Modelling malaria virulence 1/10/02. By Caroline Cross The development of malaria vaccines is of crucial importance, but Andrew Read argues that their implementation needs to be handled carefully.

"Why isn't the malaria parasite even nastier?" questions Professor Andrew Read. At present the malaria parasite kills a child every 30 seconds, but it has the capacity for rapid evolution – as evidenced by the spread of drug resistance in many parts of the world. So there is always the possibility that the parasite could become even more readily transmitted between people (more infectious) and cause a more severe disease (more virulent) than it is now.

Conventional wisdom has it that virulence is naturally kept in check because the more virulent parasites kill their hosts – so they cannot be passed onto new hosts. But what happens when vaccination is introduced into this equation? How will the parasite respond to this powerful new selection pressure?

At the University of Edinburgh, Professor Read and his colleagues Margaret Mackinnon, Sylvain Gandon and Sean Nee are using both mathematical and animal models to investigate how natural selection can influence the infectiousness and virulence of parasites, and how interventions such as vaccination will affect these traits.

Vaccines are designed to protect the host from disease and death, boosting the immune system to kill parasites. But unless a malaria vaccine leads to the death of every single parasite, the ones that survive are likely to be the nastiest. And these more virulent parasites could continue to circulate in a vaccinated population.

"We could get a situation where we have a vaccine that is protective in clinical trials and goes into widespread use," says Professor Read. "But in the medium term, the evolution of virulence by the parasite erodes away benefit seen in clinical trials. This could mean more virulent strains are circulating in populations, so unvaccinated individuals are at greater risk."

At present this scenario is only a theory, not least because as yet there is no effective vaccine against malaria. Professor Read is fully supportive of vaccine development, but at the same time believes it is important to consider the medium-term impact of such vaccines, particularly the effects they might have on the transmission of parasites between people.

To test the impact of vaccination on malaria parasite virulence, Professor Read and his team use mathematical models that make assumptions about the public health aspects of malaria – such as transmission rates and the population densities of mosquitoes. These models can be used to assess the impact of vaccination, information for public health officials to consider.

To improve the mathematical model with data from natural situations, Margaret Mackinnon and Professor Read are using an animal model of malaria disease. They have found, for example, that virulent parasites are transmitted more readily than benign, less harmful parasites, an observation that supports the theory.

"There is a lot of interplay between the different models," he explains. "The information found in the animal model can be integrated into the mathematical model, improving the latter's ability to predict the impact of a vaccine."

Patients with malaria are often infected with more than one parasite strain, and one might expect that more virulent strains would outcompete weaker strains, reducing the transmission of the weaker strains onto new hosts. But this does not appear to be the case in the animal models: "In some cases, strains that are competitively suppressed transmit as well if not better than they would have done on their own," says Professor Read.

Quite why is not clear, so in his latest research, planned as part of his recently awarded Wellcome Trust programme grant, Professor Read hopes to understand the relationship between transmission success and competitive ability. Such information should help to the mathematical models predict how vaccination would influence virulence evolution in areas where mixed infections are common.

Professor Read realises that it will be difficult to know if the theory is correct until after the event: "Of course we don't have a malaria vaccine yet, so we are trying to predict something that will ultimately be tested in the field, something that'll be taking place over a few decades".

But with careful preparation, and analyses of pathogen virulence in animal and human diseases where vaccination has been widespread, when the vaccines do arrive they should be even more effective in preventing malaria.

"There really is an urgent need to focus on the transmission consequences of these vaccines, not just the disease-reducing component. We didn't pay any attention to the likely development of drug resistance even though it was obvious a long time ago, and now we have serious problems with it," he warns.

Professor Andrew Read is at the Institute of Cell, Animal and Population Biology, University of Edinburgh.

Page of 2; 2/9/04

[WTD023856] Modelling malaria virulence.doc