A holistic approach for modeling the dynamic of chikungunya and developing control actions

Présentation du modèle de propagation du Chikungunya au Congrès ASAIHL à Montpellier le 12 juin 2014, pour lequel je suis co-auteur s'agissant de la partie clinique.

 A holistic approach for modeling the dynamic of chikungunya and developing control actions

Roger Frutos^1,2,*, Aurélie Merlin³, Siobhan Staunton⁴, Patrice Ravel⁵, Mireille Fargette³, Hervé Quiquanpoix5, Tri Baskoro Tunggul Satoto⁶, Erna Andayani⁷, Susi Damayanti⁶, Guilhem Kister⁵, Yannick Bardie⁸, Emmanuel Cornillot¹, Christian Devaux¹, Catherine Moulia⁹, Laurent Gavotte⁹ Thérèse Libourel³, Laurence Briant¹, Nathalie Chazal¹

1: Centre d’étude des Pathogènes et Biotechnologies pour la Santé, CPBS, UMR 5236 CNRS - UM1- UM2, Montpellier, France

2: Intertryp, UMR 77 CIRAD-IRD, Montpellier, France

3: Espace Dev, UMR IRD-UM2-UAG-ULR, Montpellier, France

4: Eco&Sols, UMR 210, INRA-IRD-CIRAD-Supagro, Montpellier France

5: Laboratoire de physique industrielle et traitement de l’information, EA2415 UM1, Montpellier, France

6: Center for Tropical Medicine, Faculty of Medicine, Gadjah Mada University, Yogyakarta, Indonesia

7: Sukoharjo Regency Health Center, Center of Java, Sukoharjo, Indonesia

8: Intelligence in Life – R&D Clinique SARL, Montpellier, France

9: Institut des Sciences de l’Evolution de Montpellier - ISEM, UMR 5557 CNRS – UM2, Montpellier, France

* Corresponding author: roger.frutos@univ-montp2.fr, roger.frutos@cirad.fr

Objectives: We developed integrative approaches for the study of emerging diseases. Chikungunya virus (CHIKV), a mosquito-borne pathogen from the alphavirus genius transmitted by Aedes mosquitoes (Aedes aegypti and Aedes albopictus) causes acute infection in humans characterized by rash, high fever and, its hallmark feature is a severe arthritis that can persist for years. These disease was taken as a model system. Modeling the dynamic of the disease in Asia, where the CHIKV is actively propagating, is essential to develop prevention strategies for regions with high risk of propagation, including Europe where the vector A. albopictus is now well established. Multidisciplinary and multi-scale approaches are considered in our solutions.

Methods: We have constructed an object model (UML class diagram) to describe the structure of the system. This object model has been extended to dengue. The modeling of the dynamic of the transmission of CHIKV was performed with a multi-agent system implemented on the Netlogo platform. A case study has been selected in the region of Yogyakarta-Sukoharjo, Indonesia to validate both static and dynamic models. Information relative to geolocalisation, demography, environment (rural, urban, industrial, …), clinical information, vector population, has been collected on the field using standardized procedures.

Results: The class diagram defines the compartments of the model. We carefully describe compartments concerning environment description (description referring to geographic standards, ecology, anthropic influence), sample description (vector, human, virus triptych), clinical data (including media influence) and diagnostic tools. Data have been collected in a periurban zone of Yogyakarta (RT42) and a rural section of Sukoharjo district (RT02 and RT03). A questionnaire to collect clinical information has been set up following OBOE standard and presented to the local health centers. The index case was successfully established in the rural region. SEIR (Sensitive/Exposed/Infected/Resistant) compartments were considered for the host while SEI (Sensitive/Exposed/Infected) were selected for the vector to establish the dynamic model. We have been able to test the influence of medical care policies, vector control options and external income of virus. We also develop analyses on the fate and persitance of mosquitocidal agents post treatment.

Conclusion: We have tested the influence of medical practices in the development of the illness. Isolation of patients during the acute initial phase of the infection would avoid transmission and should drastically reduce the transmission of the virus. According to our study, the presence of large factories in the region of Sukoharjo is suspected to constitute the major source of CHIKV for the region. Whether factories provide suitable habitat for local populations of vector and whether industrial activity contributes to the introduction of populations originating elsewhere will be tested by further serological and molecular analyses.