The research agenda for antiretroviral prevention – now it gets complex

This article is more than 13 years old. Click here for more recent articles on this topic

We now know that starting antiretroviral therapy early, pre-exposure prophylaxis (PrEP) and vaginal microbicides can all have an impact on HIV transmission, Victor de Gruttola told a satellite session at the International AIDS Society conference (IAS 2011) in Rome on Sunday. But researchers now need to do more than establish efficacy, he said.

Studies need to identify the mechanisms by which interventions do and do not work in different communities. They need to get to understand the characteristics of sexual networks, sexual behaviour and local epidemiology that influence their effectiveness. And they need to compare the impact of providing a stand-alone intervention with that of combined packages of interventions.

Other speakers at the satellite, which had been organised by AVAC and the European AIDS Treatment Group, emphasised the importance of implementation research – identifying barriers to the implementation of prevention interventions and developing strategies to overcome them.


mathematical models

A range of complex mathematical techniques which aim to simulate a sequence of likely future events, in order to estimate the impact of a health intervention or the spread of an infection.


A product (such as a gel or cream) that is being tested in HIV prevention research. It could be applied topically to genital surfaces to prevent or reduce the transmission of HIV during sexual intercourse. Microbicides might also take other forms, including films, suppositories, and slow-releasing sponges or vaginal rings.

pilot study

Small-scale, preliminary study, conducted to evaluate feasibility, time, cost, adverse events, and improve upon the design of a future full-scale research project.



The study of the causes of a disease, its distribution within a population, and measures for control and prevention. Epidemiology focuses on groups rather than individuals.


How well something works (in a research study). See also ‘effectiveness’.

Both Victor de Gruttola from the Harvard School of Public Health and Timothy Hallett from Imperial College London suggested there is no single best intervention – or even best package of interventions, but that this will depend on the characteristics of different communities and epidemics.

For different settings, researchers need to identify the combination of prevention interventions which could keep the spread of HIV under control. They also need to establish the breadth of programme coverage that is required.

Timothy Hallett presented some results from a basic mathematical model which aimed to identify the impact and cost of providing antiretroviral therapy to 80% of people at a number of different CD4 counts, PrEP to varying proportions of young people, PrEP to most people of all ages, or a combination thereof.

For each level of spending, Hallett identified the programme that would have the greatest impact – at the lowest levels of spending identified, this would be antiretroviral therapy alone. Should there be budget available to fund more than making therapy available for all with diagnosed HIV, policy makers should then provide PrEP for young people, and then for people of all ages.

But the model’s results change if baseline assumptions shift. If the costs of PrEP are actually lower than Hallett estimated (because drug prices come down), or if it turns out to be more expensive to get people diagnosed early and on to treatment (because testing promotion has less impact than anticipated or because new health services need to be provided), strategies with a greater reliance on PrEP would start to make more sense.

And the modelling studies need to consider other issues. Interventions – and combinations of interventions – will have different levels of effectiveness in different places, depending on a vast range of local factors which researchers are only beginning to get to grips with.

For example, Victor de Gruttola mentioned assortativity: the tendency for people who have many sexual partners to choose partners with the same characteristic. When this is the case, interventions will have less impact than when there is less assortativity.

Other important local factors are the number of transmissions that are due to people who are themselves recently infected, the proportion of people with HIV who are diagnosed and linked to care and the proportion of HIV-negative people who can be provided with an intervention.

Timothy Hallett noted that, although we know from the HPTN 052 trial that early initiation of treatment can reduce transmission to stable partners by 96%, this does not mean that changing treatment guidelines will bring about a 96% reduction in new infections.

Far too many people are diagnosed late for this to be possible. While early treatment strategies rely on early diagnosis, Sheena McCormack of the UK's Medical Research Council said that frequent HIV screening is not always an acceptable intervention.

New modelling work suggests that, even to achieve a 60% reduction in new infections through early treatment, testing would have to be so frequent that 60% are diagnosed within a year of infection, 90% of diagnosed people would have to be treated, 87% would need to be virally suppressed within six months of starting therapy, with only a 1% drop-out rate from treatment programmes.

Just minor modifications in these highly optimistic assumptions can wipe out the predicted impact. On the other hand, a combination of interventions would be more resilient in real-life conditions.

Should there not be the resources to make treatment available for everyone who needs it, its impact could be increased by prioritising its provision to people at higher risk of passing their infection on.

Sheena McCormack argued that the next prevention trials need to show that it is feasible to deliver interventions in a service setting, rather than with a great number of extra resources or with excessive demands placed on participants. Requirements for clinic visits, HIV tests and laboratory monitoring should be cut back, while users should be advised that PrEP may only be used around the time of sex, rather than on a daily basis. These measures will reduce the cost of interventions and increase their acceptability to users, she said.

She pointed out that a key question for a pilot PrEP study in the UK is whether a significant number of gay men are actually interested in taking it.

More acceptable interventions are more likely to be used consistently, and Sheena McCormack said that adherence is key to all the interventions discussed: “It’s all about behaviour,” she said. This applies as much to condom users as it does to people using a microbicide, PrEP or antiretroviral treatment.

This news report is also available in French.