Relative importance of S. pneumoniae serotypes causing invasive pneumococcal disease in young children after the introduction of conjugate vaccines

Abstract

Introduction: Pneumococcal diseases include a range of infections caused by S. pneumoniae, which is a common cause of morbidity and mortality globally. Even though pneumococcal conjugate vaccines (PCVs) are available, invasive pneumococcal disease (IPD) remains a global public health issue in young children. Given the large diversity of pneumococcal serotypes and differences in their prevalence geographically, it is important to assess and quantify their burden in a comprehensive way to inform global health interventions. Aims and objectives: This thesis aims to characterise the relative burden of S. pneumoniae serotypes in childhood in IPD in settings where PCVs have been implemented, with a focus on serotypes not included in current formulations, to inform research and future prevention strategies. Methods: I conducted two systematic reviews to answer questions related to the change in the incidence of IPD in young children, the proportional contribution of serotypes to IPD, and the relative invasive disease potential of circulating serotypes after the introduction of PCVs. Data on the first two outcomes were identified from published literature, complemented by grey literature and they were pooled through meta-analysis. Serotype data from carriage and IPD studies to estimate the relative invasive disease potential were obtained through a systematic review and from collaborators who provided re-analysed or an extension of published data. Then, data were pooled using meta-analysis. To examine the impact of PCVs on the incidence of IPD, by serotype categories, among young children in Latin America I extracted and pooled data from annual SIREVA-II reports using meta-analysis. Results: The introduction of highly valent PCVs (PCV13 and/or PCV10) has resulted in a consistent reduction of IPD incidence among young children. The decreases in the incidence of IPD have been driven by declines of cases due to vaccine-targeted serotypes. Less consistently, there is evidence of a potential lag between PCV introduction and increases of non-PCV13 serotypes. Based on data from varying years after the implementation of highly valent PCVs, a pooled analysis indicated that less than half of childhood IPD cases were found to be associated with non-PCV13 serotypes, with differences by region. Due to the high heterogeneity, these estimates should be interpreted with caution. The estimates of invasive disease potential of non-PCV13 serotypes were usually lower than that of 19A or other vaccine types. However, there is evidence to support that some non-PCV13 serotypes are more invasive than other serotypes not included in PCVs. When focusing on the Latin American region, a large, yet variable, protective effect of highly valent PCVs has been identified on childhood IPD in seven countries. Regional estimates for the different serotype categories are difficult to interpret due to heterogeneity but the average incidence of IPD associated with PCV7 and PCV10non7 has declined by over 70% after three years of highly valent PCV use when compared to the year before their introduction. The moderate heterogeneity in the estimates of these reductions (I 2 0-49%) offer support for the protective effects of PCVs against targeted serotypes. Conclusion: Incidence data of childhood IPD provide evidence for serotype replacement post-highly valent PCVs, which has not yet been sufficient to offset the protective effect of the vaccines. As vaccination programmes mature, the relative contribution of serotypes to childhood IPD is changing, with non-PCV13 serotypes accounting for a third (or more) of IPD cases in some settings with over five years of use. Estimates of invasive disease potential for non-PCV13 serotypes show that most have a low invasive disease potential. In Latin America region there is evidence of reductions in IPD due to vaccine-targeted serotypes by the third year of highly valent PCVs use. The estimates presented in this thesis show there is a need to monitor the role of non-PCV13 serotypes in IPD closely. It is not possible, as of yet, to know which, if any, of these serotype(s) could emerge as the leading serotype in IPD or have an impact that would mitigate PCVs’ high protective effects. Globally, PCV programmes need to be introduced, sustained and expanded where possible to reduce the burden of pneumococcal disease. Close attention to the diversity of serotypes that are associated with IPD in different world regions is warranted to inform the development and implementation of future pneumococcal disease vaccines.