Immuno-epidemiology of malaria Klaus Dietz Department of Medical Biometry University of Tübingen, Germany DIMACS Worksop 11-13 December 2006.

Immuno-epidemiology of malaria

Klaus DietzDepartment of Medical Biometry

University of Tübingen, Germany

DIMACS Worksop

11-13 December 2006

Outline

• What is immuno-epidemiology?• What is malaria?• Why model malaria immunity? • Malaria immunity models: a brief history• A within-host malaria model• Concluding remarks

Serological surveys as immuno-epidemiological tools

Desowitz RS, Saave JJ, Stein B.The application of the indirect haemagglutination test in recent studies on the immuno-epidemiology of human malaria and the immune response in experimental malaria.Mil Med.,131:1157-1166 (1966).

Suzuki M. Malaria immuno-epidemiology: a trial to link field study with basic science.Gaoxiong Yi Xue Ke Xue Za Zhi.7:224-232 (1991).

.....malaria serological assessment was carried out in endemic areas in Haiti, Indonesia, Sudan and in Brazil Amazon. The serological survey was useful in finding latent foci in a controlled area, for the assessment of past epidemics,....

A theoretical framework for immuno-epidemiology

Woolhouse, M.E.J.et al.: Acquired immunology and epidemiology of Schistosoma haematobium., Nature 351, 757-759, 1991 („Acquisition of this immunity seems to be related to the cumulative effects of repeated infection and provides only partial protection. These characteristics are consistent with immuno-epidemiological data for both S. mansoni and S. haematobium infections of humans.“)

Hellriegel, B.: Immunoepidemiology-bridging the gap between immunology and epidemiology. Trends in Parasitology, 17, 102-106, 2001

„Immunoepidemiology combines individual- and population-oriented approaches to create new perspectives. It examines how inter-individual differences in immune responses affect the population dynamics of micro- and macro-parasites to produce the epidemiological patterns of infection observed in heterogeneous host populations.“

A cartoon of immuno-epidemiological models

0)0,();(),0(

)0()0,();(),0(

0)0,();(),0(

,....,1

,

),(

),(),....,,,( 1

tmamam

ztzazaz

tyayay

ni

zya

m

t

m

zya

z

t

z

zyyyata

y

t

y

iiii

iiiii

iiii

iiiii

iiiii

iiiniii

Parasite density

Immunity level

Morbidity level

• Many millions of simulation runs, each in the order of hours

• Need for supercomputing

STI project on the mathematical modeling of the impact of malaria vaccines on the clinical

epidemiology and natural history of Plasmodium falciparum malaria

Thomas A. Smith and team

Network Computing

Volunteer Computing Project

Malaria cycle

Source:Wellcome Trust

Infected Red Blood Cells (IREs) by asexual (red) and sexual (green) parasites in Patient G141

n = 54; mean 212 days, median 216 days

Present geographical distribution of malaria

The challenge of malaria

In June, President Bush announced a new commitment to combat malaria. His proposal calls for an additional $1.2 billion over the next five years. The money will pay for insecticide-treated nets, it will allow for indoor spraying against mosquitoes, and it will provide effective new combination drugs to treat malaria. Our goal with this new funding is to reach more than 175 million people in 15 nations. (1.37 $/Person/year)

The President And Mrs. Bush Will Host The White House Summit On Malaria On December 14, 2006, In Washington, DC, To Discuss And Highlight Measures For Combating This Preventable Disease

Growth of the yearly number of malaria publications

Doubling time: 10 years and 7 months

Why model malaria immunity?

• Not necessary, if one is aiming for eradication:

• Macdonald, G.: Theory of the eradication of malaria. Bull. WHO 15, 369-387, 1956.

