypes 19F, 9V and 4. There was no significant difference between any of the groups for the percentage with titers ≥0.15 μg/ml.
LBW infants compared with NBW infants in the telephone interviewed subset were reported to have swelling and tenderness at the injection site as well as fever ≥38°C that were not significantly different (Table 6). Redness was significantly greater in this type analysis for NBW infants. When stratified by individual dose LBW infants were reported to have higher rates of local redness >3 cm (RR = 4.7, P = 0.02) and swelling >3 cm (RR = 3.2, P = 0.05), both after Dose 3. For fever and local reactions there was no difference between LBW infants receiving PCV and similar infants receiving MCV (Table 6).
Local reactions and fever adjusted for clustering among multiple doses per child did not show any significant difference between PT and FT infants (Table 7). However, when size of reaction is considered, PT infants had significantly more often swelling >2.4 cm than FT infants (RR 1.4, P = 0.05). For PT infants fever ≥38°C, swelling and tenderness were observed more commonly in PCV than in MCV recipients (Table 7).
For systemic events there was no difference between LBW and NBW infants. In comparing LBW recipients of PCV with those receiving MCV, PCV recipients had significantly more hives, with 6 events in PCV infants and none in MCV controls (P = 0.03). These reactions were seen within 48 h of vaccination. In the overall trial a similar difference between PCV and MCV was seen only after Dose 1 in infants receiving concomitant diphtheria-tetanus toxoids-acellular pertussis vaccine (10 vs. 2 cases, P = 0.02).
Compared with FT infants PT infants had significantly more vomiting (RR 1.2, P = 0.03). PT infants receiving PCV had more irritability (RR 1.1, P = 0.02), loss of appetite (RR 1.2, P = 0.04), vomiting (RR 1.2, P = 0.04), diarrhea (RR 1.3, P = 0.05) and hives (RR 2.2, P = 0.05) than infants receiving MCV.
In our automated database all hospitalizations and ED visits were recorded for the entire cohort. Rate comparisons were made for 61 separate diagnostic categories between study infants and NBW and FT infants who were hospitalized within 30 days. There were 73 diagnostic categories for ED visits within 30 days for which similar comparisons were made.
ED visits for LBW infants were significantly higher than those for NBW infants only for inguinal hernia with 3 cases in the LBW group vs. none in the NBW group (P < 0.001). For PT infants significant values were seen for inguinal hernia (3 vs. 0 cases, P = 0.002) and gastrointestinal reflux (RR 13.6, P = 0.05). Febrile seizure was more common in PT immunized infants than FT infants who received concomitant diphtheria-tetanus toxoids-whole cell pertussis vaccine and Haemophilus influenzae type b vaccine (RR 13.6, P = 0.05). There was no difference in the groups who received diphtheria-tetanus toxoids-acellular pertussis vaccine instead of diphtheria-tetanus toxoids-whole cell pertussis vaccine.
For hospitalizations, in comparison with NBW infants across all vaccine comparisons and diagnoses, LBW infants had significantly higher rates of inguinal hernia (RR 10.1, P < 0.001), elective procedures (RR 3.8, P = 0.02) and all diagnoses combined (RR 2.0, P = 0.006). PT infants had a higher rate of bronchiolitis (RR 2.7, P = 0.004), inguinal hernia (RR 3.8, P = 0.03), urinary tract infection (RR 4.5, P = 0.03) and all diagnoses combined (RR 2.3, P < 0.001) than FT infants. PT and LBW infants receiving PCV were twice as likely to be hospitalized as FT and NBW infants (RR 2.0, P = 0.006; RR 2.3, P < 0.001, respectively) but not more likely than their counterparts receiving MCV (RR 1.0, P = 0.92; RR 0.7, P = 0.19, respectively).
LBW and PT infants form a small but important proportion of total birth cohorts. In general data to recommend age at initiation and intervals for vaccine administration for these infants have been collected after data obtained for NBW and FT infants or not at all. Although in most cases there has been no need to modify the immunization schedule for LBW and PT infants, this is not always the case. 4 Therefore for each vaccine destined to be used in LBW and PT infants, safety, immunogenicity and efficacy data must be available for reasonable vaccination recommendations.
In our Phase III study of PCV, although infants could not be vaccinated while hospitalized, LBW and PT infants were not excluded. Therefore we obtained data on the risk, safety, immunogenicity and efficacy of the vaccine in these infants as part of the trial. It is clear that LBW and PT infants are at an increased risk for pneumococcal invasive disease. Although the numbers are small it also appears that the PCV is highly effective in these infants. What is reassuring is the fact that not only is the immunogenicity similar to those in NBW and FT infants for most of the serologies but also PT infants had significantly higher geometric mean titers than FT infants for three serotypes. These serum values were determined in a subgroup of infants that included 1 who was <750 g at birth, 12 who were between 750 and 999 g and 50 who were between 1000 and 1500 g. There were 76 infants with a gestational age of <32 weeks in the LBW immunogenicity subgroup.
PCV also appears to be as safe for LBW and PT infants as for NBW and FT infants. The elevated risk seen in the hospital and ED settings were for diagnoses commonly associated with LBW and PT in all such infants.
Of interest is the significant increase in febrile seizures seen in PT infants as compared with FT infants who were vaccinated with PCV concomitantly with diphtheria-tetanus toxoids-whole cell pertussis vaccine and H. influenzae type b vaccines. Prematures are predisposed to febrile seizures even without immunizations. In some instances these events have been associated with maternal abnormalities during pregnancy or with birth injury. 5, 6 However, in some reports there were no perinatal or neonatal complication or any association with neurodevelopmental abnormalities. 7, 8 In short febrile seizures in prematures may be the result of maternal, neonatal, perinatal or unidentified risk factors.
Increased risks of local and systemic reactions were few and generally associated with individual doses of PCV rather than overall. These isolated local and systemic reactions, including hives, were more commonly seen with PCV than with MCV, which was a pattern of reactogenicity similar to that seen in our previous study. 1
With regard to efficacy, immunogenicity and safety, these data support the use of PCV in LBW infants and PT infants.
1. Black S, Shinefield H, Fireman B, et al. Efficacy, safety and immunogenicity
of heptavalent pneumococcal conjugate vaccine in children. Pediatr Infect Dis J 2000; 19: 187–95.
2. STAT EXACT statistical package. Cambridge, MA: Cytel Statistical Software, 1998.
3. Quataert SA, Kirch CS, Wiedl LJ, et al. Assignment of weight-based antibody unites to a human anti-pneumococcal standard reference serum Lot 89-S. Clin Diag Lab Immunol 1995; 2: 590–7.
4. Lau YL, Tam AC, Ng KW. Clinical laboratory observations: response of preterm infants to hepatitis B vaccine. J Pediatr 1992; 121: 962–5.
5. Greenwood R, Golding J, Ross E, Verity C. Prenatal and perinatal antecedents of febrile convulsions and afebrile seizures: data from a national cohort study. Paediatr Perinat Epidemiol 1998; 12 (Suppl 1): 76–95.
6. Nalin A, Frigieri G, Cordioli A, et al. The risk of convulsions: a longitudinal study of normal babies and infants with neonatal damage in the first 6 years of life. Childs Nerv Syst 1990; 6: 254–63.
7. Forsgren L, Sidenvall R, Blomquist HK, et al. Pre- and perinatal factors in febrile convulsions. Acta Paediatr Scand 1991 Feb; 80: 218–25.
8. Ling SG. Clinical characteristics and risk factors for a complex first febrile convulsion. Singapore Med J 2001; 42: 264–7.