Serological testing is important for assessing past exposure and immunity, but interpretation can be complicated by antibody cross-reactivity between closely related viruses. We assess this challenge for Usutu virus (USUV) and West Nile virus (WNV), flaviviruses that recently emerged in Europe. We analysed samples from wild blackbirds collected in the Netherlands between 2016-2022. Samples (N=1742) were screened using an NS1-based protein micro-array, with positives confirmed by Focus Reduction Neutralization Tests (FRNT). We jointly estimated seroprevalence and antibody responses by fitting a Bayesian latent-variable model to FRNT values. Estimates of homologous and cross-reactive antibody responses were used to improve interpretation of observed titres for serosurveillance. Estimated seroprevalence varied across time and regions between 4.9% (95%CrI 3.5-6.7) to 18.5% (95%CrI 14.9-22.7) for USUV and between 2.4% (95%CrI 1.3-3.8) to 6.4% (95%CrI 3.9-9.6) for WNV. These were 1.5 (USUV) to 2.4 times (WNV) higher than estimates based on the current threshold-based algorithm. USUV induced a higher antibody response and was more likely to induce a cross-reactive response than WNV. Our classification algorithm informed by these estimates showed high sensitivity (WNV: 0.88, USUV: 0.97) and specificity (both: >0.99). Our results illustrate how quantitative frameworks can improve serological interpretation in settings with co-circulating pathogens.
de Wit, M. M., Hoze, N., De Jong, M. C. M., Koopmans, M., van Mastrigt, T., Sikkema, R., ten Bosch, Q.
Advertisement
Stats
- Recommendations n/a n/a positive of 0 vote(s)
- Views 9
- Comments 0