Serological inference of past primary and secondary dengue infection: implications for vaccination

Reconciling heterogeneous dengue virus infection risk estimates from different study designs

Uncovering rates at which susceptible individuals become infected with a pathogen, i.e., the force of infection (FOI), is essential for assessing transmission risk and reconstructing distribution of immunity in a population. For dengue, reconstructing exposure and susceptibility statuses from the measured FOI is of particular significance as prior exposure is a strong risk factor for severe disease. FOI can be measured via many study designs. Longitudinal serology is considered gold standard measurements, as they directly track the transition of seronegative individuals to seropositive due to incident infections (seroincidence). Cross-sectional serology can provide estimates of FOI by contrasting seroprevalence across ages. Age of reported cases can also be used to infer FOI. Agreement of these measurements, however, has not been assessed. Using 26 y of data from cohort studies and hospital-attended cases from Kamphaeng Phet province, Thailand, we found FOI estimates from the three sources to be highly inconsistent. Annual FOI estimates from seroincidence were 1.75 to 4.05 times higher than case-derived FOI. Seroprevalence-derived was moderately correlated with case-derived FOI (correlation coefficient = 0.47) with slightly lower estimates. Through extensive simulations and theoretical analysis, we show that incongruences between methods can result from failing to account for dengue antibody kinetics, assay noise, and heterogeneity in FOI across ages. Extending standard inference models to include these processes reconciled the FOI and susceptibility estimates. Our results highlight the importance of comparing inferences across multiple data types to uncover additional insights not attainable through a single data type/analysis.

Linking multiple serological assays to infer dengue virus infections from paired samples using mixture models

Dengue virus (DENV) is an increasingly important human pathogen, with already half of the globe's population living in environments with transmission potential. Since only a minority of cases are captured by direct detection methods (RT-PCR or antigen tests), serological assays play an important role in the diagnostic process. However, individual assays can suffer from low sensitivity and specificity and interpreting results from multiple assays remains challenging, particularly because interpretations from multiple assays may differ, creating uncertainty over how to generate finalized interpretations. We develop a Bayesian mixture model that can jointly model data from multiple paired serological assays, to infer infection events from paired serological data. We first test the performance of our model using simulated data. We then apply our model to 677 pairs of acute and convalescent serum collected as a part of illness and household investigations across two longitudinal cohort studies in Kamphaeng Phet, Thailand, including data from 232 RT-PCR confirmed infections (gold standard). We compare the classification of the new model to prior standard interpretations that independently utilize information from either the hemagglutination inhibition assay (HAI) or the enzyme-linked immunosorbent assay (EIA). We find that additional serological assays improve accuracy of infection detection for both simulated and real world data. Models incorporating paired IgG and IgM data as well as those incorporating IgG, IgM, and HAI data consistently have higher accuracy when using PCR confirmed infections as a gold standard (87-90% F1 scores, a combined metric of sensitivity and specificity) than currently implemented cut-point approaches (82-84% F1 scores). Our results provide a probabilistic framework through which multiple serological assays across different platforms can be leveraged across sequential serum samples to provide insight into whether individuals have recently experienced a DENV infection. These methods are applicable to other pathogen systems where multiple serological assays can be leveraged to quantify infection history.

Serodynamics: A primer and synthetic review of methods for epidemiological inference using serological data

We present a review and primer of methods to understand epidemiological dynamics and identify past exposures from serological data, referred to as serodynamics. We discuss processing and interpreting serological data prior to fitting serodynamical models, and review approaches for estimating epidemiological trends and past exposures, ranging from serocatalytic models applied to binary serostatus data, to more complex models incorporating quantitative antibody measurements and immunological understanding. Although these methods are seemingly disparate, we demonstrate how they are derived within a common mathematical framework. Finally, we discuss key areas for methodological development to improve scientific discovery and public health insights in seroepidemiology.

