This analysis is part of a larger project describing the impact of PCV dosing schedules on IPD, immunogenicity, NP carriage, pneumonia and indirect effects.^11,18–21 Details on the literature search terms and methods used in this systematic review are described elsewhere (see Methods Appendix²²). In brief, a systematic literature review was performed to collect all available English language data published from January 1994 to September 2010 (supplemented post hoc with studies from 2011) on the effect of various PCV vaccination schedules among immunized children on immunogenicity, NP colonization, IPD, pneumonia and indirect effects among unvaccinated populations. Articles published in 14 databases, from ad hoc unpublished sources and abstracts from meetings of the International Symposium on Pneumococci and Pneumococcal Disease (1998–2010) and the Interscience Conference on Antimicrobial Agents and Chemotherapeutics (1994–2010), were searched. We included all randomized controlled clinical trials, nonrandomized trials, surveillance database analyses and observational studies of any PCV schedule on 1 or more outcomes of interest. Studies were included for abstraction if pneumococcal polysaccharide vaccine was used as a booster dose, but not as a primary dose. Titles and abstracts were reviewed twice and those with relevant content on 1 of the 5 outcomes (immunogenicity, carriage, invasive disease, pneumonia and indirect effects) underwent full review using a standardized data collection instrument. We did not search non-English language literature because of the low likelihood they would have relevant data for this project. Details on the search methods are provided in the Methods Appendix.²²

Citations recovered through the literature search went through several stages of independent review to determine their eligibility, as described elsewhere.²² Citations meeting inclusion criteria were categorized on an outcome specific basis into “study families,” where each family included abstracts or publications generated from a single protocol, population, surveillance system or other data collection system relevant to that outcome. Investigators identified primary data from the individual studies making up each study family for inclusion in the analysis. The primary data were selected as the most current and complete data available for that study family. In some cases, these data were drawn from >1 publication within a family. We also defined “study arms” as a group of children distinguished by immunization schedule or PCV product.

We abstracted core information on the following: number of children in a “study arm”; PCV manufacturer, valency and conjugate protein; coadministered vaccines; country; age at each dose and date of study and publication. Additional data abstracted for the analysis on immunogenicity covariates included study population characteristics (HIV status, sickle cell disease, indigenous subgroups and other high-risk groups) and ELISA laboratory methods. Results that were abstracted included ELISA IgG GMC, the percentage of children with ST-specific antibody concentrations >0.35 µg/mL [or 0.2 µg/mL if GlaxoSmithKline (GSK) ELISA method used], and whether other assays were performed such as opsonophagocytic assay and avidity measures.

Study arms meeting the following criteria were included in the analysis: children immunized with at least 2 primary doses of PCV, with a first dose at ≤ 4 months of age and the last primary dose at ≤ 6 months of age; used licensed or similar to licensed PCV products 7-valent or higher and provided information on ELISA IgG GMC for any of the 6 STs of interest (1, 5, 6B, 14, 19F or 23F). Merck and Aventis products were excluded because they were not pursued for licensure and contained carriers that are not the same as those used in licensed products. Study arms evaluating only high-risk populations, including those with HIV infection, sickle cell disease, chronic illness and indigenous subgroups, were excluded.

Study arms with immunogenicity data after a second or third primary dose were defined as “2 primary dose” or “3 primary dose” arms, respectively. Any 3 primary dose schedules that provided immunogenicity data following the second dose were also included with the 2 primary dose schedules. “2+0” and “2+1” schedules refer to 2 primary doses without and with a booster dose, respectively. A booster dose was defined as immunization between 9 and 18 months of age where infants had already received 2 or more doses of PCV. Mean age at immunization, if available, defined age at each dose; otherwise, scheduled age was used. To collapse into schedules, age was rounded to the nearest 2 weeks. Interval between doses was determined by the number of months between first and second primary dose.

We aimed to assess the effect of PCV dosing schedules and study-specific covariates on mean ST-specific GMCs while considering that most studies did not have intrastudy comparisons of schedules. We therefore followed an ecological regression approach to compare dosing schedules across studies. We fitted random effects meta-regression models of log-transformed GMC levels by ST, weighting by the inverse of their variances and calculating robust standard errors to account for multiple arms within studies.^23,24 We evaluated the effect of geographic region, PCV product and coadministered diphtheria, tetanus and pertussis vaccine (DTP), adjusted for the effect of the other covariates and for number of doses, age at first dose, interval between doses and ELISA laboratory method. For studies that did not report the variances of the GMC, we assigned the average variance reported by studies within the same region. A detailed description of how missing values were accounted for can be found in the accompanying immunogenicity article.¹¹

Using the coefficients for each covariate, the regression model output was used to estimate the GMC for combinations of region, coadministered DTP type and schedule, including potential schedules that have not been studied in the existing literature. The most common vaccine schedule in a given region was used as the reference group to compute GMC for other schedules. The PCV9 vaccine was used as the reference to generate GMC values for STs 1 and 5, and PCV7 (Prevnar) was used as the reference for STs 6B, 14, 19F and 23F.

Analyses were done in SAS 9.2 (SAS Institute Inc., Cary, NC) and STATA 11 (StataCorp, College Station, TX).

