Treatment as prevention tool: enough evidence to say it works, says IAS president

This article is more than 16 years old. Click here for more recent articles on this topic

While experts are at loggerheads over the recent statement that HIV-positive heterosexual individuals who have undetectable viral load and no other sexually transmitted infections, should not be considered at risk of transmitting HIV to others, there was a striking degree of unanimity among a range of experts at the XVII International AIDS Conference that expanding the number of people on antiretroviral treatment will reduce the number of new infections at a population level.

“We’re in a desperate race against time for prevention that works,” said former UNAIDS Ambassador Stephen Lewis, citing disappointing results from recent vaccine and microbicide trials, together with the long lead-time needed for structural and behavioural prevention interventions to translate into reductions in HIV prevalence at the population level, as indicators of the need for new thinking.

Interest has been galvanised by mathematical modelling of the potential prevention effects of much wider treatment, a greater appreciation of the role of undiagnosed individuals in HIV transmission, the move towards earlier treatment as a result of recent US and European guidelines, and more detailed information about the effects of antiretroviral drugs on virus levels in the genital tract.


second-line treatment

The second preferred therapy for a particular condition, used after first-line treatment fails or if a person cannot tolerate first-line drugs.

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.

drug resistance

A drug-resistant HIV strain is one which is less susceptible to the effects of one or more anti-HIV drugs because of an accumulation of HIV mutations in its genotype. Resistance can be the result of a poor adherence to treatment or of transmission of an already resistant virus.


A serodiscordant couple is one in which one partner has HIV and the other has not. Many people dislike this word as it implies disagreement or conflict. Alternative terms include mixed status, magnetic or serodifferent.


In everyday language, a general movement upwards or downwards (e.g. every year there are more HIV infections). When discussing statistics, a trend often describes an apparent difference between results that is not statistically significant. 

“We believe there is now enough evidence to say to policymakers that if you roll out HIV treatment with 100% coverage, you will see a reduction in HIV transmission,” said Professor Julio Montaner, incoming President of the International AIDS Society, the organiser of the International AIDS Conference.

But the move towards viewing treatment as a prevention tool should not be seen as a measure of desperation, said Professor Myron Cohen; “a lot of science went into making Julio Montaner’s statement.”

Preceding the study, substantial evidence had already emerged that antiretroviral therapy not only reduces virus levels in semen and vaginal fluid, but also substantially reduces onward transmission. In Taiwan for example, investigators estimated that HIV transmisison fell by 53% after antiretroviral therapy was introduced in 1997.

And in Uganda, the US Centers for Disease Control-led Home Based AIDS Care (HBAC) study has shown that HIV transmission can be drastically reduced when HIV-positive individuals are treated, their partners undergo counselling and testing, couples receive prevention counselling and the HIV-positive partner receives adherence support to maintain an undetectable viral load. Based on viral load and transmission data from a study in Rakai, Uganda, the investigators estimated that the interventions reduced new HIV infections by approximately 90% over three years – despite a trend towards an increase in unprotected sex with casual partners.

In July 2008 Professor Montaner and colleagues from the University of British Columbia Centre for Excellence in HIV published a mathematical model in the Journal of Infectious Diseases showing that two-thirds of projected HIV infections in British Columbia up to 2030 would be averted if every HIV-positive person in the province began treatment at a CD4 count of 350 cells/mm3, dramatically curtailing the future financial burden that lifelong HIV treatment would pose for governments.

The analysis persuaded the government of the Canadian province of British Columbia to aggressively expand antiretroviral therapy because of the prevention benefits of earlier treatment at the population level.

However, said Dr Warren O’Briain of the British Columbia Ministry of Health, the province’s government was persuaded of the need for more aggressive attempts to bring people onto treatment earlier on the basis of the right to health, not the economic argument.

