CONTEXT:

Pneumococcal conjugate vaccines (PCVs) (pneumococcal 13-valent conjugate vaccine [PCV-13] and pneumococcal 10-valent conjugate vaccine [PCV-10]) are available for prevention of pneumococcal infections in children.

OBJECTIVE:

To determine the vaccine effectiveness (VE) of PCV-13 and PCV-10 in preventing invasive pneumococcal disease (IPD) and acute otitis media (AOM) in children <5 years.

DATA SOURCES:

Systematic searches of Medline, Embase, Cumulative Index to Nursing and Allied Health Literature, Web of Science, and Cochrane.

STUDY SELECTION:

Eligible studies examined the direct effectiveness and/or efficacy of PCV-10 and PCV-13 in reducing the incidence of disease in healthy children <5 years.

DATA EXTRACTION:

Two reviewers independently conducted data extraction and methodologic quality assessment.

RESULTS:

Significant effectiveness against vaccine-type IPD in children ≤5 years was reported for ≥1 dose of PCV-13 in the 3 + 1 (86%–96%) and 2 + 1 schedule (67.2%–86%) and for PCV-10 for the 3 + 1 (72.8%–100%) and 2 + 1 schedules (92%–97%). In children <12 months of age, PCV-13 VE against serotype 19A post–primary series was significant for the 3 + 1 but not the 2 + 1 schedule. PCV-10 crossprotection against 19A was significant in children ≤5 years with ≥1 dose (82.2% and 71%). The majority of studies did not find either PCV to be effective against serotype-3. PCV-13 was effective against AOM (86%; 95% confidence interval [CI]: 61 to 94). PCV-10 was effective against clinically defined (26.9%; 95% CI: 5.9 to 43.3) and bacteriologically confirmed AOM (43.3%; 95% CI: 1.7 to 67.3).

LIMITATIONS:

Because of the large heterogeneity in studies, a meta-analysis for pooled estimates was not done.

CONCLUSIONS:

Both PCVs afford protection against pneumococcal infections, with PCV-10 protecting against 19A IPD, but this VE has not been verified in the youngest age groups.

Streptococcus pneumoniae and Haemophilus influenzae are recognized for their infectious potential in young children.1  Nontypeable Haemophilus influenzae (NTHi) strains, characterized by their lack of a polysaccharide capsule, are associated with noninvasive mucosal diseases, such as acute otitis media (AOM) and sinusitis. Although not serious, AOM remains the leading cause for antimicrobial prescriptions in children in many countries, imposing a substantial burden on health care systems and potentially accelerating the development of antibiotic resistance.2 

To reduce the burden of invasive pneumococcal disease (IPD) in children, the pneumococcal 7-valent conjugate vaccine (PCV-7), or Prevnar 7, was licensed for use in Canada in 20013  and implemented in infant vaccination programs across all provinces and territories by 2006.4  Rapidly after implementation, the increase in non–vaccine-type IPD associated with serotype replacement threatened to offset the gains offered by the program.5  To address the problem, higher-valent vaccines were developed. The pneumococcal 10-valent conjugate vaccine (PCV-10), or Synflorix, became available in Canada in 2009, offering protection against 3 additional serotypes: 1, 5A, and 7F4 ; it also employed a novel carrier protein derived from NTHi presumed to grant protection against AOM and other diseases caused by NTHi.6  However, subjects vaccinated with PCV-10 showed a non-significant decrease in NTHi carriage in the year following booster vaccination.7 In 2010, pneumococcal 13-valent conjugate vaccine (PCV-13), or Prevnar13, was licensed, offering protection for PCV-10 serotypes with 3 additional serotypes: 3, 6A, and 19A.6  The inclusion of 19A, a serotype with high invasive potential and associated burden of disease, made this vaccine particularly interesting8  and was selected for infant immunization programs in most jurisdictions.8 

Despite their use in routine clinical practice, the comparative effectiveness of PCV-13 and PCV-10 for the prevention of IPD and AOM has yet to be assessed. Unlike PCV-7, for which efficacy was estimated from randomized controlled trials (RCTs), both PCV-10 and PCV-13 were licensed on the basis of a noninferior immunologic response for the 7 common serotypes found in PCV-7 (4, 6B, 9V, 14, 18C, 19F, and 23F6 ). Authors of recent observational studies of vaccine effectiveness (VE) have suggested that PCV-10 might afford crossprotection against the highly pathogenic 19A serotype.911  The expected superior protection of PCV-10 against AOM is however unclear.12 

With nearly 7 years after their worldwide implementation, there is an opportunity to assess the effectiveness of PCV-10 and PCV-13 to inform health policy discussions concerning pneumococcal infant immunization programs. We therefore conducted a systematic review of published studies in which authors evaluated the effectiveness of PCV-10 and PCV-13 in providing protection against IPD and AOM in children ages ≤5. We also evaluated VE at reducing pneumococcal nasopharyngeal carriage (NPC).

In June of 2016, we systematically searched Ovid Medline (1946–present), Embase (Ovid), Web of Science, and Cumulative Index to Nursing and Allied Health Literature for studies published between 2009 and 2016 examining the effectiveness or efficacy of PCV-10 and PCV-13 for protection against AOM and IPD in children. We combined free-text search terms for the concepts of “PCV-13” AND “PCV-10” and (“efficacy” OR “effectiveness” OR “safety” OR “AOM” OR “IPD”). The search was updated in July of 2018. A sample of the full search strategy is shown in detail in the Supplemental Information.

Two researchers independently assessed the eligibility of each study for inclusion. Full texts of eligible studies were obtained and assessed independently. A third reviewer was consulted when consensus could not be reached.

Studies were eligible for inclusion if they examined the direct effectiveness or efficacy of PCV-10 or PCV-13 in preventing or reducing the incidence of IPD and/or AOM in healthy children 5 years or younger. We included RCTs and cohort and case-control studies as well as surveillance studies with individual-level data about vaccination status. Pre-post studies capturing both direct and indirect VE were excluded. We restricted inclusion to studies with confirmed laboratory IPD diagnosis. For AOM, studies including diagnosis of AOM by diagnostic code, laboratory results, or clinical definitions were also included. Studies were excluded if they were only available as abstracts from conference proceedings or published in a language other than English. In addition, studies that included children who had received both PCV-13 and PCV-10 in the analysis of VE, presented data exclusively on individuals >5 years, or were specific to a subgroup of children with underlying medical conditions were excluded. Country economic statuses were based on rankings provided by the World Bank Group’s World Development Indicators13  on the year(s) the study was conducted.13 

Two reviewers independently conducted data extraction and methodologic quality assessment of included studies, using DistillerSR (Evidence Partners, Ottawa, Ontario, Canada). We extracted the country of study, study time frame, funding source, study design, sample size, follow-up time, surveillance method, how outcomes were defined and ascertained, vaccine assessed, comparator vaccine, vaccine schedule (2 + 1 or 3 + 1), method of vaccination status ascertainment, and how VE was calculated. For population characteristics, we extracted vaccine coverage in the region, whether the vaccine was publicly or privately funded, the year of pneumococcal vaccination program implementation, and the ages at which vaccination was recommended in the jurisdiction. For all studies, the main measure of interest was the effect of the vaccine at reducing the primary outcomes. For case-control and cohort studies, the relative measures were reported as odds ratios or incidence rate ratios. For RCTs, the measure was the incidence rate ratio or the hazard ratio. The VE was then calculated as VE = (1 − relative measure) × 100.

The outcomes extracted included VE and/or efficacy, effect measures and their confidence intervals (CIs), as well as confounders for which the model was adjusted. We focused on extracting VE against vaccine-type invasive pneumococcal disease (VT-IPD), vaccine-related IPD, serotypes unique to PCV-13 or PCV-10, and serotypes 19A and 3. For RCTs, only results from intention-to-treat analyses were extracted.

The risk of bias was evaluated through the National Advisory Committee on Immunization guidelines for quality assessment adapted from Harris et al,14  an adaptation of the methods employed by the US Preventive Task Force. Lack of control for age and/or underlying medical conditions15  as confounders was considered a flaw that would render a study as “fair.” Age was considered a confounder because younger children are at higher risk for AOM and/or IPD because of their anatomy and the immaturity of their immune system and may also be more or less likely to be vaccinated, depending on public health strategies. Fatal flaws were determined a priori and included inadequate selection of controls for case-control studies and inadequate maintenance of balanced groups for RCTs.

We summarized all included studies through descriptive analyses to provide an overview of studies’ characteristics, quality, and reported outcomes. Because of the heterogeneity in outcome assessment and the various stratifications for VE measures across studies, a meta-analysis was not performed. Between-study heterogeneity was evaluated by using visual assessment of forest plots. We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analysis.

Our initial search yielded a total of 3073 studies (Fig 1). No further studies were identified through Google Scholar or hand searching of relevant articles. After removing duplicates, 1331 articles remained and were screened by title and abstract. Of the 33 articles that underwent full-text screening, 12 met our inclusion criteria. Seven more articles were added after the updated search in 2018. When 2 studies employed the same data source, the earlier version was excluded.

From the 19 studies included (RCT = 6; case-control = 12; cohort = 1), 11 examined PCV-13 VE, with 1 examining VE against AOM and NPC compared to PCV-7, and the rest examined PCV-13 VE against IPD compared with unvaccinated children. Nine studies provided data for PCV-10: 3 examined VE against IPD compared to no pneumococcal vaccine; 2 examined VE against IPD when compared to either the hepatitis B vaccine or the diphtheria-tetanus-acellular pertussis–inactivated poliovirus Haemophilus influenzae type b vaccine (DTaP-IPV/Hib), or InfanrixTM inactivated poliovirus H influenzae type b, with a hepatitis A vaccine and DTaP-IPV/Hib booster. The remaining 4 studies reported on PCV-10 efficacy against AOM or a proxy measure. For NPC, 2 studies evaluated carriage after PCV-10 administration and 1 evaluated carriage after PCV-13. There were no direct comparisons of PCV-10 to PCV-13. Most studies were conducted in countries of high (n = 12) or upper-middle income status (n = 6), and only 1 was conducted in a country of lower-middle status.16  The majority of studies were rated as good (n = 16), 3 were rated as fair, and none were rated as poor (Table 1).

