There was no evidence of a confounding effect of schedule on VEC (with the coefficient β2 centered around 0 [−0.03 (95% CrI: −0.32; 0.63)] or that the waning rate differed by schedule [interaction term β3 0.01 (95% CrI: −0.24; 0.13)]. However, taken individually, the median waning coefficient β1 was smaller (ie, “flatter” slope) after a booster than after a 3 + 0 schedule (Fig., Supplemental Digital Content 4, http://links.lww.com/INF/C161 and Table 2). Our model was insensitive to the assumption on the prior of β1.
The serotype distribution among the unvaccinated children was fairly stable across studies (see Fig., Supplemental Digital Content 5, http://links.lww.com/INF/C162), with little or moderate statistical heterogeneity in the distribution of serotypes among PCV7 positive samples (serotype-specific I2 values of heterogeneity ranging from 0% to 60%). Serotypes 6B, 23F and 19F were the VT most commonly found, contributing to 26%, 22% and 28%, respectively of the isolated PCV7 serotypes overall among unvaccinated children. Serotype 14 was found in 11% and serotypes 4, 9V and 18C in 3%, 6% and 3% of PCV7 samples, respectively. Serotype 6A was found in about 9% of unvaccinated children, a little higher than the prevalence of 6B (8%, P = 0.07).
Efficacy estimates differed across PCV7 serotypes. Six months after vaccination the highest VEC was measured for serotypes 4 (80%) and 9V (79%) and the lowest for 19F (38%; Fig. 2 and Table 2).
The decline in the efficacy over time varied by serotype (Table 2), with the slowest decline for serotypes 23F and 19F (median β1, 0.09) and more rapid declines for rarer serotypes, although CrIs overlapped for all serotypes (Table 1).
We found evidence of protection against 6A, with a peak VEC of 48% (95% CrI: 18–72%), decreasing to 0 within 5 years postvaccination (Table 2 and Fig. 2).
Our sensitivity analysis showed no significant impact of any study estimate on the coefficients. Estimates were similar after excluding the cluster randomized trial,10,33 with a VEC of 62% (51–73%) at 4 months, decreasing to 40% (12–54%) at 5 years. Excluding the 2 estimates from Cheung et al29 in the Gambia, which together accounted for about 28% of all children included in the analysis, did not affect model estimates [VEC of 62% (50–74%) at 4 months and 39% (12–54%) at 5 years]. Finally, overall and booster schedule VEC estimates and respective model coefficients were similar with and without data from the PCV10 trial.26,27
We explored 2 other models of waning, in addition to the main model (Appendix, Supplemental Digital Content 2, http://links.lww.com/INF/C159). In all 3 models, there was good evidence of protective efficacy in the first few years after vaccination. A similar DIC was obtained for all 3 models estimating the aggregate VEC, as well as for serotype-specific models, except for serotypes 14 and 19F for which the model with the asymptotic time function was outperformed by the other 2. Further information can be found in Figure, Supplemental Digital Content 5, http://links.lww.com/INF/C163 and Table, Supplemental Digital Content 6, http://links.lww.com/INF/C164.
We computed pooled aggregate and PCV7 serotype-specific vaccine efficacy against NP acquisition and its waning based on a meta-regression model of cross-sectional data. Our results suggest that PCVs confers reasonable protection against acquisition of pneumococcal carriage of the 7 studies serotypes, for several years after vaccination, albeit with differences across serotypes.
Previous studies have explored PCV efficacy against carriage16 and compared schedules;5 however, a pooled estimate was not previously calculated. We found that the distribution of VT was relatively stable across settings, making the pooling of aggregate estimates possible despite differences in the VEC against individual serotypes. An analysis of the aggregate VEC based on the pooled individual serotype-specific estimates also showed results in line with that of the main model presented, albeit with wider confidence bounds given the uncertainty around serotype-specific estimates.
