The impact of antigenic escape on the evolution of virulence

Understanding the evolutionary drivers determining the transmission rate and virulence of pathogens remains an important challenge for evolutionary theory with clear implications to the control of human, agricultural and wildlife infectious disease. Although disease is often very dynamic, classical theory examines the long-term outcome of evolution at equilibrium and, in simple models, typically predicts that R0 is maximized. For example, immune escape may lead to complex disease dynamics including repeated epidemics, fluctuating selection and diversification. Here we model the impact of antigenic drift and escape on the evolution of virulence and show analytically that these non-equilibrium dynamics select for more acute pathogens with higher virulence. Specifically, under antigenic drift and when partial cross immunity leads to antigenic escape, our analysis predicts the long-term maximization of the intrinsic growth rate of the parasite resulting in more acute and virulent pathogens than those predicted by classic R0 maximization. Furthermore, it follows that these pathogens will have a lower R0 leading to implications for epidemic, endemic behavior and control. Our analysis predicts both the timings and outcomes of antigenic shifts leading to repeated epidemics and predicts the increase in variation in both antigenicity and virulence before antigenic escape. There is considerable variation in the degree of antigenic escape that occurs across pathogens and our results may help to explain the difference in virulence between related pathogens most clearly seen in the human A, B and C influenzas. More generally our results show the importance of examining the evolutionary consequences of non-equilibrium dynamics.