In every region, except Africa, we found diversity in the type of DTP vaccine coadministered with PCV; in Africa, all studies used whole-cell DTP (DTwP) as the coadministered vaccine (Table 2). DTwP was used in many studies in every region except Europe. Except Africa, every region had 2-, 4- and 6-month schedules evaluated with both acellular pertussis DTP (DTaP) and DTwP enabling comparison of the effect of DTP within and between regions without confounding by other factors. 2+1 schedules were administered with DTaP only, all from European settings (data not shown). Only in Europe could we compare DTaP with DTwP for different PCV schedules (3, 4 and 5 months and 2 and 4 months). PCV7 was evaluated in all regions. Only 1 and 2 non-PCV7 products were evaluated in North America and Asia/Oceania, respectively, and Europe was the only region with analyzable data on all 5 PCV products.

Regression analysis controlling for covariates found that 3 primary PCV doses produced significantly higher GMCs than 2 primary doses (Table 3). We also noted higher GMCs for schedules that had 2-month intervals between doses in the primary series compared with 1-month intervals for STs 6B, 14 and 23F, although these results were not significant. Age at first dose, interval between doses, number of doses and age at last dose were interrelated (3 of these factors determine the fourth); therefore, we could not evaluate all of these factors independently. All but 1 study retained the same interval between first to second and second to third primary doses. In our analysis, we evaluated age at first dose and age at last dose separately and found higher GMCs with increasing age, although this was only significant for ST6B with a 1.29-fold increase for each month of increasing age.

Geographic region was associated with both pre- and post immunization GMC values (Fig. 1). Although preimmunization GMC data were sparse, GMCs appeared lowest in North America (N = 2 for STs 1 and 5, N = 9 for other STs) and Latin America (N = 1) and highest in Africa (N = 3 for STs 1 and 5, N = 5 for other STs). For some STs, preimmunization GMCs were inversely related to postimmunization GMCs; in that, if preimmunization GMCs were low relative to other regions, their postimmunization GMCs were high relative to other regions.

Type of DTP coadministered had no effect on immunogenicity for STs 6B, 19F and 23F, but DTaP was associated with 1.6-fold higher GMCs for ST14 than DTwP (P < 0.01) and 1.4- and 1.3-fold higher GMCs for STs 1 and 5, and the results were not significant (Fig. 2). When limiting evaluations to homogeneous settings in North America and Europe, DTaP coadministration remained associated with a higher GMC for ST14.

ST-specific postprimary GMCs varied by PCV product tested. Compared with PCV7, GSK PCV10 had lower GMCs for all STs evaluated in common, but significantly higher GMC for ST19F after adjusting for ELISA method (Table 3). PCV13 was also lower than PCV7 for the 4 STs evaluated in common, but there were few PCV13 studies and the difference was not statistically significant. Immunogenicity to all GSK products was evaluated using the GSK ELISA laboratory method, which is known to produce lower absolute values than other ELISA measurement methods.

Using the output from the regression model, we estimated GMCs for plausible schedules, including some which have not been reported in the existing literature, combined with DTP type for each region (Table 4). The projected change in GMC comparing the 3-dose 6-, 10- and 14-week schedule with a 2-dose 6- and 14-week schedule in Africa is relatively small for STs 1 and 5 (changing from GMC = 5.0 µg/mL for both STs to GMC = 4.77 and 3.88 µg/mL, respectively), but for the other STs the decrease in GMC is more substantial (ie, ST6B dropped from GMC 0.97 to 0.27 µg/mL, ST14 dropped from 2.51 to 1.33 µg/mL). Although this hypothetical schedule cannot be verified directly, a study by Ota et al⁶⁹ showed a GMC of 0.05 and 1.03 for STs 6B and 14, respectively, using a similar 2- and 3-month schedule; the GMC in the 3-dose group was 3.47 for ST 6B and 4.65 for ST 14. In Asia, the predicted fold change was similar, but because GMCs were higher in Asia than in Africa, the GMCs for the 2-dose 6- and 14-week schedule in Asia are similar to the GMCs for the 3-dose 6-, 10- and 14-week schedule in Africa (eg, for ST19F in Africa, the GMC = 4.26 µg/mL with 3 doses and in Asia, the GMC = 4.25 µg/mL for 2 doses).

Predicted GMC responses followed similar trends in North America and Europe. In North America, the predicted change in GMC comparing a 2-, 4- and 6-month schedule with a 2- and 4-month schedule coadministered with DTaP remains relatively small for STs 1 and 5 (changing from 3.00 and 2.34 µg/mL to 2.73 and 1.75 µg/mL, respectively). A larger change is predicted for the other STs, with GMCs changing from 1.09 to 0.16 and 4.50 µg/mL to 1.78 µg/mL for STs 6B and 14, respectively. Increasing the interval between primary doses from 1 to 2 months also tended to increase GMCs, although less substantially than increasing the number of doses. Lowest GMCs were predicted for 2-dose schedules with 1 month between doses. Nearly all schedules produced predicted GMCs above the 0.35 µg/mL value correlated with high vaccine efficacy in children except for certain 2-dose schedules in Europe, North America, Africa and Latin America for STs 6B and 23F.