Early HIV Infection in the United States: A Virus's Eye View

Early after infection with HIV, the concentration of virus in the body increases rapidly before an immune system response begins to hold it under temporary control [1]. During that short time of elevated viral concentration, the infected individual may be much more infectious than at other times [2,3]. At advanced infection, viral concentration can increase again, potentially leading to a late surge in infectiousness. However, the actual influence of each stage of infection on the onward spread of HIV does not depend just on the biology of the infection but also on the patterns of sexual partnership formation through which transmission can occur. Understanding the contributions of the different phases of infection to the onward spread of HIV is essential in planning effective interventions to control the epidemic. To estimate these contributions epidemiologists have previously resorted to mathematical models, but uncertainties in key parameters, as well as variability between populations, have contributed to a very wide range of model estimates of what proportion of transmissions occur during different stages—for early HIV infection (EHI), estimates range from less than 5% to more than 90% [4]. Therefore, new data on this topic are of substantial interest, and in this issue of PLOS Medicine, Erik Volz and colleagues [4] present a novel approach that opportunistically leverages genetic sequence data from a population of men who have sex with men in Detroit, Michigan, to add significantly to our understanding of HIV epidemic dynamics. The basic premise of using genetic sequence data is that within an individual early in infection, the population of virus is relatively homogenous and similar to that in the individual that infected him. Over time, however, greater diversity in the genetic make-up of the virus population develops, and this will be occurring independently in the ‘‘donor’’ and ‘‘recipient’’. This process gives rise to stereotypical patterns of viral diversity across a population that depend on how much transmission happens early in infection. By constructing a model of that process and fitting it to the genetic sequence data and other demographic and epidemiological information, Volz et al. [4] can ‘‘back into’’ an estimate of how much transmission occurs during the first year of infection. The major theoretical development of this study has been in describing how to appropriately combine those diverse sources of data, which is essential for fitting a model in such a way that inferences can be drawn on model parameters, such as the rate of transmission from different phases of infection. The analysis by Volz and colleagues finds that during the first year of HIV infection individuals are eight times as infectious as during chronic infection. These estimates appear similar to independent estimates from cohort studies (making allowance for the different definitions of EHI used) [3,5]. This pattern of infectiousness means that, according to the Volz et al. model, 42%–46% of transmissions come from persons in EHI, which is also consistent with expectations from other estimates [1]. Recognising EHI’s large influence on the dynamics of HIV in the sort of epidemic seen in this population could have practical significance in the response to epidemics. First, it means that partnernotification approaches could be particularly useful, as they could lead to the discovery of undiagnosed infections at a time when they are most infectious [6]. A concentration of the transmission potential of the virus early in infection also modulates the types of sexual behaviours that are important for sustaining HIV transmission—in particular, short gap lengths between partners, or concurrent sexual partners, would be expected to be more important for the spread of HIV than if transmission rates did not change over the Linked Research Article This Perspective discusses the following new study published in PLOS Medicine: