A large international study using antiretroviral drugs (ARVs) as a prevention method has closed four years early because it has found that giving ARVs to HIV-positive people reduced the chance of them infecting their partner by 96% (see News in brief for more details).¹

The result of the HPTN 052 study puts the results of other HIV prevention trials in the shade. In recent years we’ve seen three trials showing that men who get circumcised are about 65% less likely to acquire HIV;^2,3 that prescribing daily Truvada (combined tenofovir and FTC) pills as pre-exposure prophylaxis (PrEP) to HIV-negative gay men reduced their risk of acquiring HIV by 42%;⁴ that giving women a microbicide gel containing tenofovir to use during sex reduced their risk by 39%;⁵ and even that, to everyone’s surprise, an HIV vaccine few thought would work reduced infections by 31%.⁶

All promising results, but not ones you’d want to base an entire HIV prevention strategy on. A twentyfold reduction in risk, on the other hand, is the sort of epidemic-halting result we’d like to see in a vaccine. And of course the great thing about the HPTN 052 concept is, as activists pointed out, that it’s almost “prevention for free”: many people who could be taking ARVs as a prevention method should be taking them as treatment anyway, or would be pretty soon. The fact that there was an 82% reduction in TB cases in positive partners who took early treatment underlines this. Buy treatment, get prevention thrown in.

All very exciting, although not entirely unexpected. More than a year ago, in February 2010, the Partners in Prevention study, which was designed to find out if treating herpes might reduce HIV transmission, found that the minority of its participants who started taking ARVs were 92% less likely to transmit HIV to their partners once they started therapy.⁷

So does HPTN 052 imply that, given these huge reductions in transmission, we should regard ARV treatment as essentially the answer to the HIV epidemic? No, for a number of reasons.

Firstly, we are having a hard enough time getting ARVs to people who need them as vital treatment. Currently just over 50% of diagnosed people in the world with a CD4 count below the old World Health Organization threshold of 200 cells/mm³ are getting them, and roughly a third of people with CD4 counts below the new threshold of 350 cells/mm³.⁸ Some people think it may be unaffordable to put more HIV-positive people on ARVs in order to reduce the risk of them infecting others, when their health would be okay for several years without taking them. We need some good cost-effectiveness studies to find out if the prevention bonus of putting more people on ARVs would justify the cost in the long run.

Secondly, although people have got the impression HPTN 052 was a solely heterosexual study, they did manage to recruit 38 gay couples, or 3% of the total. But that is not nearly enough to establish whether the same reduction in risk applies to gay men and, while we suspect it does, we still need the studies to provide data to convince doubters. The same applies to injecting drug users.

Thirdly, there was the sneaky fact that there were another eleven HIV infections in the study that came from people other than the primary partner (that’s 27% – an almost identical proportion to the 28% seen in Partners in Prevention). People ‘play away’, and, in many parts of the world, the main way HIV is transmitted is through casual sex, often from people who don’t know they have HIV.

This reminds us that you can put every diagnosed person in the world on ARVs, but if you don’t drive testing and diagnosis rates up to the point where the undiagnosed are a small minority, HIV will continue to be passed on.

For ‘treatment as prevention’ to start really working worldwide, we need to: increase rates of testing; reduce the proportion of undiagnosed people to a minimum; link the diagnosed to care; provide them with ARVs; support them to take them; and monitor their health to ensure they remain virally suppressed.

If any one of these six links in the chain is weak, the proportion of people with HIV who are virtually non-infectious due to ARVs will be a minority. Even in San Francisco, a city which is starting to provide evidence in the form of reduced diagnoses that treatment as prevention might work, the proportion of HIV-positive people in the city who are on ARVs and have viral loads below 50 copies/ml is less than 50%.⁹

So, people will say, what about good old-fashioned safer sex, and getting people to use condoms? After all, the absolute lowest-possible price of combination therapy in low-income countries (which involves regimens including the toxic d4T) is about $90 a year,¹⁰ whereas the price for one condom a day for 365 days, as supplied to donor programmes, would be about $9.¹¹

This would be fine, if people used condoms consistently. But although determined programmes have increased the rate of condom use in casual sex in many countries to well over 50%, it seems pretty impossible to get long-established partners to use them.¹² Condoms remain part of the answer, but if they were the whole answer, the epidemic would have finished by now. We probably have no alternative but to start using ARV drugs as prevention much more systematically.