Innovative methodological approach to assess dengue transmission: findings from the SERODEN study in Africa

Abstract

Introduction

A limited number of dengue seroprevalence surveys have been conducted across Africa, with only 21 studies reported to date. Implementing new surveys to assess dengue transmission can be costly, time-consuming and resource intensive. In SERODEN we developed new simulation-based methods to calculate the optimal number and age-distribution of existing blood samples –collected in the context of previous household and community-based surveys, most notably against SARS-CoV-2 – to be tested for dengue. We used three different assays, namely the enzyme-linked immunosorbent assay (ELISA) IgG type 1-4, ELISA IgG NS1 type 1-4 and Plaque Reduction Neutralization Test (PRNT) for the 4 dengue serotypes, to characterise age-specific seroprevalence and dengue transmission intensity in 19 locations across Senegal, the Democratic Republic of the Congo (DRC), and Ghana.

Method

We designed a simulation-based framework to inform serosurvey design when leveraging existing blood samples. We also developed a Bayesian approach to combine the results obtained from the different assays in unifying age-stratified seroprevalence and force of infection estimates, estimating and accounting for the tests’ specificity and sensitivity.

Results

Our simulation-based framework identified the optimal sample sizes and age-distribution of the available blood samples, and overall, we reduced the total number of samples required for testing by 20%. The age-group prioritised for testing depended on the expected transmission intensity, with younger age-groups targeted for testing in high transmission settings. By combining multiple tests, including IgG ELISA type 1-4, ELISA IgG NS1, and PRNTs, we were able to quantify the specificity and sensitivity of each test and the dengue transmission in different African settings.

Discussion

Our results unveiled significant heterogeneity in dengue transmission both within and across countries and underscored the endemic nature of dengue transmission in Senegal, DRC, and Ghana. The methods developed in the SERODEN study demonstrates the feasibility and benefits of utilising existing blood samples for the implementation of dengue serosurveys.

Conclusion

By leveraging existing resources and combining different tests, we can provide valuable insights into dengue transmission intensity in Africa, which sheds new light on the dengue infection burden and can help inform dengue surveillance.