In children <5 years of age receiving at least 1 dose of PCV-13, VE against VT-IPD was consistently high for the 3 + 1 schedule17  (n = 4; range: 86% [95% CI: 74 to 93] to 96% [95% CI: 43 to 100]) and 2 + 1 schedule (n = 3; range: 67.2% [95% CI: 2.3 to 90] to 86% [95% CI: 62 to 95]) (Table 2). In the same age group, when restricting to IPD caused by serotypes unique to PCV-13, the estimate remained positive and significant (Supplemental Table 3). Only 1 study reported a nonsignificant VE of PCV-13 against VT-IPD with the 2 + 1 schedule among the subgroup of children up to date with their vaccine (Table 2).

For PCV-10, VE against VT-IPD in jurisdictions employing the 3 + 1 schedule (n = 4; range: 72.8% [95% CI: 44.1 to 86.7] to 100% [95% CI: 83 to 100]) or 2 + 1 schedule (n = 2; 92% [95% CI: 58 to 100] and 97% [95% CI: 84 to 99]) was also reliably high among children <5 years. Because no study was able to estimate PCV-10 VE against serotypes unique to its formulation, the VT-IPD estimates reflect VE against PCV-7 serotypes. PCV-13 and PCV-10 vaccine-type VE ± their 95% CIs are depicted in Fig 2 for comparison.

For the age subgroups with the highest incidence of IPD (12 and 24 months of age), effectiveness estimates were only provided for PCV-13. Against serotypes unique to the 13-valent formulation, the effectiveness post–primary series (ie, after the 2 or 3 priming doses but before the booster dose) in children <12 months of age hovered at ∼80% for both schedules (n = 2; 80% [95% CI: 46 to 93] and 80% [95% CI: 43 to 93] for 3 + 1 and 2 + 1 schedule, respectively). Among children 12 and 24 months old, VE of the 2 + 1 schedule against VT-IPD, although statistically significant, was lower than the corresponding estimate obtained with the 3 + 1 vaccine schedule (Supplemental Table 3).

The question of PCV-10’s crossprotection was addressed in 3 separate studies.10,18,19  Included in these analyses were serotypes in the same group as the vaccine-type serotypes (ie, 6A, 6C, 6D, 7C, 9N,18A, 18B, 19A, and 23A)10,18  or just serotypes 6A, 9N, and 19A.19  For children <5, Tregnaghi et al19  reported a nonsignificant vaccine-related VE of at least 1 dose of PCV-10 against serotypes 6A, 9N, or 19A (−99.5%; 95% CI: −2100 to 81.9). For IPD caused by 6A, 6C, 6D, 7C, 9N,18A, 18B, 19A, and 23A serotypes, PCV-10’s VE ranged from 64.8% (95% CI: 15.3 to 85.4)18  to 77.9% (95% CI: 41.0 to 91.7)10  for children who were up to date with their vaccination status, whereas for at least 1 dose, VE was 61.3%18  (95% CI: 14.5 to 82.5).

The VE for PCV-10 and PCV-13 against IPD caused by serotypes 19A or 3 could only be compared descriptively for the age strata including all children <5 years because this was the only strata for which both PCV-10 and PCV-13 had VE estimates. Four studies16,17,20,21  reported on the VE of at least 1 dose of PCV-13 against 19A IPD offering estimates ranging between 77%17  (95% CI: 47 to 90) and 85.6%21  (95% CI: 70.6 to 93.5) for the 3 + 1 schedule and reduced but significant estimates for the 2 + 1 schedule (Table 2). For the 10-valent formulation, 1 study evaluated the crossprotection against 19A IPD afforded by at least 1 dose of the vaccine in children ≤5 years11  (71.3%; 95% CI: 16.6 to 90.1). Although initially significant, this crosseffectiveness decreased postbooster (63.5%; 95% CI: −16.8 to 88.6). In the same age group, Domingues et al11  reported a 19A VE of 82.2% (95% CI: 10.7 to 96.4) in those who were up to date with the vaccine schedule at the time of assessment (Supplemental Fig 3). One study8  examined the 2 + 1 schedule VE against 19A IPD for both PCV-10 and PCV-13, offering comparable estimates of 74% (95% CI: 11 to 92) and 71% (95% CI: 24 to 89) for the 13- and 10-valent formulations, respectively8  (Table 2).

For serotype 3, PCV-13 VE in children <12 months post–primary series was not statistically significant in either the 3 + 1 schedule or 2 + 1 schedule16,22  (Supplemental Table 3). The same trend was found for both schedules after the booster dose for children <5 years (Table 2). VE estimates against serotype 3 were only statistically significant when considering all children <5 years of age who received at least 1 dose of PCV-13 in the 3 + 1 schedule because the sample size was larger for this category. These reported estimates ranged between 74%16  (95% CI: 2 to 93) and 79.5%21  (95% CI: 30.3 to 94.8).

The end points for the assessment of VE against AOM were variable across studies, thus limiting quantitative comparisons between vaccines. Only 1 study reported on PCV-13 VE against episodes of AOM.23  This was a prospective cohort study comparing the frequency of serotypes unique to PCV-13 in middle-ear fluid of children vaccinated in either a PCV-7 or PCV-13 cohort. In children <12 months of age who had received the primary series of PCV-13, the estimated VE of PCV-13 against AOM caused by the 6 additional serotypes in PCV-13 was 86% (95% CI: 61 to 94) with a relative VE against 19A of 91% (95% CI: 58 to 97) and against serotype 3–AOM of 15% (95% CI: −181 to 72).23 

For PCV-10, VE was examined through RCTs; thus, estimates represent the vaccine’s efficacy. For children <12 months of age, 1 study24  found a positive VE at preventing at least 1 clinically defined AOM episode (26.9%; 95% CI: 5.9 to 43.3) as well as for all episodes of AOM (23.7%; 95% CI: 1.3 to 41).24  The efficacy tended to fall off in older age groups (Supplemental Table 4). When the analysis was restricted to bacteriologically confirmed AOM, the efficacy estimate in children <12 months of age was significant when considering the first episode (43.3%; 95% CI: 1.7 to 67.3) but not for all AOM episodes (40.3%; 95% CI: −4 to 65.7)24 . Importantly, VE against NTHi AOM could not be demonstrated in children <5 years of age who had received at least 1 dose of the vaccine24  (Supplemental Table 4).

Pichichero et al23  examined differences in NPC in children vaccinated with PCV-7 or PCV-13. Post–primary series, PCV-13 resulted in reduced carriage of all 6 additional serotypes included in its formulation (76%; 95% CI: 58 to 85), with a relative effectiveness of 73% (95% CI: 52 to 84) for reducing carriage of serotype 19A.23  PCV-10 was not effective at reducing vaccine-type carriage when measured 1 month after the primary series of the 2 + 1 schedule (1.3%; 95% CI = −21.2 to 19.8)25  but was at the other time points examined.24,25  For both the 3 + 1 and 2 + 1 schedules, when NPC was measured at either the 1- or 6-month post–primary series or at the 3-month postbooster time point, PCV-10 VE at reducing NTHi carriage was not statistically significant.24,25  However, these studies saw low carriage levels in all groups across all time points, resulting in wide CIs.

Sixteen of the 19 studies examined met all of Harris et al’s14  stipulated criteria for good internal validity. The RCTs employed to assess PCV-10 VE had balanced groups and ensured maintenance of randomization by reporting intention-to-treat results.19,24,2628  Although control selection is always a challenge in case-control designs, most case-control studies matched on age and neighborhood of residence to reduce the bias in ascertainment of controls. Likewise, all but 3 case-control studies16,21,29  adjusted for underlying medical conditions when calculating VE estimates (Table 1).

Indirect cohort studies circumvent the issue of control selection by comparing VT-IPD cases to non–vaccine-type IPD controls. Four studies16,17,18,20  employed Broome et al’s30  indirect cohort method for VE calculation. In this method, differential serotype replacement in vaccinated and unvaccinated individuals can introduce potential bias and overestimate VE. Of the 4 studies employing the indirect cohort method for VE assessment, only 2 acknowledged the potential bias introduced by serotype replacement.16,17  Weinberger et al20  and Verani et al18  did not measure the potential effect of this bias; thus, the VE estimates may be overestimated. Three of the included studies were rated as fair.16,21,29  One had several imbalances between cases and controls, and the reported estimates were unadjusted for age or underlying medical conditions18 ; another did not adjust for underlying comorbidities, which was significantly different between cases and their matched controls.29  Lastly, Su et al16  did not adjust for underlying medical conditions, and additionally, the study was judged to have a significant risk of nondifferential information bias because of the surveillance network employed to confirm vaccination status.

The higher-valent pneumococcal conjugate vaccines (PCVs) were licensed on the basis of comparative immunogenicity to their precursor PCV-7. PCVs are among the most expensive vaccines currently available,31  and the question of their comparative effectiveness remains highly salient for public health initiatives worldwide. We identified 19 studies examining the direct VE of PCV-10 or PCV-13, but none compared the vaccines to each other. Differences in comparators, jurisdiction schedules, and the reported age group and/or dosing end points excluded the potential for a meta-analysis.

Across age groups, schedules, and number of doses, all studies reported high and statistically significant VE against VT-IPD for both vaccines in children <5 years of age. For PCV-10, no study was able to assess VE against serotypes unique to its formulation (1, 5A, and 7F); thus, its VE reflects PCV-7 VE. For serotype-specific effectiveness, all studies struggled with small numbers of serotype-specific IPD cases. Regardless of schedule, most of the included studies that examined PCV-13 reported statistically significant protection against IPD caused by the 19A serotype for at least 1 dose in all children <5 years.9,16,24,2931  Importantly, for children <12 months, the 19A-IPD VE post–primary series varied by schedule, with the 3 + 1 schedule giving a statistically significant VE that was not established in the 2 + 1 schedule.17,22  Three studies reported statistically significant crossprotection of PCV-10 against 19A IPD in all children <5 years of age receiving at least 1 dose and in those being up to date with the vaccine given their age at the time of assessment.8,10,18  Nevertheless, children <2 years of age are especially susceptible to IPD,33  and the question of expandability of the crossprotection to younger age groups is pending on future studies.

PCV-10 effectiveness against serotype 19A was observed in other studies that did not meet inclusion for the present review.10,34  A study in Chile reported a decrease in 19A-specific pneumococcal IPD, with 19A cases decreasing from 13 to 8 from pre-to-post– PCV-10 introduction.34  In Finland, a population follow-up study comparing a cohort of children who had received PCV-10 to a nonvaccinated season-age matched historical cohort reported a VE against serotype 19A of 62% (95% CI: 20 to 85).10  VE against 19A was shown consistently for PCV-13, whereas the results varied for PCV-10. Although PCV-10 was shown to induce an increase in antibody against serotype 19A after the primary series, this increase was consistently lower than that observed in infants receiving PCV-13, regardless of schedule.29 

For serotype-3, PCV-13 effectiveness was observed only when all children <5 years who received at least 1 dose of vaccine were included in the analysis. PCV-13 effectiveness against AOM caused by serotype 3 in infants <12 months of age was limited. The immunogenicity of PCVs against serotype 3 in infants was shown to be relatively low with no clear evidence of toddler boosting by several different serotype-3 conjugate vaccines as measured by enzyme-linked immunosorbent assay.7,3537 

The question concerning which of the higher-valent vaccines affords superior protection against AOM is of interest considering its high burden in children.2,38  Of studies examining PCV-10, most reported moderate to minimal direct protection against AOM. As reported from the Finnish Invasive Pneumococcal disease vaccine trial in Finland, estimates of PCV-10 efficacy against AOM episodes, or against tympanostomy procedures or antimicrobial purchases, were not significant. However, the authors emphasize that by design, the trial was underpowered to detect any differences in these outcomes. The Latin American Clinical Otitis Media and Pneumonia Study trial and resulting studies19,24  found positive estimates for clinically and culture-confirmed AOM episodes. However, with regard to NTHi-linked AOM, PCV-10 efficacy was positive but nonsignificant. PCV-10 is currently marketed and sold under the presumption that it affords superior protection against NTHi-associated AOM.38  Yet, as observed in this review, this has yet to be objectively established.

Although the overall frequency of AOM between cohorts vaccinated with PCV-7 and PCV-13 did not change, the relative frequency of AOM caused by the 6 additional S pneumoniae serotypes included in PCV-13 significantly decreased. No decrease in AOM caused by serotype 3 was detected in this study.23  A prospective study examined the effects of PCV-7 and PCV-13 sequential introduction on the incidence of pneumococcal-confirmed AOM. The authors reported a decline of 85% in the incidence of AOM caused by the additional serotypes included in PCV-13.39  However, given the nature of the pre-post design, it is difficult to untangle the direct effects from the indirect protection afforded by years of PCV-7 administration.

In a previous systematic review, de Oliveira et al31  examined the impact and effectiveness of the 13- and 10-valent vaccines on hospitalizations and mortality due to IPD in children <5 years of age residing in Latin American countries. Their study included articles published in any language and expanded the search to include the gray literature, including 22 published and unpublished studies. As in the current review, none of the studies compared PCV-10 to PCV-13, and meta-analysis was not possible. For radiologically confirmed pneumonia, the all-serotypes VE of PCV-13 in children <12 months of age ranged from 33% (95% CI: 25 to 41) to 44.6% (95% CI: 24.6 to 59.3). For PCV-10, VE ranged from 13.6% (95% CI: 3 to 24.3) to 25.3% (95% CI: 24.6 to 26.1). For meningitis hospitalizations and deaths, only studies examining PCV-10 met the inclusion criteria, giving a high vaccine-type VE in the 2 studies reporting on this outcome (0–23 months; VE = 77% [95% CI: 20 to 94]; 2 to 49 months, VE = 87.7% [95% CI: 61.4 to 96.1]); for death due to all serotypes, the estimates provided were high but had no associated CIs.31  The authors concluded that with respect to the outcomes examined, in children <5 years, there was no evidence to assert the superiority of one vaccine over the other.31  The review did not focus on serotype-specific PCV effectiveness and did not differentiate between different vaccine schedules.31  Additionally, the inclusion of unpublished data and pre-post studies, although minimizing publication bias, decreased the overall quality of the evidence analyzed, a limitation that we tried to address in the current study.

