DIMACS Workshop on Immuno-epidemiology

December 11 - 13, 2006
DIMACS Center, CoRE Building, Rutgers University

Hans Heesterbeek, Universiteit Utrecht, The Netherlands, j.a.p.heesterbeek at vet.uu.nl
Rob de Boer, Universiteit Utrecht, The Netherlands, r.j.deboer@bio.uu.nl
Presented under the auspices of the Special Focus on Computational and Mathematical Epidemiology.


Daniel Coombs, University of British Columbia

Title: Virus competition at multiple scales

Viruses compete and are subject to natural selection at multiple levels: within-cell, within-host and within-population (of hosts). We looked at how viruses can optimally exploit their hosts and how this behaviour may influence the most successful strategy at the between-host, or epidemiological level. I will present a fairly general way to consistently combine models of disease process and disease spread with the goal of understanding the net selection pressure on a model virus. The method is illustrated using two popular models at the within- and between-host levels.

Rob De Boer, Utrecht University, The Netherlands

Title: Simple models of the immune response

If we are to extend current epidemiological models with the within host dynamics of the pathogen and the host immune response, we need simple immune system models. The immune system is complicated and immune responses to particular pathogens are typically poorly characterized and understood. It is therefore not feasible to link simple epidemiological models with large and realistic immune system models. Simple immune response models however have to allow for variability between hosts, and for the typical different phases of the immune response. An important parameter in the immunological variability between hosts is the polymorphism of the MHC. The effect of MHC molecules on the immune response is complicated, and I will discuss how this could be incorporated in the host immune response.

Klaus Dietz, University of Tuebingen, Germany

Title: Immuno-epidemiology of malaria

Plasmodium falciparum (P.f.) malaria is still among the leading causes of childhood mortality in Africa. Sir Ronald Ross, the discoverer of the malaria transmission cycle via Anopheles mosquitos developed already in 1911 the first SIS model, which implicitly assumes premunition, i.e. the presence of one infection prevents superinfections. After a survey of a whole series of malaria immunity models the presentation will summarize the attempts to model the impact of malaria vaccines on the clinical epidemiology and the natural history of P.f. malaria (Supplement No. 2 to the American Journal of Tropical Medicine and Hygiene, 75, August 2006, 1-143) A within-host model for innate and adaptive immunity which was fitted to malaria therapy data will be described in detail.

Vitaly Ganusov, Utrecht University

Title: Evolution of pathogens: a within-host approach

I review one of the current theoretical approaches used to understand evolution of pathogens. In this approach, we currently restrict our analysis to pathogens causing acute infections in vertebrates and focus on detailed description of the within-host dynamics of pathogens including pathogen's growth within the host, generation of the host's immune response controlling and clearing the pathogen, and pathogen's transmission from the infected host. The approach has allowed us to reach several interesting conclusions such as the following: i) host immune response may be one of the forces that maintain virulence of pathogens in vertebrates; ii) in spite of general expectations, higher levels of host heterogeneity may select for pathogens with higher virulence; iii) the trade-offs between pathogen transmissibility, the rate of host recovery and virulence naturally arise from the within-host dynamics; iv) these simple ``within-host'' models are not robust at predicting optimal virulence of pathogens since changing parameters or assumptions of the model may dramatically change the level of virulence to which pathogens are expected to evolve. I discuss the use of these "within-host" models to understand the impact of vaccination on the evolution of pathogens, and potential applications of these models for fine tuning of epidemiological models for disease spread.

Gabriela Gomes, Paula Rodrigues, and Sander van Noort, Instituto Gulbenkian de Ciência

Title: Epidemiological consequences of species/strain competition and selection: tuberculosis and malaria

All human populations are subject to an intensity of Mycobacterium tuberculosis (Mtb) transmission that varies according to demographic and socioeconomic backgrounds. They are also exposed to populations of environmental mycobacteria (EM) whose composition varies between geographical regions. The bacille Calmette-Guérin (BCG) vaccine is used worldwide but its efficacy against pulmonary tuberculosis has revealed great variability. We develop mathematical models that describe the transmission of Mtb under constraints that are imposed by host immunity elicited by previous exposures to Mtb, EM and BCG. We describe how vaccine efficacy (protection at the individual level) and effectiveness (protection at the population level) depend on the intensity of exposure to both Mtb and EM. We conclude that variations in Mtb transmissibility can explain extreme variability in the effectiveness of vaccination programs even when vaccine efficacy is constant. We also verify that variations in EM abundance can explain some variability in both vaccine efficacy and effectiveness but the interpretation needs special care as the two trends may seem conflicting. Standard estimates of vaccine efficacy may give a misleading impression of the impact that a vaccination program can have at the population level.

Each Plasmodium falciparum parasite has a repertoire of ~60 variant surface antigens (VSAs), which are expressed by infected red blood cells. These VSAs are extremely diverse but some families are relatively more conserved. It has been suggested that an antigenically conserved group of VSAs correlates with severe disease (VSA_SM) while an antigenically more diverse group is associated with uncomplicated malaria (VSA_UM). We develop a mathematical model where each parasite is characterized be multiple loci and multiple alleles and identify the expected patterns of parasite diversity circulating in a host population. In particular we investigate the assumption that VSA_SM are positively selected to be expressed in naïve hosts. The next step is to combine transmission models with within-host models that dictate which VSAs are expressed based on immune history.

Can Kesmir, Utrecht University

Title: Pathogen adaptation in MHC Class I antigen presentation pathway: within-host and population level

In the recent years we have been developing a number of tools to predict the specificity of several steps involved in antigen presentation and processing pathways. Using these methods we can now study the adaptation of viruses to their host immune responses. This research benefits heavily from large amount of data available on HIV, and therefore is rather focused on evolution of HIV with its host immune responses. The first part of my talk will be on adaptation of HIV-1 to the host antigen processing. The second part will be on MHC effect in HIV disease progression. And finally I will discuss my recent results of adaptation of the chimpanze to SIV.

D. Klinkenberg, M. Severins, H. Heesterbeek, Utrecht University

Title: Immuno-epidemiology of coccidiosis

Coccidiosis in chickens is an intestinal infection caused by protozoans of the genus Eimeria. Infection occurs by ingestion of oocysts, the obligatory environmental stage of the parasite, and the uptake history determines the gradual build up of immunity, which protects against tissue damage and excretion of oocysts. We first modelled the within-host dynamics of the parasite and its interaction with the immune system, thus obtaining a model that gives a good description of the relation between uptake history and excretion of oocysts. Then we added a variable for environmental contamination, enabling coupling of the excretion of oocysts to re-uptake. This resulted in a model for Eimeria dynamics during a cohort of chickens in a shed, and between subsequent cohorts. Recently, we took a different approach to model the same system: a spatial and stochastic individual-based model. Both models can explain the experimental observation that intermediate oocyst levels at the start of a chicken production cycle may give rise to minimal tissue and production damage.

Björn Peters, La Jolla Institute for Allergy and Immunology

Title: Defining and characterizing epitopes in infectious pathogens ? from prediction to protection

A number of experimental and theoretical tools have been developed to identify peptides contained in a given pathogen which are likely to be recognized by T cells of an infected host. Such tools evaluate the ability of peptides to bind MHC molecules, the efficiency with which peptides are being processed by proteasomal cleavage and TAP transport, and the availability of a T cell repertoire capable of recognizing the peptides. In this presentation, the capability of these tools to identify peptides that can be used to protectively vaccinate will be discussed. Also, the use of the Immune Epitope Database to extract such experimental information will be demonstrated.

Mick Roberts, Massey University

Title: A mathematical model for the within-host evolution and transmission of a virus

A model for the evolution and transmission of a virus is described. Its essential features are: within-host evolution through random mutation during virus replication; host heterogeneity in HLA types; and between-host transmission with dynamic viral diversity due to the outgrowth of quasispecies that are most fit in the recipient host. The model, expressed as a system of ordinary differential equations, has been used to describe the dynamics of the HIV virus. The results illustrate how an HIV infection progresses until the generation of an escape mutant causes host death.

Collaborative work with Angela McLean, Oxford University.

Tanya Kostova - Vassilevska, Lawrence Livermore National Laboratory

Title: Viral spread in a network of hosts

I will describe a general framework to model the spread of virus within a population of hosts which combines viral replication and immune response at the in-host level and between host infection at the population level which is prone to some mathematical analysis. The population - level model is based on a fully heterogeneous contact network. The in-host model is sufficiently general to represent various modelling assumptions.

Previous: Program
Workshop Index
DIMACS Homepage
Contacting the Center
Document last modified on November 28, 2006.