NS1-based IgG ELISA with OD ratio improves ZIKV and dengue serology in endemic regions

Researchers in Thailand have evaluated an in-house enzyme-linked immunosorbent assay designed to distinguish past Zika virus infections from past dengue virus infections in a population where dengue circulates widely. The study, published in Virology Journal, addresses a persistent problem in flavivirus serology: antibodies raised against one mosquito-borne flavivirus cross-react strongly with antigens from related viruses, making it difficult to interpret IgG-based tests in dengue-endemic regions. The team, led by Micaella C. Dato and Pornsawan Leaungwutiwong with colleagues from several Thai institutions, tested whether combining a specific viral antigen with a ratio-based interpretation of optical density readings could sharpen this discrimination.

The reference standard for confirming prior flavivirus exposure, plaque reduction neutralization testing, measures the ability of a patient’s antibodies to block viral infection of cells in culture. Although PRNT is regarded as the confirmatory assay, the authors note that it is labor-intensive and time-consuming, and that interpretation is itself complicated in flavivirus-endemic settings, where cross-reactive neutralizing antibodies generated by prior dengue exposure can obscure ZIKV-specific signals. Those limitations restrict its use in large-scale surveillance. The researchers therefore built their evaluation around an NS1-based IgG ELISA, a format that is faster and more scalable, and asked how well it could separate samples from patients whose acute infections had already been confirmed by RT-PCR as either Zika virus or dengue virus.

Three Zika virus antigens entered the comparison. The first was envelope protein domain III. The other two were versions of nonstructural protein 1: one derived from the African-lineage Zika strain MR766 and one from the Asian-lineage strain SV0127. Rather than reading each antigen’s signal in isolation, the team calculated three ZIKV/DENV ratio metrics for every sample: the positive-to-negative ratio, the optical density ratio, and the relative optical density ratio. Receiver operating characteristic curve analysis quantified how well each antigen-metric combination separated the RT-PCR-confirmed ZIKV samples from the DENV samples, with the area under the curve serving as the summary measure of discriminatory performance.

In the antigen selection phase, the Asian-lineage SV0127 NS1 antigen paired with the ZIKV/DENV relative OD ratio produced the highest area under the curve among the combinations tested. That pairing was carried forward into a diagnostic validation phase. Optimal cut-off values were chosen using Youden’s index, a standard method that balances sensitivity and specificity by maximizing their sum across possible thresholds. The validation set was then assessed with the selected antigen and the optical density ratio interpretation, and the resulting performance was benchmarked against a commercial Zika IgG ELISA using correlation analysis, paired t-tests, and Bland–Altman agreement analysis.

The validation results define both the promise and the limits of the approach. With a cut-off of 2.35 or higher on the ZIKV/DENV OD ratio, the in-house ELISA achieved an area under the curve of 0.7083, with a 95 percent confidence interval of 0.6335 to 0.7831. Sensitivity was 58.3 percent and specificity 79.2 percent, a balance the authors describe as even-handed rather than skewed toward either error type. Those figures are modest by the standards of a standalone diagnostic, and the study frames the assay accordingly: as a serological differentiation tool for surveillance and retrospective investigation, not as a replacement for neutralization-based confirmation.

The comparison with the commercial assay added an instructive nuance. Correlation analysis demonstrated a strong relationship between the optical density values generated by the in-house NS1 ELISA and the commercial Zika IgG ELISA, indicating that the two assays track the same underlying antibody signal. The commercial test produced slightly higher OD values overall, yet the in-house NS1-based format delivered better discriminatory performance between the RT-PCR-confirmed ZIKV and DENV samples. This suggests that the choice of antigen and, critically, the ratio-based interpretation of the dengue cross-reactive background contributed more to discrimination than raw signal strength did.

The ratio logic is the technical core of the study. In a dengue-endemic population, most adults carry flavivirus antibodies that bind Zika antigens to some degree, so an absolute Zika IgG reading alone cannot separate recent Zika infection from prior dengue exposure. By expressing each sample’s Zika signal relative to its dengue signal, the assay asks a different question: whether the antibody response is proportionally stronger against Zika than against dengue. The Youden-derived threshold of 2.35 sets how large that proportional excess must be before a sample is called Zika-reactive. Because the threshold was optimized on the study’s own samples, its performance in other populations with different dengue exposure histories and infection sequences would need independent verification.

The authors position the validated combination as a resource-conscious strategy for settings where neutralization assays are impractical. Epidemiological surveillance programs and retrospective serological investigations often require processing large sample volumes, a scale at which PRNT’s cell-culture demands become prohibitive. An ELISA that incorporates a dengue-referenced ratio into its interpretation offers a way to extract more epidemiological signal from routine IgG testing without new instrumentation. The study’s scope should be kept in view: the evaluation used RT-PCR-confirmed samples from a single dengue-endemic setting, the reported sensitivity leaves a substantial share of Zika cases undetected at the chosen cut-off, and the assay measures IgG rather than acute-phase markers, so it speaks to exposure history rather than current infection. Within those boundaries, the work provides a quantified benchmark for NS1-based ratio serology and a template for further optimization of flavivirus differentiation in endemic regions.

Subject of Research: Evaluation of an NS1-based ZIKV/DENV IgG ELISA with optical density ratio interpretation for serological differentiation of Zika and dengue virus infections in a dengue-endemic setting

Article Title: Evaluation of an NS1-Based ZIKV/DENV IgG ELISA approach for serological differentiation in a dengue-endemic setting

Article References: Dato, M. C., Sittikul, P., Sriburin, P., Rattanamahaphoom, J., Maneekan, P., Jangpatarapongsa, K., Chatchen, S., & Leaungwutiwong, P. (2026). Evaluation of an NS1-Based ZIKV/DENV IgG ELISA approach for serological differentiation in a dengue-endemic setting. Virology Journal. https://doi.org/10.1186/s12985-026-03340-w

Image Credits: AI Generated

DOI: 10.1186/s12985-026-03340-w

Keywords: Zika virus, dengue virus, IgG ELISA, NS1 antigen, serological differentiation, flavivirus cross-reactivity, optical density ratio, ROC analysis, diagnostic validation, dengue-endemic setting, PRNT, surveillance