We analyzed the evidence for direct effectiveness of the higher-valent pneumococcal vaccines from published studies worldwide. We tried to maximize quality by restricting inclusion to published studies, which is reflected by the preponderance of studies that were assigned a “good” quality rating in analyses of internal validity. We provided a thorough summary of all the available evidence concerning serotype-specific VE, with a specific focus on serotypes 19A and 3, which remain contentious for PCV-10 and PCV-13, respectively. Notwithstanding these strengths, the present review has several potential limitations. We restricted the inclusion to published articles and excluded pre-post incidence studies. This may have discounted potentially relevant results, particularly those arising from developing economies. We further limited our inclusion to studies published in English and excluded the gray literature, which increases the potential for publication bias. Finally, because of the large heterogeneity in study design, end points, and age group and/or dosing stratification across studies, we were unable to perform a meta-analysis and provide pooled estimates of VE.

Although the effectiveness against VT-IPD was confirmed for both vaccines, the comparative assessment of PCV-10 and PCV-13 against AOM and serotypes 3 and 19A is precluded by the profound heterogeneity in the reported dose, age, and schedule combinations across studies. PCV-13 VE against 19A IPD was confirmed. Although PCV-10 seems to afford crossprotection against 19A IPD, the question of whether the VE is sufficiently effective in younger children remains unanswered. Finally, PCV-10’s superior protection against AOM caused by NTHi is of public health importance but has to yet to be confirmed in field studies of VE.

All authors contributed to the conception and design of the study, extracting, analyzing and interpretation of the data, and drafting the final manuscript, approved the final manuscript as submitted, and agree to be accountable for all aspects of the work.

FUNDING: No external funding.

AOM

acute otitis media

CI

confidence interval

DTaP-IPV/Hib

diphtheria-tetanus-acellular pertussis–inactivated poliovirus Haemophilus influenzae type b vaccine

IPD

invasive pneumococcal disease

NPC

nasopharyngeal carriage

NTHi

nontypeable Haemophilus influenzae

PCV

pneumococcal conjugate vaccine

PCV-7

pneumococcal 7-valent conjugate vaccine

PCV-10

pneumococcal 10-valent conjugate vaccine

PCV-13

pneumococcal 13-valent conjugate vaccine

RCT

randomized controlled trial

VE

vaccine effectiveness

VT-IPD

vaccine-type invasive pneumococcal disease

1
World Health Organization
.
Measuring impact of Streptococcus pneumoniae and Haemophilus influenzae type b conjugate vaccination.
2012
. Available at: www.who.int/iris/handle/10665/75835. Accessed September 10, 2017
2
Dubé
E
,
De Wals
P
,
Gilca
V
, et al
.
Burden of acute otitis media on Canadian families
.
Can Fam Physician
.
2011
;
57
(
1
):
60
65
3
Kellner
J
.
Update on the success of the pneumococcal conjugate vaccine
.
Paediatr Child Health
.
2011
;
16
(
4
):
233
240
4
Public Health Agency of Canada
.
Invasive pneumococcal disease.
2016
. Available at: https://www.canada.ca/en/public-health/services/immunization/vaccine-preventable-diseases/invasive-pneumococcal-disease/health-professionals.html. Accessed September 10, 2017
5
Torres
A
,
Bonanni
P
,
Hryniewicz
W
,
Moutschen
M
,
Reinert
RR
,
Welte
T
.
Pneumococcal vaccination: what have we learnt so far and what can we expect in the future?
Eur J Clin Microbiol Infect Dis
.
2015
;
34
(
1
):
19
31
6
Lee
H
,
Choi
EH
,
Lee
HJ
.
Efficacy and effectiveness of extended-valency pneumococcal conjugate vaccines
.
Korean J Pediatr
.
2014
;
57
(
2
):
55
66
7
Prymula
R
,
Peeters
P
,
Chrobok
V
, et al
.
Pneumococcal capsular polysaccharides conjugated to protein D for prevention of acute otitis media caused by both Streptococcus pneumoniae and non-typable Haemophilus influenzae: a randomised double-blind efficacy study
.
Lancet
.
2006
;
367
(
9512
):
740
748
8
Earnshaw
SR
,
McDade
CL
,
Zanotti
G
,
Farkouh
RA
,
Strutton
D
.
Cost-effectiveness of 2 + 1 dosing of 13-valent and 10-valent pneumococcal conjugate vaccines in Canada
.
BMC Infect Dis
.
2012
;
12
:
101
9
Deceuninck
G
,
De Serres
G
,
Boulianne
N
,
Lefebvre
B
,
De Wals
P
.
Effectiveness of three pneumococcal conjugate vaccines to prevent invasive pneumococcal disease in Quebec, Canada
.
Vaccine
.
2015
;
33
(
23
):
2684
2689
10
Jokinen
J
,
Rinta-Kokko
H
,
Siira
L
, et al
.
Impact of ten-valent pneumococcal conjugate vaccination on invasive pneumococcal disease in Finnish children--a population-based study
.
PLoS One
.
2015
;
10
(
3
):
e0120290
11
Domingues
CM
,
Verani
JR
,
Montenegro Renoiner
EI
, et al;
Brazilian Pneumococcal Conjugate Vaccine Effectiveness Study Group
.
Effectiveness of ten-valent pneumococcal conjugate vaccine against invasive pneumococcal disease in Brazil: a matched case-control study
.
Lancet Respir Med
.
2014
;
2
(
6
):
464
471
12
Schuerman
L
,
Borys
D
,
Hoet
B
,
Forsgren
A
,
Prymula
R
.
Prevention of otitis media: now a reality?
Vaccine
.
2009
;
27
(
42
):
5748
5754
13
World Bank Group
.
World development indicators.
2018
. Available at: https://datacatalog.worldbank.org/dataset/world-development-indicators. Accessed September 10, 2017
14
Harris
RP
,
Helfand
M
,
Woolf
SH
, et al;
Methods Work Group, Third US Preventive Services Task Force
.
Current methods of the US Preventive Services Task force: a review of the process
.
Am J Prev Med
.
2001
;
20
(
3
suppl
):
21
35
15
Sullivan
SG
,
Tchetgen Tchetgen
EJ
,
Cowling
BJ
.
Theoretical basis of the test-negative study design for assessment of influenza vaccine effectiveness
.
Am J Epidemiol
.
2016
;
184
(
5
):
345
353
16
Su
WJ
,
Lo
HY
,
Chang
CH
, et al
.
Effectiveness of pneumococcal conjugate vaccines of different valences against invasive pneumococcal disease among children in taiwan: a nationwide study
.
Pediatr Infect Dis J
.
2016
;
35
(
4
):
e124
e133
17
van der Linden
M
,
Falkenhorst
G
,
Perniciaro
S
,
Fitzner
C
,
Imöhl
M
.
Effectiveness of pneumococcal conjugate vaccines (PCV7 and PCV13) against invasive pneumococcal disease among children under two years of age in Germany
.
PLoS One
.
2016
;
11
(
8
):
e0161257
18
Verani
JR
,
Domingues
CM
,
de Moraes
JC
;
Brazilian Pneumococcal Conjugate Vaccine Effectiveness Study Group
.
Indirect cohort analysis of 10-valent pneumococcal conjugate vaccine effectiveness against vaccine-type and vaccine-related invasive pneumococcal disease
.
Vaccine
.
2015
;
33
(
46
):
6145
6148
19
Tregnaghi
MW
,
Sáez-Llorens
X
,
López
P
, et al;
COMPAS Group
.
Efficacy of pneumococcal nontypable Haemophilus influenzae protein D conjugate vaccine (PHiD-CV) in young Latin American children: a double-blind randomized controlled trial [published correction appears in PLoS Med. 2015;12(6):e1001850]
.
PLoS Med
.
2014
;
11
(
6
):
e1001657
20
Weinberger
R
,
van der Linden
M
,
Imöhl
M
,
von Kries
R
.
Vaccine effectiveness of PCV13 in a 3+1 vaccination schedule
.
Vaccine
.
2016
;
34
(
18
):
2062
2065
21
Moore
MR
,
Link-Gelles
R
,
Schaffner
W
, et al
.
Effectiveness of 13-valent pneumococcal conjugate vaccine for prevention of invasive pneumococcal disease in children in the USA: a matched case-control study
.
Lancet Respir Med
.
2016
;
4
(
5
):
399
406
22
Andrews
NJ
,
Waight
PA
,
Burbidge
P
, et al
.