Estimating the force of infection of four dengue serotypes from serological studies in two regions of Vietnam

Dengue is endemic in Vietnam with circulation of all four serotypes (DENV1-4) all year-round. It is hard to estimate the disease's true serotype-specific transmission patterns from cases due to its high asymptomatic rate, low reporting rate and complex immunity and transmission dynamics. Seroprevalence studies have been used to great effect for understanding patterns of dengue transmission. We tested 991 population serum samples (ages 1-30 years, collected 2013 to 2017), 531 from Ho Chi Minh City and 460 from Khanh Hoa in Vietnam, using a flavivirus protein microarray assay. By applying our previously developed inference framework to the antibody profiles from this assay, we can (1) determine proportions of a population that have not been infected or infected, once, or more than once, and (2) infer the infecting serotype in those infected once. With these data, we then use mathematical models to estimate the force of infection (FOI) for all four DENV serotypes in HCMC and KH over 35 years up to 2017. Models with time-varying or serotype-specific DENV FOI assumptions fit the data better than constant FOI. Annual dengue FOI ranged from 0.005 (95%CI: 0.003-0.008) to 0.201 (95%CI: 0.174-0.228). FOI varied across serotypes, higher for DENV1 (95%CI: 0.033-0.048) and DENV2 (95%CI: 0.018-0.039) than DENV3 (95%CI: 0.007-0.010) and DENV4 (95%CI: 0.010-0.016). The use of the PMA on serial age-stratified cross-sectional samples increases the amount of information on transmission and population immunity, and should be considered for future dengue serological surveys, particularly to understand population immunity given vaccines with differential efficacy against serotypes, however, there remains limits to what can be inferred even using this assay.

Age-dependent heterogeneity in the antigenic effects of mutations to influenza hemagglutinin

Human influenza virus evolves to escape neutralization by polyclonal antibodies. However, we have a limited understanding of how the antigenic effects of viral mutations vary across the human population and how this heterogeneity affects virus evolution. Here, we use deep mutational scanning to map how mutations to the hemagglutinin (HA) proteins of two H3N2 strains, A/Hong Kong/45/2019 and A/Perth/16/2009, affect neutralization by serum from individuals of a variety of ages. The effects of HA mutations on serum neutralization differ across age groups in ways that can be partially rationalized in terms of exposure histories. Mutations that were fixed in influenza variants after 2020 cause greater escape from sera from younger individuals compared with adults. Overall, these results demonstrate that influenza faces distinct antigenic selection regimes from different age groups and suggest approaches to understand how this heterogeneous selection shapes viral evolution.

Age-time-specific transmission of hand-foot-and-mouth disease enterovirus serotypes in Vietnam: A catalytic model with maternal immunity

Hand, foot and mouth disease (HFMD) is highly prevalent in the Asia Pacific region, particularly in Vietnam. To develop effective interventions and efficient vaccination programs, we inferred the age-time-specific transmission patterns of HFMD serotypes enterovirus A71 (EV-A71), coxsackievirus A6 (CV-A6), coxsackievirus A10 (CV-A10), coxsackievirus A16 (CV-A16) in Ho Chi Minh City, Vietnam from a case data collected during 2013-2018 and a serological survey data collected in 2015 and 2017. We proposed a catalytic model framework with good adaptability to incorporate maternal immunity using various mathematical functions. Our results indicate the high-level transmission of CV-A6 and CV-A10 which is not obvious in the case data, due to the variation of disease severity across serotypes. Our results provide statistical evidence supporting the strong association between severe illness and CV-A6 and EV-A71 infections. The HFMD dynamic pattern presents a cyclical pattern with large outbreaks followed by a decline in subsequent years. Additionally, we identify the age group with highest risk of infection as 1-2 years and emphasise the risk of future outbreaks as over 50% of children aged 6-7 years were estimated to be susceptible to CV-A16 and EV-A71. Our study highlights the importance of multivalent vaccines and active surveillance for different serotypes, supports early vaccination prior to 1 year old, and points out the potential utility for vaccinating children older than 5 years old in Vietnam.

