Sensitivity and negative predictive value for a rapid dengue test

01 May 2019
Flasche S, Smith PG,

In 2016, WHO recommended Dengvaxia, the first-ever licensed dengue vaccine, for use in children aged 9 years or older residing in high-burden settings. The vaccine had shown efficacy against all four dengue serotypes in two large phase 3 trials.

 A safety signal of an excess risk of severe dengue was detected in vaccinated children aged 2–5 years.

 This finding prompted further analyses with a new immunological assay to assess whether the risk was associated with age per se or was due to a higher proportion of children in this age group having had no previous dengue infections. These analyses

showed that although the vaccine offered substantial protection among children who had been previously infected with dengue, dengue-naive vaccinees were at increased risk of dengue hospitalisation and severe dengue during the 5-year trial follow-up compared with similar children in the placebo group. This effect had been previously postulated in mathematical models.

In response to this new evidence, WHO revised their previous advice and now recommends a pre-vaccination screening strategy to avoid the vaccination of dengue-naive children.

 In the absence of a licensed rapid diagnostic test to detect previous dengue infection, discussions around a target product profile for such a test have focused on the need for high test specificity to limit the number of false-positive test results, and thus minimise the risk of vaccinating dengue-naive children. However, increasing the specificity of a test typically decreases its sensitivity, and a test with poor sensitivity would leave many unvaccinated who would have potentially benefited from vaccination.