Serotype-specific effectiveness and correlates of protection for the 13-valent pneumococcal conjugate vaccine: a postlicensure indirect cohort study
.
Lancet Infect Dis
.
2014
;
14
(
9
):
839
846
23
Pichichero
M
,
Kaur
R
,
Scott
DA
, et al
.
Effectiveness of 13-valent pneumococcal conjugate vaccination for protection against acute otitis media caused by Streptococcus pneumoniae in healthy young children: a prospective observational study
.
Lancet Child Adolesc Health
.
2018
;
2
(
8
):
561
568
24
Sáez-Llorens
X
,
Rowley
S
,
Wong
D
, et al
.
Efficacy of 10-valent pneumococcal non-typeable Haemophilus influenzae protein D conjugate vaccine against acute otitis media and nasopharyngeal carriage in Panamanian children - a randomized controlled trial
.
Hum Vaccin Immunother
.
2017
;
13
(
6
):
1
16
25
Vesikari
T
,
Forsten
A
,
Seppä
I
, et al
.
Effectiveness of the 10-valent pneumococcal nontypeable Haemophilus influenzae protein D-conjugated vaccine (PHID-CV) against carriage and acute otitis media-a double-blind randomized clinical trial in Finland
.
J Pediatric Infect Dis Soc
.
2016
;
5
(
3
):
237
248
26
Palmu
AA
,
Jokinen
J
,
Borys
D
, et al
.
Effectiveness of the ten-valent pneumococcal Haemophilus influenzae protein D conjugate vaccine (PHiD-CV10) against invasive pneumococcal disease: a cluster randomised trial
.
Lancet
.
2013
;
381
(
9862
):
214
222
27
Palmu
AA
,
Jokinen
J
,
Nieminen
H
, et al
.
Effectiveness of the ten-valent pneumococcal conjugate vaccine against tympanostomy tube placements in a cluster-randomized trial
.
Pediatr Infect Dis J
.
2015
;
34
(
11
):
1230
1235
28
Palmu
AA
,
Jokinen
J
,
Nieminen
H
, et al
.
Effect of pneumococcal Haemophilus influenzae protein D conjugate vaccine (PHiD-CV10) on outpatient antimicrobial purchases: a double-blind, cluster randomised phase 3-4 trial
.
Lancet Infect Dis
.
2014
;
14
(
3
):
205
212
29
Cohen
C
,
von Mollendorf
C
,
de Gouveia
L
, et al;
South African IPD Case-Control Study Group
.
Effectiveness of the 13-valent pneumococcal conjugate vaccine against invasive pneumococcal disease in South African children: a case-control study
.
Lancet Glob Health
.
2017
;
5
(
3
):
e359
e369
30
Broome
CV
,
Facklam
RR
,
Fraser
DW
.
Pneumococcal disease after pneumococcal vaccination: an alternative method to estimate the efficacy of pneumococcal vaccine
.
N Engl J Med
.
1980
;
303
(
10
):
549
552
31
de Oliveira
LH
,
Camacho
LA
,
Coutinho
ES
, et al
.
Impact and effectiveness of 10 and 13-valent pneumococcal conjugate vaccines on hospitalization and mortality in children aged less than 5 years in Latin American countries: a systematic review
.
PLoS One
.
2016
;
11
(
12
):
e0166736
32
Domínguez
A
,
Ciruela
P
,
Hernández
S
, et al
.
Effectiveness of the 13-valent pneumococcalconjugate vaccine in preventing invasive pneumococcal disease in children aged 7–59 months
.
PLoS One
.
2017
;
12
(
8
):
e0183191
33
Tan
TQ
.
Pediatric invasive pneumococcal disease in the United States in the era of pneumococcal conjugate vaccines
.
Clin Microbiol Rev
.
2012
;
25
(
3
):
409
419
34
Valenzuela
MT
,
Seoane
M
,
Canals
A
, et al
.
Laboratory surveillance of Streptococcus pneumoniae from invasive disease, Chile 2007-2012] [in Spanish]
.
Rev Chilena Infectol
.
2014
;
31
(
6
):
651
658
35
Kieninger
DM
,
Kueper
K
,
Steul
K
, et al;
006 Study Group
.
Safety, tolerability, and immunologic noninferiority of a 13-valent pneumococcal conjugate vaccine compared to a 7-valent pneumococcal conjugate vaccine given with routine pediatric vaccinations in Germany
.
Vaccine
.
2010
;
28
(
25
):
4192
4203
36
Puumalainen
T
,
Dagan
R
,
Wuorimaa
T
, et al
.
Greater antibody responses to an eleven valent mixed carrier diphtheria- or tetanus-conjugated pneumococcal vaccine in Filipino than in Finnish or Israeli infants
.
Pediatr Infect Dis J
.
2003
;
22
(
2
):
141
149
37
Togashi
T
,
Yamaji
M
,
Thompson
A
, et al;
3003 Study Group
.
Immunogenicity and safety of a 13-valent pneumococcal conjugate vaccine in healthy infants in Japan
.
Pediatr Infect Dis J
.
2013
;
32
(
9
):
984
989
38
GlaxoSmithKline. Synflorix: Pneumococcal Conjugate Vaccine (Non-Typeable Haemophilus influenzae (NTHi) Protein D, Diphtheria or Tetanus Toxoid Conjugates) Adsorbed. Mississauga, Ontario, Canada: GlaxoSmithKline;
2016
. Available at: https://ca.gsk.com/media/591956/synflorix.pdf. Accessed April 11, 2016
39
Ben-Shimol
S
,
Givon-Lavi
N
,
Leibovitz
E
,
Raiz
S
,
Greenberg
D
,
Dagan
R
.
Near-elimination of otitis media caused by 13-valent pneumococcal conjugate vaccine (PCV) serotypes in southern Israel shortly after sequential introduction of 7-valent/13-valent PCV
.
Clin Infect Dis
.
2014
;
59
(
12
):
1724
1732
40
Tomczyk
S
,
Lessa
FC
,
Sánchez
J
, et al
.
Effectiveness of 13-pneumococcal conjugate vaccine (PCV13) against invasive pneumococcal disease in children in the Dominican Republic
.
BMC Infect Dis
.
2018
;
18
:
152
41
Guevara
M
,
Ezpeletac
C
,
Gil-Setas
A
, et al
.
Reduced incidence of invasive pneumococcal disease after introduction of the 13-valentconjugate vaccine in Navarre, Spain, 2001–2013
.
Vaccine
.
2014
;
2553
2562

Competing Interests

POTENTIAL CONFLICT OF INTEREST: The authors have indicated they have no potential conflicts of interest to disclose.

FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.

Supplementary data