Age-dependent heterogeneity in the antigenic effects of mutations to influenza hemagglutinin

Human influenza virus evolves to escape neutralization by polyclonal antibodies. However, we have a limited understanding of how the antigenic effects of viral mutations vary across the human population, and how this heterogeneity affects virus evolution. Here we use deep mutational scanning to map how mutations to the hemagglutinin (HA) proteins of the A/Hong Kong/45/2019 (H3N2) and A/Perth/16/2009 (H3N2) strains affect neutralization by serum from individuals of a variety of ages. The effects of HA mutations on serum neutralization differ across age groups in ways that can be partially rationalized in terms of exposure histories. Mutations that fixed in influenza variants after 2020 cause the greatest escape from sera from younger individuals. Overall, these results demonstrate that influenza faces distinct antigenic selection regimes from different age groups, and suggest approaches to understand how this heterogeneous selection shapes viral evolution.

Previous Dengue Infection among Children in Puerto Rico and Implications for Dengue Vaccine Implementation

Limited dengue virus (DENV) seroprevalence estimates are available for Puerto Rico, which are needed to inform the potential use and cost-effectiveness of DENV vaccines. The Communities Organized to Prevent Arboviruses (COPA) is a cohort study initiated in 2018 in Ponce, Puerto Rico, to assess arboviral disease risk and provide a platform to evaluate interventions. We recruited participants from households in 38 study clusters, who were interviewed and provided a serum specimen. Specimens from 713 children aged 1 to 16 years during the first year of COPA were tested for the four DENV serotypes and ZIKV using a focus reduction neutralization assay. We assessed the seroprevalence of DENV and ZIKV by age and developed a catalytic model from seroprevalence and dengue surveillance data to estimate the force of infection for DENV during 2003-2018. Overall, 37% (n = 267) were seropositive for DENV; seroprevalence was 9% (11/128) among children aged 1 to 8 years and 44% (256/585) among children aged 9 to 16 years, exceeding the threshold over which DENV vaccination is deemed cost-effective. A total of 33% were seropositive for ZIKV, including 15% among children aged 0 to 8 years and 37% among children aged 9 to 16 years. The highest force of infection occurred in 2007, 2010, and 2012-2013, with low levels of transmission from 2016 to 2018. A higher proportion of children had evidence of multitypic DENV infection than expected, suggesting high heterogeneity in DENV risk in this setting.

An open-label, Phase 3 trial of TAK-003, a live attenuated dengue tetravalent vaccine, in healthy US adults: immunogenicity and safety when administered during the second half of a 24-month shelf-life

Dengue is caused by a mosquito-transmitted flavivirus. The disease is now endemic to many tropical and subtropical regions, manifesting as approximately 96 million symptomatic cases of dengue each year. Clinical trials have shown TAK-003 (Qdenga®), a live attenuated dengue tetravalent vaccine, to be well-tolerated, immunogenic, and efficacious in adults with no prior exposure to dengue virus infection living in non-endemic regions, as well as in adults and children living in dengue-endemic areas. This open-label, single-arm phase 3 trial (NCT03771963) was conducted in two dengue non-endemic areas of the USA, and it evaluated the immunogenicity and safety of naturally-aged TAK-003 administered to adult participants. Overall, the immunogenicity data from this trial are consistent with those reported from other TAK-003 phase 2 and 3 trials, and the safety data are consistent with the broader integrated safety data analysis. The data show that naturally-aged TAK-003 had a well-tolerated reactogenicity and adverse events profile when administered in the second half of its clinical 24-month shelf-life and that it still elicited an immune response that persisted up to 6 months after the second dose against all four dengue serotypes, with no important safety risks identified during the trial.

Seroprotection against tetanus in southern Vietnam

BACKGROUND

Ongoing tetanus cases and sporadic outbreaks of vaccine-preventable diseases associated with routine vaccination programmes remain problems in many low and middle-income countries, including Vietnam. With no human-to-human transmission or natural immunity, tetanus antibody levels indicate both individual risk of tetanus and gaps in vaccination programmes.

METHODS

To investigate gaps in immunity to tetanus in Vietnam, a country with a historically high level of tetanus vaccination coverage, tetanus antibodies were measure by ELISA from samples selected from a long-term serum bank, established for the purposes of general-population seroepidemiological investigations in southern Vietnam. Samples were selected from 10 provinces, focussing on age-groups targeted by national vaccination programmes for infants and pregnant women (Expanded Programme on Immunization, EPI, and Maternal and Neonatal Tetanus, MNT).

RESULTS

Antibodies were measured from a total of 3864 samples. Highest tetanus antibody concentrations occurred in children under 4 years old, over 90 % of whom had protective levels. Approximately 70 % of children aged 7-12 years had protective antibody concentrations although there was variation among provinces. For infants and children, there were no significant differences in tetanus protection between males and females, but for adults aged 20-35 years, in five of the ten provinces surveyed, protection against tetanus was higher in females (p < 0.05) who are eligible for booster doses under the MNT programme. In seven of ten provinces, antibody concentrations were inversely related to age (p < 0.01) and protection of older individuals was generally low.

CONCLUSION

Widespread immunity to tetanus toxoid is seen in infants and young children consistent with the high coverage rates reported for diptheria tetanus toxoid and pertussis (DTP) in Vietnam. However, the lower antibody concentrations seen in older children and men suggest reduced immunity to tetanus in populations not targeted by EPI and MNT programmes.

Vaccine-induced antibodies to contemporary strains of dengue virus type 4 show a mechanistic correlate of protective immunity

The four dengue virus serotypes (DENV1-4) are mosquito-borne flaviviruses of humans. Several live-attenuated tetravalent DENV vaccines are at different stages of clinical development and approval. In children with no baseline immunity to DENVs, a leading vaccine (Dengvaxia) is efficacious against vaccine-matched DENV4 genotype II (GII) strains but not vaccine-mismatched DENV4 GI viruses. We use a panel of recombinant DENV4 viruses displaying GI or GII envelope (E) proteins to map Dengvaxia-induced neutralizing antibodies (NAbs) linked to protection. The vaccine stimulated antibodies that neutralize the DENV4 GII virus better than the GI virus. The neutralization differences map to 5 variable amino acids on the E protein located within a region targeted by DENV4 NAbs, supporting a mechanistic role for these epitope-specific NAbs in protection. In children with no baseline immunity to DENVs, levels of DENV4 serotype- and genotype-specific NAbs induced by vaccination are predictive of vaccine efficacy.

Dengue virus seroprevalence study in Bangphae district, Ratchaburi, Thailand: A cohort study in 2012-2015

Background

To determine the seroprevalence and transmission dynamics of dengue virus (DENV), age-stratified longitudinal serological surveys were conducted in Bangphae district, Ratchaburi province, Thailand, for 3 years between April 2012 and April 2015.

Methodology

The surveys enrolled 2012 healthy children and adults between 1 and 55 years-of-age, and a longitudinal serosurvey of six repeated bleeds of the same cohort of individuals was conducted every 8 months for the first 2 years (M0, M8, M16) and every half a year (M24, M30, M36) for the rest of the study period. All samples were tested using in-house indirect sandwich dengue IgG ELISA to determine DENV antibody titer, and 640 paired samples which showed rising of DENV IgG titers in paired serum were further tested using in-house neutralization assay, Plaque Reduction Neutralization Test (PRNT₅₀).

Principal findings

When compared against the gold standard based on the results of PRNT₅₀, sensitivity and specificity of indirect ELISA were found to be both about 85%. The overall DENV IgG positivity determined by ELISA was 74.3% in 2012 and increased to 79.4% by the final sample collection in 2015. In our study sample, more than 98% of subjects older than 25 years were found to be seropositive. Among 518 IgG negative subjects at enrollment, the seroconversion rates were measured in paired bleeds; the rates (between successive visits, approximately 6 months) ranged between 4.8% (between M16 and M24) and 14.7% (between M0 and M8). The dominant serotype of primary DENV infection cases based on seroconversion was identified from the PRNT results and it was DENV-2.

Conclusions

Our study documented high levels of seroprevalence and rate of transmission. Given the importance of the serostatus and disease burden in consideration for dengue vaccine introduction, our data could be used in decision-making on implementation of various dengue control and preventive measures.

To estimate the proportion of individuals with past exposure to dengue virus (DENV), we conducted repeated serological surveys in Bangphae district, Ratchaburi province, Thailand, between April 2012 and April 2015. About 45% of the subjects were under 15 years-of-age. The surveys enrolled 2012 healthy individuals between 1 and 55 years-of-age, and a longitudinal serosurvey of six repeated bleeds of the same cohort of individuals was conducted approximately every 6 months. All samples were tested using an indirect dengue IgG ELISA and a subset of paired samples which showed rising of IgG values in pairs were tested using neutralization assay, used as the gold standard. When compared against the gold standard, sensitivity and specificity of indirect ELISA were found to be both 85%. The proportion of the study population with past DENV infection measured by IgG ELISA was 74.3% in 2012 and increased to 79.4% by 2015. More than 98% of subjects older than 25 years showed to have past exposure to dengue. Among 518 subjects without past exposure, the rate of getting infection was between 4.8–14.7% (between successive visits, approximately 6 months). The dominant serotype was DENV-2. The level of proportion of individuals with past DENV infection is an important consideration, and our data could be used in decision-making for dengue vaccine introduction.

Age-seroprevalence curves for the multi-strain structure of influenza A virus

The relationship between age and seroprevalence can be used to estimate the annual attack rate of an infectious disease. For pathogens with multiple serologically distinct strains, there is a need to describe composite exposure to an antigenically variable group of pathogens. In this study, we assay 24,402 general-population serum samples, collected in Vietnam between 2009 to 2015, for antibodies to eleven human influenza A strains. We report that a principal components decomposition of antibody titer data gives the first principal component as an appropriate surrogate for seroprevalence; this results in annual attack rate estimates of 25.6% (95% CI: 24.1% - 27.1%) for subtype H3 and 16.0% (95% CI: 14.7% - 17.3%) for subtype H1. The remaining principal components separate the strains by serological similarity and associate birth cohorts with their particular influenza histories. Our work shows that dimensionality reduction can be used on human antibody profiles to construct an age-seroprevalence relationship for antigenically variable pathogens.

Estimating the annual dengue force of infection from the age of reporting primary infections across urban centres in endemic countries

BACKGROUND

Stratifying dengue risk within endemic countries is crucial for allocating limited control interventions. Current methods of monitoring dengue transmission intensity rely on potentially inaccurate incidence estimates. We investigated whether incidence or alternate metrics obtained from standard, or laboratory, surveillance operations represent accurate surrogate indicators of the burden of dengue and can be used to monitor the force of infection (FOI) across urban centres.

METHODS

Among those who reported and resided in 13 cities across the Philippines, we collected epidemiological data from all dengue case reports between 2014 and 2017 (N 80,043) and additional laboratory data from a cross-section of sampled case reports (N 11,906) between 2014 and 2018. At the city level, we estimated the aggregated annual FOI from age-accumulated IgG among the non-dengue reporting population using catalytic modelling. We compared city-aggregated FOI estimates to aggregated incidence and the mean age of clinically and laboratory diagnosed dengue cases using Pearson's Correlation coefficient and generated predicted FOI estimates using regression modelling.

RESULTS

We observed spatial heterogeneity in the dengue average annual FOI across sampled cities, ranging from 0.054 [0.036-0.081] to 0.249 [0.223-0.279]. Compared to FOI estimates, the mean age of primary dengue infections had the strongest association (ρ -0.848, p value<0.001) followed by the mean age of those reporting with warning signs (ρ -0.642, p value 0.018). Using regression modelling, we estimated the predicted annual dengue FOI across urban centres from the age of those reporting with primary infections and revealed prominent spatio-temporal heterogeneity in transmission intensity.

CONCLUSIONS

We show the mean age of those reporting with their first dengue infection or those reporting with warning signs of dengue represent superior indicators of the dengue FOI compared to crude incidence across urban centres. Our work provides a framework for national dengue surveillance to routinely monitor transmission and target control interventions to populations most in need.

A mixture model to assess the the immunogenicity of an oral rotavirus vaccine among healthy infants in Niger

Analysis of immunogenicity data is a critical component of vaccine development, providing a biological basis to support any observed protection from vaccination. Conventional methods for analyzing immunogenicity data use either post-vaccination titer or change in titer, often defined as a binary variable using a threshold. These methods are simple to implement but can be limited especially in populations experiencing natural exposure to the pathogen. A mixture model can overcome the limitations of the conventional approaches by jointly modeling the probability of an immune response and the level of the immune marker among those who respond. We apply a mixture model to analyze the immunogenicity of an oral, pentavalent rotavirus vaccine in a cohort of children enrolled into a placebo-controlled vaccine efficacy trial in Niger. Among children with undetectable immunoglobulin A (IgA) at baseline, vaccinated children had 5.2-fold (95% credible interval (CrI) 3.7, 8.3) higher odds of having an IgA response than placebo children, but the mean log IgA among vaccinated responders was 0.9-log lower (95% CrI 0.6, 1.3) than among placebo responders. This result implies that the IgA response generated by vaccination is weaker than that generated by natural infection. Multivariate logistic regression of seroconversion defined by ≥ 3-fold rise in IgA similarly found increased seroconversion among vaccinated children, but could not demonstrate lower IgA among those who seroresponded. In addition, we found that the vaccine was less immunogenic among children with detectable IgA pre-vaccination, and that pre-vaccination infant serum IgG and mother's breast milk IgA modified the vaccine immunogenicity. Increased maternal antibodies were associated with weaker IgA response in placebo and vaccinated children, with the association being stronger among vaccinated children. The mixture model is a powerful and flexible method for analyzing immunogenicity data and identifying modifiers of vaccine response and independent predictors of immune response.

Serotype-specific detection of dengue viruses in a nonstructural protein 1-based enzyme-linked immunosorbent assay validated with a multi-national cohort

BACKGROUND

Dengue virus (DENV) infections pose one of the largest global barriers to human health. The four serotypes (DENV 1-4) present different symptoms and influence immune response to subsequent DENV infections, rendering surveillance, risk assessments, and disease control particularly challenging. Early diagnosis and appropriate clinical management is critical and can be achieved by detecting DENV nonstructural protein 1 (NS1) in serum during the acute phase. However, few NS1-based tests have been developed that are capable of differentiating DENV serotypes and none are currently commercially available.

METHODOLOGY/PRINCIPLE FINDINGS

We developed an enzyme-linked immunosorbent assay (ELISA) to distinguish DENV-1-4 NS1 using serotype-specific pairs of monoclonal antibodies. A total of 1,046 antibodies were harvested from DENV-immunized mice and screened for antigen binding affinity. ELISA clinical performance was evaluated using 408 polymerase chain reaction-confirmed dengue samples obtained from patients in Brazil, Honduras, and India. The overall sensitivity of the test for pan-DENV was 79.66% (325/408), and the sensitivities for DENV-1-4 serotyping were 79.1% (38/48), 80.41% (78/97), 100% (45/45), and 79.6% (98/123), respectively. Specificity reached 94.07-100%.

SIGNIFICANCE

Our study demonstrates a robust antibody screening strategy that enabled the development of a serotype NS1-based ELISA with maximized specific and sensitive antigen binding. This sensitive and specific assay also utilized the most expansive cohort to date, and of which about half are from Latin America, a geographic region severely underrepresented in previous similar studies. This ELISA test offers potential enhanced diagnostics during the acute phase of infection to help guide patient care and disease control. These results indicate that this ELISA is a promising aid in early DENV-1-4 diagnosis and surveillance in regions of endemicity in addition to offer convenient monitoring for future vaccine interventions.