Three serotypes (6B, 19F and 23F) accounted for about 75% of all PCV7 serotypes, but the VEC for each of those differed, with high efficacy against 6B and a weaker anti-19F efficacy. A possible reason for this divergence is the difference in the amount of antibody required for protection as well as differences in the vaccine-induced opsonophagocytic activity (ie, the ingestion and killing of pathogens by phagocytes), despite similar antibody geometric mean concentrations following PCV7 vaccination.30,38 Interestingly, a recent study in the United Kingdom on the vaccine effectiveness and immune correlates of protection against invasive pneumococcal disease39 showed that much less antibody is required for 6B and 23F protection than for 19F protection. The polysaccharide capsule of 19F is more resistant to complement deposition than 6B and requires higher levels of antibodies for opsonophagocytosis.38 However, although trials37,40,41 have shown persistence of serum antibodies several years after vaccination, the exact mechanism underlying the protection against acquisition of carriage remains unclear. Such mechanisms could perhaps involve mucosal immunological responses42 in addition to preexisting circulating serum immunoglobulin (IgG), with serological markers incompletely capturing the mucosal response.
Although natural immunity to colonization in infancy is poor, conjugate vaccines stimulate B-cell responses and the generation of memory B-cells,43 which can be naturally boosted. If boosting does contribute to maintaining a protective efficacy against carriage, then one might expect efficacy to wane faster for rarer serotypes and slower for the more prevalent ones. Our results support such a hypothesis to some extent, showing a slower VEC decline for the most prevalent serotypes. This would also mean that VEC may wane more rapidly after routine implementation of the vaccine than in trial conditions.
We also found evidence of cross-protective efficacy against 6A acquisition based on data from PCV7 and PCV9 trials. Such evidence is supported by trials and observational studies showing an impact of PCV7 on 6A disease,44 as well as immunological evidence with the vaccine eliciting functional antibody (ie, antibodies inducing opsonophagocytosis) against 6A.45
Efficacy estimates against carriage and their waning have several implications for vaccination programs. Despite a stable distribution of serotypes across studies in this analysis, it is likely that some geographical variation occurs, and serotype-specific efficacies are, therefore, important in predicting the impact of PCV in various epidemiological settings. Our results show good evidence of a direct protection against carriage in the first 5 years of life, when the pneumococcal burden is particularly high, and vaccinated children, therefore, also contribute to reducing transmission for several years. This may be particularly important in settings with low vaccine uptake or interrupted delivery.
The population impact on carriage and disease of introducing PCVs under different epidemiological scenarios could be explored through dynamic transmission models of pneumococcal disease, and our estimates of VEC and their waning are essential parameters for such models.
The direct impact on disease is not solely conditioned on the VEC, but also on the efficacy of the vaccine against progressing to disease following carriage.1 This explains the higher efficacy of PCV against invasive disease, at around 80%.1,3,39 In contrast, the efficacy on disease progression against mucosal forms of disease, such as acute otitis media (AOM), is small with most of the disease impact predicted by VEC only.1,3 Interestingly, the efficacy against pneumococcal AOM diagnosed by myringotomy with middle ear fluid aspiration among Finnish children enrolled in a large PCV7 trial30 was 62% (48–72%) in the year following the booster dose, and serotype-specific efficacies were lowest for serotype 19F, at about 37%, and high for 6B (79%), 4 (75%) and 9V (82%). Those estimates are similar to our aggregate and serotype-specific VEC estimates, adding to the evidence that VEC is a close measure of the efficacy against AOM.
An important question is the applicability of our results to 10-valent and 13-valent vaccines, given that many countries have introduced—or are planning to do so—those vaccines into their routine vaccination programs. Data on immunological correlates of protection from trials generally suggest comparable responses after PCV13 and PCV10 vaccination compared with PCV7.45–47 However, a recent study comparing IgG concentration and functional antibodies in PCV7 and PCV13 vaccinated Navajo and White Mountains Apache children in the US48 found higher functional antibody activity against 19F after PCV13 vaccination, compared with PCV7, possibly explained by the inclusion of 19A in PCV13 and the additional activity of anti-19A antibodies against 19F. Another trial in the UK showed lower IgG concentrations for serotypes 4 and 23F for PCV13 compared with PCV7 recipients after a booster dose at 12 months of age. Those differences could translate in differences in aggregate VEC.
The estimation of the efficacy against carriage acquisition from cross-sectional data relies on several assumptions, the most important being that of stationarity—ie, that the relationship between carriage incidence and carriage prevalence is stable.15,16 Vaccination will introduce some temporary disturbance in the carriage rates of different serotypes, with the average prevalence estimates stabilizing after some time.49 Auranen et al17 suggest that stationary levels should not be considered before at least twice the duration of carriage since vaccination. We included studies from 4 months after vaccination to account for this, which we considered to be a good trade-off between ensuring steady-state carriage levels and avoiding peak estimates to be affected by waning VEC.
The assumption that PCV do not affect clearance is based on limited evidence.9,10 Similarly, studies have suggested that the vaccine may also impact carriage density.10 In both scenarios (reduced duration and reduced density), VEC could represent a combined efficacy estimate against acquisition and transmission under the assumption that a reduction in duration of carriage and/or carriage density is associated with both a reduction in the likelihood of detection and of transmission, as discussed elsewhere.15,16
Our study has a number of additional limitations.
First, our analysis was limited by the number of data points, with wider uncertainty as time since vaccination increases and the smaller study sizes for serotype specific analyses, with substantial uncertainty around model estimates for the least prevalent serotypes. The small number of data points in each schedule subgroup may have limited our ability to detect any difference between schedules. In addition, we were unable to provide concise predictions about the waning later than 5 years after completion of the vaccination schedule. This was because of the lack of data and the absence of strong statistical evidence of the superiority of the main model of waning presented compared with the 2 other models tested.
Second, studies were based on the identification of the dominant serotype in single colonies, and multiple colonization was not taken into account. If the prevalence of multiple colonization is low and if there are no differences in the propensity of detecting 1 serotype over another, VEC estimates based on single colonization would nonetheless adequately capture VEC.15
There are several other factors related to vaccine schedules and delivery that may impact on VEC (and on the heterogeneity between studies), which we were unable to explore, including the timing and spacing of doses and the coadministration of PCV with different childhood vaccines.46 For example, a recent systematic review of the impact of PCV vaccination schedules on immunological responses46 suggests that immune responses to serotype 14 may be influenced by coadministration of PCV with DTP vaccines, with significantly higher geometric mean concentrations observed with acellular pertussis compared with the whole cell pertussis vaccine.
In addition, although the description of the swabbing and sample processing techniques used in the studies included—although sometimes limited—seem to conform to WHO guidelines,50 we cannot rule out that some of the between-study heterogeneity may be because of differences in such techniques.
Finally, further research to obtain more precise estimates of VEC after noncomplete schedules, particularly single catch-up doses, is warranted. This is particularly relevant in the context of PCV roll out in low-income settings, as some countries may opt for catch-up campaigns at the introduction of the vaccine.
In conclusion, through this study, we provide consistent evidence for a lasting efficacy against carriage acquisition of PCV in children during the first few years after completion of vaccination, although with differences in efficacy and duration of protection between serotypes.
We would like to thank Shabir Madhi, Ron Dagan, Noga Givon-Lavi, Adam Finn and Arto Palmu for providing us with summary data from their studies.
1. Simell B, Auranen K, Käyhty H, et al.Pneumococcal Carriage
Group. The fundamental link between pneumococcal carriage
and disease. Expert Rev Vaccines. 2012;11:841–855
2. Davis S, Deloria-Knoll M, Kassa HT, et al. Impact of pneumococcal conjugate vaccines
on nasopharyngeal carriage
and invasive disease among unvaccinated people: Review of evidence on indirect effects. Vaccine. 2013;32:133–145
3. Goldblatt D, Ramakrishnan M, O’Brien K.. Using the impact of pneumococcal vaccines on nasopharyngeal carriage
to aid licensing and vaccine implementation; a PneumoCarr meeting report March 27–28, 2012, Geneva. Vaccine. 2013;32:146–152
4. Weinberger DM, Bruden DT, Grant LR, et al. Using pneumococcal carriage
data to monitor postvaccination changes in invasive disease. Am J Epidemiol. 2013;178:1488–1495
5. Fleming-Dutra KE, Conklin L, Loo JD, et al. Systematic review of the effect of pneumococcal conjugate vaccine dosing schedules on vaccine-type nasopharyngeal carriage
. Pediatr Infect Dis J. 2014;33(Suppl 2):S152–S160
6. Dagan R, Muallem M, Melamed R, et al. Reduction of pneumococcal nasopharyngeal carriage
in early infancy after immunization with tetravalent pneumococcal vaccines conjugated to either tetanus toxoid or diphtheria toxoid. Pediatr Infect Dis J. 1997;16:1060–1064
7. Obaro SK, Adegbola RA, Chang I, et al. Safety and immunogenicity of a nonavalent pneumococcal vaccine conjugated to CRM197 administered simultaneously but in a separate syringe with diphtheria, tetanus and pertussis vaccines in Gambian infants. Pediatr Infect Dis J. 2000;19:463–469
8. Dagan R, Zamir O, Tirosh N, et al. Nasopharyngeal (NP) carriage
of Streptococcus pneumoniae
(Pnc) in toddlers vaccinated during infancy with an 11 valent pneumococcal vaccine conjugated to diphtheria and tetanus toxoids (PCV-DT). Abstracts of the Interscience Conference on Antimicrobial Agents & Chemotherapy. 2000;40:236
9. Dagan R, Givon-Lavi N, Zamir O, et al. Reduction of nasopharyngeal carriage
of Streptococcus pneumoniae
after administration of a 9-valent pneumococcal conjugate vaccine to toddlers attending day care centers. J Infect Dis. 2002;185:927–936
10. O’Brien KL, Millar EV, Zell ER, et al. Effect of pneumococcal conjugate vaccine on nasopharyngeal colonization among immunized and unimmunized children in a community-randomized trial. J Infect Dis. 2007;196:1211–1220
11. Abdullahi O, Karani A, Tigoi CC, et al. Rates of acquisition and clearance of pneumococcal serotypes in the nasopharynges of children in Kilifi District, Kenya. J Infect Dis. 2012;206:1020–1029
12. Lipsitch M, Abdullahi O, D’Amour A, et al. Estimating rates of carriage
acquisition and clearance and competitive ability for pneumococcal serotypes in Kenya with a Markov transition model. Epidemiology. 2012;23:510–519
13. Turner P, Turner C, Jankhot A, et al. A longitudinal study of Streptococcus pneumoniae carriage
in a cohort of infants and their mothers on the Thailand-Myanmar border. PLoS One. 2012;7:e38271
14. Andrews N, Waight PA, Borrow R, et al. Using the indirect cohort design to estimate the effectiveness of the seven valent pneumococcal conjugate vaccine in England and Wales. PLoS One. 2011;6:e28435
15. Auranen K, Rinta-Kokko H, Halloran ME.. Estimating strain-specific and overall efficacy
of polyvalent vaccines against recurrent pathogens from a cross-sectional study. Biometrics. 2013;69:235–244
16. Rinta-Kokko H, Dagan R, Givon-Lavi N, et al. Estimation of vaccine efficacy
against acquisition of pneumococcal carriage
. Vaccine. 2009;27:3831–3837
17. Auranen K, Rinta-Kokko H, Goldblatt D, et al.Pneumococcal Carriage
Group (PneumoCarr). Colonisation endpoints in Streptococcus pneumoniae
vaccine trials. Vaccine. 2013;32:153–158
18. Higgins JP, Thompson SG, Deeks JJ, et al. Measuring inconsistency in meta-analyses. BMJ. 2003;327:557–560
19. Russell FM, Carapetis JR, Satzke C, et al. Pneumococcal nasopharyngeal carriage
following reduced doses of a 7-valent pneumococcal conjugate vaccine and a 23-valent pneumococcal polysaccharide vaccine booster. Clin Vaccine Immunol. 2010;17:1970–1976
20. Lakshman R, Murdoch C, Race G, et al. Pneumococcal nasopharyngeal carriage
in children following heptavalent pneumococcal conjugate vaccination in infancy. Arch Dis Child. 2003;88:211–214
21. Spiegelhalter DJ, Best NG, Carlin BP, et al. Bayesian measures of model complexity and fit. J Roy Stat Soc B. 2002;64
23. Watson K, Carville K, Bowman J, et al.Kalgoorlie Otitis Media Research Project Team. Upper respiratory tract bacterial carriage
in Aboriginal and non-Aboriginal children in a semi-arid area of Western Australia. Pediatr Infect Dis J. 2006;25:782–790
24. Prymula R, Kriz P, Kaliskova E, et al. Effect of vaccination with pneumococcal capsular polysaccharides conjugated to Haemophilus influenzae-derived protein D on nasopharyngeal carriage
of Streptococcus pneumoniae
and H. influenzae in children under 2 years of age. Vaccine. 2009;28:71–78
25. Kilpi T, Syrjänen R, Palmu A, et al. Parallel evaluation of 7-valent PncCRM on pneumococcal carriage
and AOM. In: European Society for Paediatric Infectious Diseases (ESPID) Conference. 2001
26. Prymula R, Hanovcova I, Splino M, et al. Impact of the 10-valent pneumococcal non-typeable Haemophilus influenzae
Protein D conjugate vaccine (PHiD-CV) on bacterial nasopharyngeal carriage
. Vaccine. 2011;29:1959–1967
27. Usuf E, Bottomley C, Adegbola RA, et al. Pneumococcal carriage
in sub-Saharan Africa–a systematic review. PLoS One. 2014;9:e85001
28. Dagan R, Givon-Lavi N, Porat N, et al. The effect of an alternative reduced-dose infant schedule and a second year catch-up schedule with 7-valent pneumococcal conjugate vaccine on pneumococcal carriage
: a randomized controlled trial. Vaccine. 2012;30:5132–5140
29. Cheung YB, Zaman SM, Nsekpong ED, et al. Nasopharyngeal carriage
of Streptococcus pneumoniae
in Gambian children who participated in a 9-valent pneumococcal conjugate vaccine trial and in their younger siblings. Pediatr Infect Dis J. 2009;28:990–995
30. Kilpi T, Ahman H, Jokinen J, et al.Finnish Otitis Media Study Group. Protective efficacy
of a second pneumococcal conjugate vaccine against pneumococcal acute otitis media in infants and children: randomized, controlled trial of a 7-valent pneumococcal polysaccharide-meningococcal outer membrane protein complex conjugate vaccine in 1666 children. Clin Infect Dis. 2003;37:1155–1164
31. Madhi SA, Adrian P, Kuwanda L, et al. Long-term effect of pneumococcal conjugate vaccine on nasopharyngeal colonization by Streptococcus pneumoniae
—and associated interactions with Staphylococcus aureus and Haemophilus influenzae colonization—in HIV-Infected and HIV-uninfected children. J infect Dis. 2007;196:1662–1666
32. Mbelle N, Huebner RE, Wasas AD, et al. Immunogenicity and impact on nasopharyngeal carriage
of a nonavalent pneumococcal conjugate vaccine. J infect Dis. 1999;180:1171–1176
33. Millar EV, O’Brien KL, Watt JP, et al. Effect of community-wide conjugate pneumococcal vaccine use in infancy on nasopharyngeal carriage
through 3 years of age: a cross-sectional study in a high-risk population. Clin Infect Dis. 2006;43:8–15
34. Palmu AA, Verho J, PH M.. Long-term efficacy
of the seven valent PncCRM vaccine on nasopharyngeal carriage
. In: 3rd International Symposium on Pneumococci and Pneumcooccal Disease (ISPPD-3). 2002 Anchorage, Alaska;
35. van Gils EJ, Veenhoven RH, Hak E, et al. Effect of reduced-dose schedules with 7-valent pneumococcal conjugate vaccine on nasopharyngeal pneumococcal carriage
in children: a randomized controlled trial. JAMA. 2009;302:159–167
36. Yeh SH, Zangwill KM, Lee H, et al. Heptavalent pneumococcal vaccine conjugated to outer membrane protein of Neisseria meningitidis serogroup b and nasopharyngeal carriage
of Streptococcus pneumoniae
in infants. Vaccine. 2003;21:2627–2631
37. Madhi SA, Adrian P, Kuwanda L, et al. Long-term immunogenicity and efficacy
of a 9-valent conjugate pneumococcal vaccine in human immunodeficient virus infected and non-infected children in the absence of a booster dose of vaccine. Vaccine. 2007;25:2451–2457
38. Ekstrom N, Vakevainen M, Verho J, et al. Functional antibodies elicited by two heptavalent pneumococcal conjugate vaccines
in the Finnish Otitis Media Vaccine Trial. Infect Immun. 2007;75:1794–1800
39. 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;14:839–846
40. Ekström N, Ahman H, Palmu A, et al.FinOM Study Group. Concentration and high avidity of pneumococcal antibodies persist at least 4 years after immunization with pneumococcal conjugate vaccine in infancy. Clin Vaccine Immunol. 2013;20:1034–1040
41. Trück J, Snape MD, Tatangeli F, et al. Pneumococcal serotype-specific antibodies persist through early childhood after infant immunization: follow-up from a randomized controlled trial. PLoS One. 2014;9:e91413
42. Zhang Q, Bernatoniene J, Bagrade L, et al. Regulation of production of mucosal antibody to pneumococcal protein antigens by T-cell-derived gamma interferon and interleukin-10 in children. Infect Immun. 2006;74:4735–4743
43. Kelly DF, Moxon ER, Pollard AJ.. Haemophilus influenzae type B conjugate vaccines
. Immunology. 2004;113:163–174
44. Whitney CG, Pilishvili T, Farley MM, et al. Effectiveness of seven-valent pneumococcal conjugate vaccine against invasive pneumococcal disease: a matched case-control study. Lancet. 2006;368:1495–1502
45. Esposito S, Tansey S, Thompson A, et al. Safety and immunogenicity of a 13-valent pneumococcal conjugate vaccine compared to those of a 7-valent pneumococcal conjugate vaccine given as a three-dose series with routine vaccines in healthy infants and toddlers. Clin Vaccine Immunol. 2010;17:1017–1026
46. Park DE, Johnson TS, Nonyane BA, et al. The differential impact of coadministered vaccines, geographic region, vaccine product and other covariates on pneumococcal conjugate vaccine immunogenicity. Pediatr Infect Dis J. 2014;33(Suppl 2):S130–S139
47. Snape MD, Klinger CL, Daniels ED, et al. Immunogenicity and reactogenicity of a 13-valent-pneumococcal conjugate vaccine administered at 2, 4, and 12 months of age: a double-blind randomized active-controlled trial. Pediatr Infect Dis J. 2010;29:e80–e90
48. Grant LR, O’Brien SE, Burbidge P, et al. Comparative immunogenicity of 7 and 13-valent pneumococcal conjugate vaccines
and the development of functional antibodies to cross-reactive serotypes. PLoS One. 2013;8:e74906
49. Auranen K, Rinta-Kokko H, Goldblatt D, et al. Design questions for Streptococcus pneumoniae
vaccine trials with a colonisation endpoint. Vaccine. 2013;32:159–164
50. O’Brien KL, Moulton LH, Reid R, et al. Efficacy
and safety of seven-valent conjugate pneumococcal vaccine in American Indian children: group randomised trial. Lancet. 2003;362:355–361
pneumococcus; conjugate vaccines; efficacy; carriage
Supplemental Digital Content
Copyright © 2015 Wolters Kluwer Health, Inc. All rights reserved.