A framework for assessing the risk of resistance for anti-malarials in development

Abstract

Resistance is a constant challenge for anti-infective drug development. Since they kill sensitive organisms, anti-infective agents are bound to exert an evolutionary pressure toward the emergence and spread of resistance mechanisms, if such resistance can arise by stochastic mutation events. New classes of medicines under development must be designed or selected to stay ahead in this vicious circle of resistance control. This involves both circumventing existing resistance mechanisms and selecting molecules which are resilient against the development and spread of resistance. Cell-based screening methods have led to a renaissance of new classes of anti-malarial medicines, offering us the potential to select and modify molecules based on their resistance potential. To that end, a standardized in vitro methodology to assess quantitatively these characteristics in Plasmodium falciparum during the early phases of the drug development process has been developed and is presented here. It allows the identification of anti-malarial compounds with overt resistance risks and the prioritization of the most robust ones. The integration of this strategy in later stages of development, registration, and deployment is also discussed.

Figure 1

Resistance risk assessment workflow. The resistance risk assessment workflow encompasses three goals: cross-resistance determination (goal I), de novo resistance selection frequency determination (goal II), and resistance mode-of-action determination (goal III). These can be achieved trough a straightforward set of quantitative experiments applied to compounds at the lead and preclinical developmental stages. A resistant IC₅₀ corresponds to a 20-fold increase as compared to a fully sensitive strain (NF54 or HB3 in the case of sulphonamides). C is the theoretical cost of fitness associated with resistance (see main text). C<0 indicates that resistance provides a fitness advantage, which is a major risk factor. Ultimately, the overall risk level can be classified as low, elevated, or major and allows to prioritize the development of robust compounds and to establish risk mitigation strategies for the others.

Figure 2

Known genetic determinants of naturally occurring resistance mechanisms. Mutations (red dot) of the dihydrofolate reductase (PfDHFR) enzyme prevent its inhibition by the antifolate drugs pyrimethamine (PYR) and cycloguanil (CYC). Similarly, sulphadoxine (SDX) resistance is mediated by mutations of its target dihydropteroate synthetase (PfDHPS). Atovaquone (ATO) binds to the cytochrome bc₁ complex (PfCYTB), mutations of which have been shown to induce high level of ATO resistance. Chloroquine (CHQ) is believed to prevent haeme detoxification within the digestive vacuole. Mutations of the CHQ resistance transporter (PfCRT) as well as of the multidrug resistance protein-1 (PfMDR1), including copy number variations, have been shown to compromise CHQ action by preventing its accumulation within the digestive vacuole. Mutations of these two transporters have also been implicated with mefloquine resistance, although definite marker has not been established for this drug.

Figure 3

In vitro resistance selection assessment. (a) A standard in vitro protocol for resistance selection frequency measurement uses defined starting inocula of a P. falciparum strain pressured with a constant level of drug nearing the IC₉₀. Parasitemia falls below detection limits but eventual resistant parasites are able to recrudesce and to be cloned for subsequent determination of the IC₅₀ fold increase. The minimal inoculum for resistance (MIR) is a measure of the resistance selection frequency, while the IC₅₀ fold increase measures the level of resistance. (b) These two endpoints are used to classify anti-malarial compounds according to risk levels (see main text). It is advisable to run control experiments in parallel with compounds known to select resistance readily, such as atovaquone.

Figure 4

Resistance profiling and clinical development.In vitro selection experiments typically generate resistant parasites from which resistance markers can be identified. This permits the identification of more robust combinations by assessing acquired and de novo cross-resistance studies with parasites already resistant to potential partner drugs. Resistance markers can be monitored during Phase II and III to include resistance selection as a clinical factor and to insure the appropriate resistance data package for registration. Post-marketing surveillance will also directly benefit from the a priori knowledge of resistance markers.