Pharmacological modelling and data analysis in decision support of antimalarial drug dosing regimens

Successful treatment of an individual infected with malaria means that the drug resolves their symptoms and clears all parasites from their blood without causing side effects that harm the patient or make them hesitant to use the drug in the future. Treatment success at population level means that the drug is effective and safe, develops a good reputation that optimises patient compliance, reduces malaria transmission from treated patients, reduces the likelihood of new drug resistant mutations arising, and slows the rate of spread of any existing resistant parasites. A clear, explicit methodology is needed to design and optimise dosing regimens to achieve these aims.

Treatment outcome rests on a complex variety of factors, including natural variability in how patients processes the drug (PK), variation in parasite sensitivity to the drug (PD) and variation in patient drug intake depending on their weight, age or height, and patient compliance. Recent developments have resulted in increasingly powerful and accurate simulation models for malaria which has allowed us to simulate drug treatment more accurately. As these models mature it is increasingly recognised that an integrated modelling method that combines different sources of data will help support the development of optimal drug combinations and dosing regimens for new and currently implemented antimalarial drugs.

We have been developing and refining simulation models of antimalarial drug treatment as one of the partners of a large malaria modelling research consortium, OpenMalaria, to the state that they can now successfully capture the broad patterns of variation in treatment outcome noted in real-life. We now propose to further refine and calibrate this methodology, and specifically develop them to combine data from several field and clinical sources to address operational questions surrounding deployment of the latest generation of antimalarial drugs, the artemisinin-combination therapies (ACTs). We do this with a team of experts that range in expertise from modelling, pharmacology, clinical and policy knowledge on antimalarials to ensure that the models are developed according to the needs of drug developers and policy makers.

Antimalarial drug treatment failures have a complicated aetiology associated with a range of factors, including human pharmacokinetic variation, parasites variation in drug sensitivity, dosing regimens and adherence, and variability caused by dosing patients based on weight/age bands. It is increasingly recognized that models which incorporate these sources of variation have the potential to support and guide the drug implementation process through identification of a drug?s future therapeutic potential.

Over the past two years, scientists at LSTM have made good progress in this area. As one of the partners of a large malaria modelling research consortium, Dr Hastings has been developing and refining pharmacokinetic/ pharmacodynamic (PK/PD) models of antimalarial drug treatment to the state that they can now successfully capture the broad quantitative patterns noted in the field and clinic. Similarly, Dr Terlouw has led the development of weight-for-age reference curves that are representative of patient populations in the endemic areas of Africa, Asia and Latin America, and has used this to identify optimal age-based dosing regimens, and the associated level of under and overdosing at population level.

We propose to further refine, integrate and calibrate the two research agendas to generate a tool that can address operational questions surrounding deployment of the latest generation of antimalarial drugs, the artemisinin-combination therapies (ACTs). Specifically we propose: (i) We shall incorporate drug absorption and conversion phases; these are important determinants of peak, possible toxic, drug concentrations and occur on important timescales for short half-life artemisinins. (ii) Malaria parasites are insensitive to drugs in certain stages of their lifecycle, and may even enter ?dormant? stages that are unaffected by drugs; we shall develop a methodology to allow for these stages of insensitivity. (iii) We shall develop methodologies to optimise the design of age/height dosing bands. (iv) We shall explicitly calibrate PK/PD variation for the current generation of antimalarial drugs (the ACTs) to investigate the dynamics of drug resistance.

Placing these factors in an explicit PK/PD framework s constitutes a significant advance on current modelling strategies for antimalarial drugs and has the potential to guide and optimise several key aspects of antimalarial deployment in resource-poor settings.