Abstract

As with many pathogens, most dengue infections are subclinical and therefore unobserved1. Coupled with limited understanding of the dynamic behaviour of potential serological markers of infection, this observational problem has wide-ranging implications, including hampering our understanding of individual- and population-level correlates of infection and disease risk and how these change over time, between assay interpretations and with cohort design. Here we develop a framework that simultaneously characterizes antibody dynamics and identifies subclinical infections via Bayesian augmentation from detailed cohort data (3,451 individuals with blood draws every 91 days, 143,548 haemagglutination inhibition assay titre measurements)2,3. We identify 1,149 infections (95% confidence interval, 1,135–1,163) that were not detected by active surveillance and estimate that 65% of infections are subclinical. After infection, individuals develop a stable set point antibody load after one year that places them within or outside a risk window. Individuals with pre-existing titres of ≤1:40 develop haemorrhagic fever 7.4 (95% confidence interval, 2.5–8.2) times more often than naive individuals compared to 0.0 times for individuals with titres >1:40 (95% confidence interval: 0.0–1.3). Plaque reduction neutralization test titres ≤1:100 were similarly associated with severe disease. Across the population, variability in the size of epidemics results in large-scale temporal changes in infection and disease risk that correlate poorly with age.

Access optionsAccess options

Rent or Buy article

Get time limited or full article access on ReadCube.

from$8.99

All prices are NET prices.

Additional information

Publisher’s note: Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

References

  1. 1.

    Undurraga, E. A., Halasa, Y. A. & Shepard, D. S. Use of expansion factors to estimate the burden of dengue in Southeast Asia: a systematic analysis. PLoS Negl. Trop. Dis. 7, e2056 (2013).

  2. 2.

    Endy, T. P. et al. Epidemiology of inapparent and symptomatic acute dengue virus infection: a prospective study of primary school children in Kamphaeng Phet, Thailand. Am. J. Epidemiol. 156, 40–51 (2002).

  3. 3.

    Cauchemez, S. & Ferguson, N. M. Methods to infer transmission risk factors in complex outbreak data. J. R. Soc. Interface 9, 456–469 (2012).

  4. 4.

    Bhatt, S. et al. The global distribution and burden of dengue. Nature 496, 504–507 (2013).

  5. 5.

    Halstead, S. B. Dengue (Imperial College Press, London, 2008).

  6. 6.

    Vaughn, D. W. et al. Dengue in the early febrile phase: viremia and antibody responses. J. Infect. Dis. 176, 322–330 (1997).

  7. 7.

    Harris, E. et al. Clinical, epidemiologic, and virologic features of dengue in the 1998 epidemic in Nicaragua. Am. J. Trop. Med. Hyg. 63, 5–11 (2000).

  8. 8.

    Venturi, G. et al. Humoral immunity and correlation between ELISA, hemagglutination inhibition, and neutralization tests after vaccination against tick-borne encephalitis virus in children. J. Virol. Methods 134, 136–139 (2006).

  9. 9.

    Katzelnick, L. C. et al. Antibody-dependent enhancement of severe dengue disease in humans. Science 358, 929–932 (2017).

  10. 10.

    Halstead, S. B., Rojanasuphot, S. & Sangkawibha, N. Original antigenic sin in dengue. Am. J. Trop. Med. Hyg. 32, 154–156 (1983).

  11. 11.

    Clapham, H. E. et al. Dengue virus (DENV) neutralizing antibody kinetics in children after symptomatic primary and postprimary DENV infection. J. Infect. Dis. 213, 1428–1435 (2016).

  12. 12.

    Salje, H. et al. Variability in dengue titer estimates from plaque reduction neutralization tests poses a challenge to epidemiological studies and vaccine development. PLoS Negl. Trop. Dis. 8, e2952 (2014).

  13. 13.

    Salje, H. et al. Dengue diversity across spatial and temporal scales: local structure and the effect of host population size. Science 355, 1302–1306 (2017).

  14. 14.

    Rodríguez-Barraquer, I. et al. Revisiting Rayong: shifting seroprofiles of dengue in Thailand and their implications for transmission and control. Am. J. Epidemiol. 179, 353–360 (2014).

  15. 15.

    Fried, J. R. et al. Serotype-specific differences in the risk of dengue hemorrhagic fever: an analysis of data collected in Bangkok, Thailand from 1994 to 2006. PLoS Negl. Trop. Dis. 4, e617 (2010).

  16. 16.

    Duong, V. et al. Asymptomatic humans transmit dengue virus to mosquitoes. Proc. Natl Acad. Sci. USA 112, 14688–14693 (2015).

  17. 17.

    D’Agostino, R. B. et al. Relation of pooled logistic regression to time dependent Cox regression analysis: the Framingham Heart Study. Stat. Med. 9, 1501–1515 (1990).

  18. 18.

    Ferguson, N. M. et al. Benefits and risks of the Sanofi–Pasteur dengue vaccine: modeling optimal deployment. Science 353, 1033–1036 (2016).

  19. 19.

    Kliks, S. C., Nimmanitya, S., Nisalak, A. & Burke, D. S. Evidence that maternal dengue antibodies are important in the development of dengue hemorrhagic fever in infants. Am. J. Trop. Med. Hyg. 38, 411–419 (1988).

  20. 20.

    Guzmán, M. G., Alvarez, M. & Halstead, S. B. Secondary infection as a risk factor for dengue hemorrhagic fever/dengue shock syndrome: an historical perspective and role of antibody-dependent enhancement of infection. Arch. Virol. 158, 1445–1459 (2013).

  21. 21.

    Villar, L. et al. Efficacy of a tetravalent dengue vaccine in children in Latin America. N. Engl. J. Med. 372, 113–123 (2015).

  22. 22.

    Moodie, Z. et al. Neutralizing antibody correlates analysis of tetravalent dengue vaccine efficacy trials in Asia and Latin America. J. Infect. Dis. 217, 742–753 (2018).

  23. 23.

    Cummings, D. A. T., Schwartz, I. B., Billings, L., Shaw, L. B. & Burke, D. S. Dynamic effects of antibody-dependent enhancement on the fitness of viruses. Proc. Natl Acad. Sci. USA 102, 15259–15264 (2005).

  24. 24.

    Wearing, H. J. & Rohani, P. Ecological and immunological determinants of dengue epidemics. Proc. Natl Acad. Sci. USA 103, 11802–11807 (2006).

  25. 25.

    Flasche, S. et al. The long-term safety, public health impact, and cost-effectiveness of routine vaccination with a recombinant, live-attenuated dengue vaccine (dengvaxia): a model comparison study. PLoS Med. 13, e1002181 (2016).

  26. 26.

    Adams, B. et al. Cross-protective immunity can account for the alternating epidemic pattern of dengue virus serotypes circulating in Bangkok. Proc. Natl Acad. Sci. USA 103, 14234–14239 (2006).

  27. 27.

    Grenfell, B. T., Bjørnstad, O. N. & Kappey, J. Travelling waves and spatial hierarchies in measles epidemics. Nature 414, 716–723 (2001).

  28. 28.

    Earn, D. J., Rohani, P., Bolker, B. M. & Grenfell, B. T. A simple model for complex dynamical transitions in epidemics. Science 287, 667–670 (2000).

  29. 29.

    Cobey, S. & Lipsitch, M. Niche and neutral effects of acquired immunity permit coexistence of pneumococcal serotypes. Science 335, 1376–1380 (2012).

  30. 30.

    Mammen, M. P. et al. Spatial and temporal clustering of dengue virus transmission in Thai villages. PLoS Med. 5, e205 (2008).

  31. 31.

    Salje, H. et al. Revealing the microscale spatial signature of dengue transmission and immunity in an urban population. Proc. Natl Acad. Sci. USA 109, 9535–9538 (2012).

  32. 32.

    Endy, T. P. et al. Spatial and temporal circulation of dengue virus serotypes: a prospective study of primary school children in Kamphaeng Phet, Thailand. Am. J. Epidemiol. 156, 52–59 (2002).

  33. 33.

    Smith, T. & Vounatsou, P. Estimation of infection and recovery rates for highly polymorphic parasites when detectability is imperfect, using hidden Markov models. Stat. Med. 22, 1709–1724 (2003).

  34. 34.

    Reich, N. G. et al. Interactions between serotypes of dengue highlight epidemiological impact of cross-immunity. J. R. Soc. Interface 10, 20130414 (2013).

  35. 35.

    Rudolph, K. E., Lessler, J., Moloney, R. M., Kmush, B. & Cummings, D. A. T. Incubation periods of mosquito-borne viral infections: a systematic review. Am. J. Trop. Med. Hyg. 90, 882–891 (2014).

  36. 36.

    White, M. T. et al. A combined analysis of immunogenicity, antibody kinetics and vaccine efficacy from phase 2 trials of the RTS,S malaria vaccine. BMC Med. 12, 117 (2014).

  37. 37.

    Therneau, T., Crowson, C. & Atkinson, E. Using time dependent covariates and time dependent coefficients in the Cox model. R package Survival Vignettes version 2.41 https://cran.r-project.org/web/packages/survival/vignettes/ (2017).

  38. 38.

    Lessler, J., Salje, H., Grabowski, M. K. & Cummings, D. A. T. Measuring Spatial Dependence for Infectious Disease Epidemiology. PLoS ONE 11, e0155249 (2016).

Download references

Acknowledgements

H.S. and D.A.T.C. acknowledge funding form the National Institutes of Health (R01AI114703-01). S.C. acknowledges financial support from the Investissement d’Avenir program, the Laboratoire d’Excellence Integrative Biology of Emerging Infectious Diseases program (grant ANR-10-LABX-62-IBEID), the Models of Infectious Disease Agent Study of the National Institute of General Medical Sciences, and the AXA Research Fund. The manuscript has been reviewed by the Walter Reed Army Institute of Research. There is no objection to its presentation and/or publication. The opinions or assertions contained herein are the private views of the author, and are not to be construed as official, or as reflecting the true views of the Department of the Army or the Department of Defence. C.K., B.T., A.N., A.W., D.E., I.-K.Y., L.M., R.J., S.T., A.M. and T.E. acknowledge funding from the National Institutes of Health (P01 AI0345333) and the Military Infectious Disease Research Program. The investigators have adhered to the policies for protection of human subjects as prescribed in AR 70–25.

Reviewer information

Nature thanks P. B. Gilbert, S. Hay and the other anonymous reviewer(s) for their contribution to the peer review of this work.

Author information

Author notes

  1. These authors jointly supervised this work: Timothy Endy, Simon Cauchemez.

Affiliations

  1. Mathematical Modelling of Infectious Diseases Unit, Institut Pasteur, Paris, France

    • Henrik Salje
    •  & Simon Cauchemez
  2. CNRS UMR2000, Génomique évolutive, modélisation et santé (GEMS), Institut Pasteur, Paris, France

    • Henrik Salje
    •  & Simon Cauchemez
  3. Center of Bioinformatics, Biostatistics and Integrative Biology, Institut Pasteur, Paris, France

    • Henrik Salje
    •  & Simon Cauchemez
  4. Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Johns Hopkins University, Baltimore, MD, USA

    • Henrik Salje
    • , Derek A. T. Cummings
    •  & Justin Lessler
  5. Department of Biology, University of Florida, Gainesville, FL, USA

    • Derek A. T. Cummings
    •  & Leah C. Katzelnick
  6. Emerging Pathogens Institute, University of Florida, Gainesville, FL, USA

    • Derek A. T. Cummings
  7. University of California, San Francisco, San Francisco, CA, USA

    • Isabel Rodriguez-Barraquer
  8. Department of Virology, Armed Forces Research Institute of Medical Sciences, Bangkok, Thailand

    • Chonticha Klungthong
    • , Butsaya Thaisomboonsuk
    • , Ananda Nisalak
    • , Alden Weg
    • , Damon Ellison
    •  & Louis Macareo
  9. International Vaccine Institute, Seoul, South Korea

    • In-Kyu Yoon
  10. Viral Diseases Branch, Walter Reed Army Institute of Research, Silver Spring, MD, USA

    • Richard Jarman
  11. Department of Medicine, Upstate Medical University of New York, Syracuse, NY, USA

    • Stephen Thomas
    •  & Timothy Endy
  12. Institute for Immunology and Informatics, Department of Cell and Molecular Biology, University of Rhode Island, Providence, RI, USA

    • Alan L. Rothman

Authors

  1. Search for Henrik Salje in:

  2. Search for Derek A. T. Cummings in:

  3. Search for Isabel Rodriguez-Barraquer in:

  4. Search for Leah C. Katzelnick in:

  5. Search for Justin Lessler in:

  6. Search for Chonticha Klungthong in:

  7. Search for Butsaya Thaisomboonsuk in:

  8. Search for Ananda Nisalak in:

  9. Search for Alden Weg in:

  10. Search for Damon Ellison in:

  11. Search for Louis Macareo in:

  12. Search for In-Kyu Yoon in:

  13. Search for Richard Jarman in:

  14. Search for Stephen Thomas in:

  15. Search for Alan L. Rothman in:

  16. Search for Timothy Endy in:

  17. Search for Simon Cauchemez in:

Contributions

H.S., D.A.T.C. and S.C. developed the methods, performed analyses and co-wrote the paper, T.E. conceived the cohort study, T.E., C.K., B.T., A.N., A.W., D.E., L.M., I.-K.Y., R.J., S.T. and A.L.R. ran the study and collected and stored the cohort study results, I.R.-B., J.L. and L.C.K. aided in the interpretation of the results. All authors commented on and edited the paper.

Competing interests

The authors declare no competing interests.

Corresponding author

Correspondence to Henrik Salje.

Extended data figures and tables

  1. Extended Data Fig. 1 Comparison of biphasic versus exponential decay.

    Biphasic and exponential decay curves fitted to haemagglutination inhibition antibody measurements following observed symptomatic infections.

  2. Extended Data Fig. 2 Variability in titre responses and measurement error and bias by serotype.

    a, Variability in titre responses. Violin plots showing median (black square), 25% and 75% quantiles (thick black line) and 95% distribution (in grey) of net titre increases at different time points after infection (n = 1,420). b, Estimated underlying differences across serotypes in the measurement of antibody levels by haemagglutination inhibition assay over that attributable to infection (DENV1 is reference (Ref.)) with 95% credible intervals (fitted to data from 140,612 titre measurements). c, Mean estimated error in the haemagglutination inhibition assay estimated with 95% credible intervals using our model results (grey) and empirically derived (blue) results from 795 repeated measurements on the same serum compared to the values from previously empirically derived estimates12 for PRNTs (blue).

  3. Extended Data Fig. 3 Serotype distributions.

    a, Distribution of serotypes by year comparing the detected symptomatic infections by PCR and the augmented primary infections for which we could confidently assign the serotype (>50% of model iterations inferring the same serotype). We could confidently assign the serotype in 60% of cases. b, Serotype distribution for detected symptomatic primary infections and augmented subclinical primary infections for which the infecting serotype could be confidently assigned (>50% of model iterations inferring the same serotype). c, Distribution of serotypes by year comparing the detected symptomatic infections by PCR and the augmented primary infections using a more stringent cutoff that >75% of model iterations infer the same serotype. In this scenario, we could confidently assign the serotype in 32% of instances.

  4. Extended Data Fig. 4 Cox proportional hazards model versus logistic regression.

    Comparison of results using time varying Cox proportional hazards model (dashed line) with that from logistic regression (solid line) for the annualized probability of infection (a), developing any symptoms (b), being hospitalized (c) and developing DHF (d) as a function of the mean measured antibody titre across all serotypes at the time of exposure using titre data from all study subjects (n = 3,451). The open circles on the left represent primary infections (that is, those with no detectable titres for any serotype before exposure). The shaded regions represent 95% bootstrap confidence intervals. To calculate probabilities, the relative hazards from the Cox model are multiplied by the baseline hazard for those with measured titres of 0 (calculated as proportion of person time with an infection time among those with measured titres of 0).

  5. Extended Data Fig. 5 ROC for the identification of DHF infections.

    Ability of modelled relationship between measured haemagglutination inhibition titre and risk of DHF to identify those with DHF, using those with DHF compared to randomly selected matched controls from individuals in the cohort who had detectable titres at the same time (n = 36 with DHF with the same number of matched controls). AUC, area under the curve.

  6. Extended Data Fig. 6 Probability of disease as a function of haemagglutination inhibition and PRNT titre.

    Probability of disease as a function of mean titre across the four serotypes at the time of infection. a, For those infected during the surveillance windows, the probability of developing any symptoms as a function of mean titre (n = 781). b, For those infected during the surveillance windows, the probability of being hospitalized (n = 781). c, For those infected during the surveillance windows, the probability of developing DHF as a function of mean titre (n = 781). d, For those infected during the surveillance windows (n = 781), the probability of developing any symptoms as a function of mean PRNT titre. e, For those infected, the probability of being hospitalized as a function of mean PRNT titre. f, For those infected, the probability of developing DHF as a function of mean PRNT titre. In each panel, the open circles on the left represent primary infections. The shaded region represents 95% confidence intervals.

  7. Extended Data Fig. 7 Population-level distribution of titres by birth cohort and age.

    a, Proportion of each cohort who are naive as a function of time. b, Proportion of each cohort who are naive as a function of age. c, d, Proportion of each cohort with titres above risk zone (that is, greater than 3) as a function of time (c) and age (d).

  8. Extended Data Fig. 8 ROC for infection detection using different testing protocols.

    The ROC for different assay approaches and time between blood draws calculated from 100,000 simulated titre responses. a, Single serotype assay—when haemagglutination inhibition tests are conducted for only a single serotype at two time points. b, Haemagglutination inhibition tests conducted against all four serotypes. Infections are considered to occur when the ratio of any of the four titres at time point 2 versus time point 1 is greater than the threshold value. c, Haemagglutination inhibition tests conducted against all four serotypes. Infections are considered to occur when the ratio of the mean of the four titres at time point 2 versus the mean at time point 1 is greater than the threshold value.

  9. Extended Data Fig. 9 Performance of assay is dependent on time between blood draws and measurement error.

    Optimization of assays for the detection of events in which the specificity is maintained at >95%. ac, We explore the performance of three different assay testing protocols: current practice for which infection events are defined as a rise above a cutoff point in any serotype across two blood draws (a), a ‘mean approach’ for which the mean across all serotypes is first calculated before comparing the means across time points (b), a ‘mean approach’ for which titres are available on a continuous scale (c). For each protocol, we identify the optimal cutoff point for a range of assay measurement errors from 100,000 simulated titres based on the fitted titre responses from infections in our study population, that maintains a specificity of >95% (top row). We then calculate the sensitivity of the approach for different time intervals between blood draws using 50% held-out data (bottom row). df, Same as ac but using a more stringent cutoff of 99%.

  10. Extended Data Fig. 10 Clustering of symptomatic (n = 274) and subclinical cases (mean n = 507 across 100 reconstructed datasets) by school by time and serotype.

    a, Probability of observing an augmented subclinical infection (irrespective of serotype) occurs at different time intervals within the same school of a detected symptomatic case relative to the probability of observing an augmented subclinical infection occurring in a different school in that same time interval. b, For augmented primary infections that are consistently of the same serotype (defined as >50% of augmented datasets having a primary infection in the same individual caused by the same serotype in the same six-month time window). Probability that an augmented primary infection that occurs within a fixed time window of a PCR-confirmed case and in the same school is of the same serotype relative to the probability that an augmented primary infection that occurs within the same time window in a different school is of the same serotype. Note that the modelling framework can only allow differentiation of serotypes for primary infections. Cross-reaction prevents differentiation in subsequent infections. Overall, 60% of primary infections have a consistent serotype for a primary infection across augmented datasets. Each box plot presents the 2.5%, 25%, 75% and the 97.5% quantiles of the distribution as well as the mean.

Supplementary information

  1. Supplementary Information

    This file contains Supplementary Tables 1-8.

  2. Life Sciences Reporting Summary

  3. Supplementary Table

    This file contains the Titer Database titled Hemagglutination Inhibition titer data. For each individual, the days since recruitment for each blood draw and the Hemagglutination Inhibition titers to each of the four serotypes at that blood draw.

  4. Supplementary Table

    This file contains the Case Database titled Disease event data. For each individual who experienced symptomatic dengue during the surveillance windows, the day since recruitment of the onset of symptoms, the serotype of the infection (where known), whether the case was hospitalized and whether the case developed dengue hemorrhagic fever.

Source Data

About this article

Publication history

Received

Accepted

Published

Issue Date

DOI

https://doi.org/10.1038/s41586-018-0157-4

Further reading

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate.