We elucidate the complexities involved in serological assessments in areas where multiple arboviruses co-circulate, underscoring the urgency for the developing of more refined and accurate methodologies in such contexts. Our study highlights that arboviral diseases are likely underestimated in endemic regions, emphasizing the need for comprehensive serological surveys to determine their actual extent. Additionally, the substantial underreporting of arboviral diseases in official records obscures the true extent of these burdens.
Our results indicate higher levels of antibodies against ZIKV and CHIKV in women compared to men. This disparity may be attributed to sociocultural practices wherein women are more likely to engage in domestic activities, thereby increasing their exposure to the vector [32]. In contrast, no significant differences between men and women in exposure to arboviruses[33, 34]. Additionally, significant differences were observed between age groups and educational levels. Consistent with previous studies[35–37] indicate older age and lower education, greater exposure to arboviruses.
During phase 2 of the study, coinciding with the COVID-19 pandemic, there was an increase in ZIKV IgM prevalence and a rise in borderline ZIKV IgG results. Concurrently, the prevalence of CHIKV IgM increased, coinciding with the transition to a different serological test manufacturer – Vircell Microbiologists. These findings may suggest that the widespread presence of anti-SARS-COV-2 IgG antibodies increased antigen affinity to arbovirus IgM antibodies [38], leading to false positives [39]. Additionally, an excess of IgG can competitively inhibit IgM binding to antigen epitopes, potentially causing false negatives [40, 41]. Therefore, it is crucial to pretreat samples with anti-IgG adsorbent in serological assays to mitigate this interference. The failure to remove IgG in Vircell Microbiologists kits may explain the significant prevalence of individuals appearing to be in the acute phase of Zika and chikungunya in our evaluation.
Regarding DENV, it is known that secondary infections are associated with increased susceptibility to severe dengue [42, 43]. Our findings indicate that between 5.7% and 9.0% of the population of São Sebastião may have been exposed to DENV more than once, suggesting the circulation of multiple viral subtypes in the area [44]. During the largest dengue epidemic in the history of the Federal District in 2024, a shift to the DENV-2 subtype occurred, which may be linked to increased severity and mortality [26].
In our study, Vircell Microbiologists kits exhibit compromised specificity and overestimate seroprevalence. We observed an almost 46% seropositivity rate for FLAV, with increased seropositivity for ZIKV IgG among individuals with DENV infection, suggesting cross-reactivity. Despite Vircell Microbiologists offering a broad array of immunodiagnostic tests for various diseases [45–51], options for arbovirus assessment remain limited. Most available studies primarily focus on ZIKV IgM [52–55].
According to the manufacturer's instructions [56], the antigens used for ZIKV in the Vircell Microbiologists kit consist of native proteins derived from purified lysates of viral cell cultures, specifically structural proteins. However, using native antigens may result in cross-reactivity in individuals with antibodies against other flaviviruses [57, 58]. Validation tests conducted by Vircell Microbiologists revealed that out of 10 samples positive for DENV, six exhibited cross-reactivity to ZIKV IgG. Similarly, among 15 samples positive for DENV, one showed cross-reactivity with ZIKV IgM [56]. Consequently, the use of structural proteins with lower specificity highlights the limitations of the Vircell Microbiologists kit in the accurate diagnosis of ZIKV in endemic regions.
Conversely, the prevalence of previous ZIKV infection, estimated at 14.9% (60/403), using the Euroimmun ZIKV IgG kit [59] appears to align more closely with previous reports [60–64]. The instructions for using the Euroimmun kit highlight that cross-reactivity with other arboviruses can be minimized due to recombinant non-structural proteins (NS1) as antigens. In fact, the use of NS1 proteins may confer greater specificity to the test, as these proteins are associated with the surface or secreted by infected cells [65, 66]. In independent evaluations, cross-reactions were not observed between cases positive for DENV [67, 68].
In summary, Euroimmun indicates that instead of cross-reactivity, co-infection should be considered based on the performance of the kit [69]. However, in our study, we observed that out of the Euroimmun ZIKV IgG positive samples, 59 out of 60 (98.3%) were also positive for DENV IgG. Other observations also highlight cross-reactivity in the manufacturer Euroimmun in dengue-positive samples [70, 71]. This raises questions about whether these individuals indeed had infections with both viruses or if the manufacturer's kit may also have limitations related to cross-reactions.
Our estimates should be interpreted cautiously due to a significant proportion of borderline results. Although manufacturers recommend retesting for indeterminate results after 1 to 2 weeks, logistical and resource constraints have made retesting impossible. In other assessments [60–62, 72], borderline results are typically excluded and not retested, hence rarely discussed. Here, the seroprevalence of antibody classes remained unchanged with the exclusion of borderline results, except for ZIKV IgG, as previously discussed. Therefore, we chose to include them in this study.
Importantly, all kits, except for the anti-ZIKV Euroimmun, were donated by the Ministry of Health and are part of the agency's official diagnostic scheme. This indicates that manufacturer variability may limit our study and health services, potentially impacting case notification.
Seropositivity rates for DENV (64.3%), ZIKV (51.4%), and CHIKV (5.4%) notably exceed those reported by local surveillance [26]. However, it is crucial to highlight that the estimated seroprevalence for DENV and ZIKV may be biased due to cross-reactions between flavivirus in the serological assays, as almost 46% of the samples showed dual positivity for DENV and ZIKV. Despite inaccuracies in comparing seroprevalence with surveillance data for DENV and ZIKV, CHIKV results, which show minimal cross-reaction, suggest that chikungunya cases could be up to 84 times higher than officially reported. This underreporting has been highlighted in other studies [60, 73], along with instances of misclassification, such as confirmed cases of CHIKV being erroneously categorized as probable cases of DENV [73, 74]. Furthermore, the substantial proportion of exposed individuals unaware of their infection underscores the significance of seroepidemiological studies and, notably, their advantages over monitoring reliant on traditional surveillance systems [61, 75].
This study represents the first comprehensive evaluation in the Central-West region of Brazil, presenting a robust and well-distributed sample size. Given that the sample size commonly used to validate diagnostic tests often falls short of capturing imperfections, particularly in areas where these arboviruses co-circulate, our research allowed for the identification and exhaustive discussion of the limitations of the serological tests employed. Despite this, we reaffirm serology as the most viable option regarding of cost-effectiveness. However, considering that the limitations of alternative tests are considerably more complex to resolve, the imperative developing of more specific serological tests is imminent. This urgency is highlighted by the vector's growing and resilient activity, which could potentially reintroduce ZIKV and CHIKV epidemics, as we estimate that at least 36,000 individuals remain susceptible to infection by these arboviruses in the assessed area. Finally, despite the meticulously outlined limitations, our results reveal alarming levels of underreporting in São Sebastião, which consequently reflects the entire Federal District.