Context
Proven HIV prevention methods are becoming increasingly available worldwide, yet there is little consensus on how to effectively deploy them, or how best to evaluate their impact. It is unknown whether treatment as prevention can be effectively targeted at either high-risk individuals or subpopulations, or what the contributions to the epidemic of different risk groups are. The role of acute and early infection is still disputed, particularly for heterosexual epidemics, and there is an increasing realisation that movement and migration may be an important epidemic driver.
Sources and Sinks
An epidemic can be described by a model of connected sources, sinks, and hubs. Sources are groups in a population that disproportionately pass on infections, sinks are groups that disproportionately receive
infections, and hubs are both sources and sinks. Population groups can be defined by age, gender, riskiness of sexual behaviour, geography, occupation, cultural preferences and norms, migrational behaviour, or other characteristics and combinations thereof. Identifying these groups allows prevention to be tailored.
Phylogenetics
The most powerful tool available to study HIV transmission is phylogenetics. A viral phylogeny reveals who is close to whom in the transmission network. With additional epidemiological information or modelling one can identify which of these close individuals are likely sources of new infections. Following theoretical work on incorporating within-host diversity into phylogenetic inference, we have developed a tool called phyloscanner that can be used for source attribution in HIV next generation sequencing (NGS) data. Once sources have been identified with quantified uncertainty, epidemiological questions about the characteristics of transmitters can be addressed. For example, transmission networks will enable us to detect individuals that are at high risk of infecting others and quantify their contribution to transmission.
Themes
The project covers four interlinked themes: 1. Molecular Epidemiology and Mathematical Modelling, 2. Phylodynamics, 3. Mobility and Migration, and 4. Clinical Science, Drug Resistance and Ethics. The four themes are each represented by one of the four PIs.
Molecular Epidemiology and Mathematical Modelling (Christophe Fraser)
This theme is building a transmission network from the PANGEA sequences, to answer several epidemiological questions, e.g. which demographical, clinical and virological correlates identify transmitters, whether stratifying the analysis by age and gender reveals transmission cycles across different age groups, and how drug-sensitive and drug-resistant viruses are transmitted. Mathematical modelling is a key tool for designing public health interventions, but several uncertainties have been difficult to resolve in HIV models, e.g. infectiousness of acute infection, transmission patterns between age and sex groups, heterogeneity of transmission rates and the contribution of hidden risk populations. As new technologies result in more costly interventions, it becomes ever more important to estimate key parameters using phylogenetics to prioritise target groups, to assess the impact of interventions and to use models to predict their likely effect.
Phylodynamics (Andrew Rambaut)
This theme takes the analysis of phylogenetic trees a step further, looking at spatial movement patterns of the virus across cohorts and countries and the identification of ‘outbreaks’, unusually dense transmission clusters in geographically or demographically defined groups which might indicate concentrated epidemics within the epidemic. The theme will also dig deeper into the trees, linking the branching length to time, thereby identifying core groups that disproportionally spread the epidemic and assessing the impact that the stage of infection has on onward transmission. The Phylodynamics team also works on visualising the results and on further developing the analysis software used in this project. They will adopt existing phylodynamics tools to account for genetic variation and for use with large datasets.
Mobility and Migration (Kate Grabowski)
This theme investigates how mobility and migration influences the spread of the HIV epidemic in different African countries. There is growing evidence that mobile populations are at higher risk of for HIV and can sustain epidemics by not being eligible for treatment. Analysing mobility can also inform us to what extent high prevalence areas fuel epidemics in surrounding areas of low prevalence. When this happens in areas of population trials (which is the case for three of the PANGEA sites), this might impact on the interpretation of the results. Phylogenetics offers a unique way to track the spread of the virus along with mobile populations and define local sources, sinks and hubs.
Clinical Science, Drug Resistance and Ethics (Deenan Pillay)
This theme investigates why incidence in many parts of Africa is not falling as fast as expected despite a dramatic increase of antiretroviral coverage and how new prevention approaches will impact on the epidemic. It also monitors to what extent the rapidly expanding use or antiretroviral therapy is associated with the emergence of HIV drug resistance. The phylogenetic approach will make it possible to find the groups that are still at high risk of HIV infection and target them in prevention approaches. Finally, this theme will look at the ethical issues around phylogenetic studies in a highly stigmatised disease like HIV and how the risks and benefits can best be explained to study participants.
Social Sciences
We have collected thousands of samplesd and behind every sample there is a life and a story. We are keen to attract more social scientists, especially those working in the countires covered by PANGEA, to help us bring out the stories and inform the behavioural aspects of our mathematical models. Limited funding for small-scale projects is available via PANGEA and we will also help you to apply for external funding. For more information please contact project manager Lucie Abeler-Dörner (lucie.abeler-dorner@bdi.ox.ac.uk).