• WHO global malaria eradication campaign 1955

Original Eradication Plans

• Interruption of transmission of main species infecting humans by DDT spraying

• Malaria disappears spontaneously in under 3 years

Source: Gabaldon

Pampana: Textbook on malaria eradication

• If malaria eradication will be achieved it will be first of all due to the computer

• Macdonald: DDT spraying and mass drug administration every two months will interrupt malaria transmission in Africa

Force of infection (per year)

Ce

rebr

al m

ala

ria

inci

den

ce

(per

10

00 p

er y

ear)

(age

s 0-

9 y

ea

rs)

0

0.5

1.0

1.5

2.0

2.5

3.0

0 0.5 1.0 1.5 2.0 2.5

Bakau(Gambia)

Sukuta(Gambia)

Kilifi N(Kenya)

Kilifi S(Kenya)

Siaya(Kenya)

Snow et al. (1997) The Lancet

hypo- meso- hyper- holo-endemic

(graph by Martin Eichner)

Points of attack of potential malaria vaccines

Smith et al (2006)

Malaria immunity models

• Dietz, K: Mathematical models for transmission and control of malaria. 1091-1133 (1988)

• Molineaux, L and Dietz, K: Review of intra-host models of malaria. Parassitologia 41:221-231 (1999)

• McKenzie, FE and Bossert, WH: An integrated model of Plasmodium falciparum dynamic. J Theor. Biology 232, 411-426 (2005)

Ronald Ross (1857-1932)

Second Nobel price in Medicine in 1902

Infection without (Ross) and with (Macdonald)

superinfectionsRoss („premunition“)

Macdonald

0

0 1 k

1

Age-specific prevalence

)(1

)(exp1

1exp1

1)(

infection one from raterecovery

rate inoculaion gicalepidemiolo

R

RM

aM

aR

p

pp

eap

eap

Prevalence by age

6.2;yearper 8.1;yearper 6.4

Implications of unrealistic assumptions about the strength of immunity

• Underestimating the strength of immunity leads to an

• underestimate of the basic reproduction number and consequently an

• underestimate of the necessary eradication efforts and an

• overestimate of the expected control effects

Garki Model (Dietz, Molineaux, Thomas, 1974)Garki Book (Molineaux L and Gramiccia G, 1980

http://whqlibdoc.who.int/publications)

Garki Data: http://www.sti.ch/de/forschung/biostatistics/downloads.html

1y

2y

1y 2y 3y

1x 2x

Garki model fit

Fit to Garki data of the STI project

Prevalence by vectorial capacity

Gupta, S. and Day, K.P.: A theoretical framework for the immunoepidemiology of Plasmodium falciparum malaria.

Parasite Immunology, 16, 361-370 (1994)

0 1 2 k n

n 1n 2n kn

20;30;40

25.30

n

R

3.25 5; 7.5;

20

0 R

n

„Might malaria yield to mathematics“

(Economist, March 12th 1994)

„Recent epidemiological work suggests that there may be a stronger chance of controlinling malaria than was once thought likely…

Comparing the distribution of the strains among people by age with the spread produced by their* model, they found a very close match if the average number of secondary cases caused by a single case was between six and seven.“

*Gupta, Anderson, Day and Trenholme

Innate and adaptive immunityStevenson and Riley (2004)

• Research on the immunology of malaria has tended to focus on adaptive immunity

• Accumulating evidence…indicates a crucial role for innate immune responses in protective immunity to malaria

• Innate responses are essential to limit the initial phase of parasite replication, controlling the first wave of parasitemia and allowing the host time to develop specific adaptive responses that will enable the infection to be cleared

Fig 1A

Fig 1B

Fig 1C

Correlations between innate immunity threshold,

parasite growth rate and peak density

Correlation between innate and adaptive immunity thresholds and between two

parameters for adaptive immunity

Feverthresholds

Vaccine effect on initial multiplication factor

Fig 3

A B

4 3 2 1 0

Vaccine efficacy

Conclusions from the within-host model for the first wave

• The 800 data points for 100 patients could well be described by a simple model with four interpretable parameters per patient.

• All parameters show large variation.• The maximum parasitaemia is mainly controlled by

innate immunity.• The benefit of a vaccine targetting the asexual blood

stages is expected to be strongly host dependent.• At low immunogenicity the expected vaccine efficacy

against severe malaria is much higher than against fever.

Concluding remarks• Historic data (neurosyphilis patients, Garki

project) are still relevant today and in the future because new data about the natural course of malaria can for ethical reasons no longer be collected.

• Estimates for basic reproduction numbers are model dependent.

• In spite of the fact that malaria is probably the disease with the largest number of models there is still no generally acceptable model.

Acknowledgements

• Louis Molineaux, Geneva

• Tom Smith and his team at the STI, Basle

• Martin Eichner, Günter Raddatz, Tübingen