In 2006 Dr Montaner co-authored a study based on mathematical modelling which concluded that providing antiretroviral therapy to everyone in the world infected with HIV would stop the HIV epidemic in its tracks within 50 years, at a cost of $7 billion a year.

But said Montaner on Tuesday, “no one here is talking about treating our way out of the epidemic – treatment is not enough.” Indeed, there has been much discussion at this conference of combination prevention technologies, and Highly Active HIV Prevention.

Nevertheless, “on the basis of personal experience, if we were able to say to a number of governments that treatment might reduce new infections by 30%, it would move things forward significantly between now and 2010 on the treatment front,” said Stephen Lewis.

Professor Myron Cohen of the University of North Carolina dismissed suggestions that the wider availability of treatment is directly responsible for an increased rate of new HIV infections among men who have sex with men in European countries such as Germany.

He said that the greater problem was the lack of diagnosis and treatment coverage, and the lack of treatment of sexually transmitted infections.

“If we were doing a good job with prevention in the US, the average CD4 count at diagnosis would be rising. It’s not. We’re doing a bad job at finding these people.”

Indeed, studies from the United Kingdom and the United States suggest that a very high proportion of new HIV infections are being generated by people who are undiagnosed – either because they are recently infected themselves, or because they have never been offered an HIV test by medical services (see, for example, this UK estimate). The introduction of more comprehensive opt-out HIV testing in the United States and the United Kingdom will have the secondary effect of reducing the number of undiagnosed and untreated people who can transmit HIV in the long-run.

Earlier treatment for HIV disease is now recommended in Europe and the United States, and a number of African countries are considering revising their treatment guidelines to recommend treatment when the CD4 cell count falls below 350 cells/mm3.

At present there is no compelling scientific case for treatment above the 350 cell threshold, guidelines’ committees agree, but if evidence emerged that treatment of individuals with CD4 counts above this level was safe and durable, and led to a reduction in onward transmission, earlier treatment might become attractive to governments if the additional cost led to substantial savings.

The HIV Prevention Trials Network study HPTN 052 is designed to answer this question. The study is randomising HIV-positive partners in 1750 serodiscordant couples to start treatment in the CD4 range 350-550, or to defer treatment to the threshold currently recommended by the World Health Organization (200 – 250 cells/mm3). The study will observe the rate of HIV transmission in these serodiscordant couples despite prevention counselling and condom provision The study is being carried out in Malawi, South Africa, India, Brazil and Thailand, and is designed to last five years, reporting by 2013.

Other studies are also likely to report findings on this question in the interim.

However, the big unknown question about the contribution of treatment to prevention is the extent to which it will have a durable effect in settings where viral load testing is not available to identify cases of treatment failure early. In these settings the lack of viral load testing may lead not only to HIV transmission due to undetected viral rebound, but also to the transmission of drug-resistant virus, compromising second-line treatment.

Evidence from Europe and North America accumulated between 1996 and 2003 shows that after the introduction of triple-drug combination therapy, there was a burst of sexual transmission of drug-resistant HIV that led to a peak prevalence of drug resistance between 2001-2003 of around 10-15% in untreated, recently diagnosed individuals. Many of these patients are likely to have acquired their drug-resistant HIV infection from sexual partners who had received sub-optimal treatment with a succession of drugs that drove the development of resistance.

But since then, the prevalence of drug resistance in newly infected people has fallen sharply because the proportion of patients with undetectable viral load has risen steadily. Some clinics in the United Kingdom now report that over 90% of their patients, including those on second-line or third-line treatment, now have undetectable viral load, limiting the pool of people with drug resistance and high viral load who can transmit HIV.

This trend suggests that it is not just a question of having viral load tests in order to prevent the transmission of drug-resistant HIV – it is also necessary to have access to a range of well-tolerated drugs for first- and second-line treatment. Until cheap point-of-care viral load tests and affordable second-line regimens are widely available in resource-limited settings, the prevention impact of treatment may be a promise rather than a solution for countries with hyper-epidemics.