RESEARCH ARTICLE Within-Host Stochastic Emergence Dynamics

Predicting the emergence of new pathogenic strains is a key goal of evolutionary epidemiol-ogy. However, the majority of existing studies have focussed on emergence at the popula-tion level, and not within a host. In particular, the coexistence of pre-existing and mutated strains triggers a heightened immune response due to the larger total pathogen population; this feedback can smother mutated strains before they reach an ample size and establish. Here, we extend previous work for measuring emergence probabilities in non-equilibrium populations, to within-host models of acute infections. We create a mathematical model to investigate the emergence probability of a fitter strain if it mutates from a self-limiting strain that is guaranteed to go extinct in the long-term. We show that ongoing immune cell prolifer-ation during the initial stages of infection causes a drastic reduction in the probability of emergence of mutated strains; we further outline how this effect can be accurately mea-sured. Further analysis of the model shows that, in the short-term, mutant strains that en-large their replication rate due to evolving an increased growth rate are more favoured than strains that suffer a lower immune-mediated death rate (‘immune tolerance’), as the latte