We found that drug treatment could double the fitness of a resistant parasite or increase it 10 000-fold, depending on in-host ecology and choice of fitness measure. Conversely, in the absence of treatment, the absolute fitness of resistant parasites was halved or completely eliminated when sensitive parasites were present.
Background and objectives: The rate of evolution of drug resistance depends on the fitness of resistant pathogens. The fitness of resistant pathogens is reduced by competition with sensitive pathogens in untreated hosts and so enhanced by competitive release in drug-treated hosts. We set out to estimate the magnitude of those effects on a variety of fitness measures, hypothesizing that competitive suppression and competitive release would have larger impacts when resistance was rarer to begin with. Methodology: We infected mice with varying densities of drug-resistant Plasmodium chabaudi malaria parasites in a fixed density of drug-sensitive parasites and followed infection dynamics using strain-specific quantitative PCR. Results: Competition with susceptible parasites reduced the absolute fitness of resistant parasites by 50–100%. Drug treatment increased the absolute fitness from 2- to >10 000-fold. The ecological context and choice of fitness measure was responsible for the wide variation in those estimates. Initial population growth rates poorly predicted parasite abundance and transmission probabilities. Conclusions and implications: (i) The sensitivity of estimates of pathogen fitness to ecological context and choice of fitness measure make it difficult to derive field-relevant estimates of the fitness costs and benefits of resistance from experimental settings. (ii) Competitive suppression can be a key force preventing resistance from emerging when it is rare, as it is when it first arises. (iii) Drug treatment profoundly affects the fitness of resistance. Resistance evolution could be slowed by developing drug use policies that consider in-host competition.