Predicted serotype distribution in invasive pneumococcal disease (IPD) among children less than five years prior to the introduction of the Pneumococcal Conjugate Vaccine (PCV) in Nigeria

ABSTRACT Background: The 10-valent pneumococcal conjugate vaccine (PCV10) was introduced in Nigeria without any baseline data on serotype distribution in invasive pneumococcal disease (IPD). To estimate the proportion of IPD attributable to different serotypes, in children aged <5 years, we used statistical models based on the serotype-specific nasopharyngeal carriage prevalence and invasive capacity (IC). Methods: We used the carriage data from one urban and one rural setting in Nigeria, collected within five months of PCV10 introduction (2016). For Model A, we used serotype-specific adult case-fatality ratios from Denmark as proxy for IC. In the second model, we used the ratio of IPD proportions to carriage prevalence (case-carrier ratios) from Kenya (Model B) and the ratio of IPD incidence to carriage acquisition (attack rates) from the UK (Model C) as measures of serotype IC. Results: The models predict that serotypes with high carriage prevalence (6A, 6B, 19F and 23F) will dominate IPD. Additionally, Models B and C predictions emphasize serotypes 1, 4, 5, and 14, which were not prevalent in carriage but had high IC estimates. Non-PCV10 serotypes,6A and 19A, also dominated IPD predictions across models and settings. The predicted proportion of IPD attributed to PCV10 serotypes varied between 56% and 74% by model and setting. Conclusion: Carriage data can provide preliminary insights into IPD serotypes in settings that lack robust IPD data. The predicted PCV10-serotype coverage for IPD was moderately high. However, predictions for non-PCV10 serotypes indicate that higher-valency PCVs that cover serotypes 6A and 19A may have a larger impact on IPD reductions.