"Evolutionary rescue" is a term used primarily in conservation biology to describe the situation in which genetic adaptation prevents extinction of a declining population. This scenario arises, for example, in macro-organisms facing habitat destruction or climate change. It also arises in micro-organisms and cell populations, including pathogens or cancerous tumors in a patient taking drug therapy. The emergence of drug resistance, which can be seen as evolutionary rescue of a pathogen population facing severe environmental change, is a major problem in public health, compromising successful treatment of infections. Here I will present new theoretical results on the within-host emergence of drug resistance in viruses. We are particularly interested in whether rescue occurs from pre-existing (standing) genetic variation, or de novo mutation during therapy. I will describe an extension of a widely-used viral dynamics model, which allows us to investigate the roles of viral "life history", competition between strains, and stochasticity. Analytical approximations for the probability of rescue can be derived from a
simplied stochastic process description, allowing identication of the parameters playing a key role in determining rescue. Finally, I will discuss some links between the largely separate drug resistance and evolutionary rescue literature and the potential for greater integration between these elds.