IAS Conference special
The 8th International AIDS Society Conference on HIV Pathogenesis, Treatment and Prevention (IAS 2015) is already being spoken of as one of those pivotal HIV conferences that mark a step-change in our ability to respond to the HIV epidemic. This puts it alongside the 1996 International AIDS Conference, also in Vancouver, where the effectiveness of antiretroviral therapy was first demonstrated, and the Durban conference in 2000, where getting treatment to Africa and the global south first really became possible. The 2015 conference has been termed the ‘90/90/90’ conference because of the presentation of the results from the START trial (see below) as well as further results from HPTN 052 (see Other recent news headlines). These results boost the UNAIDS 90/90/90 target, first announced at last year’s international conference in Melbourne. The target is that by 2020, 90% of all people living with HIV know their status: 90% of those are on treatment: and 90% of those have an undetectable viral load.
On the opening day of the conference, leading figures in the HIV response endorsed a call for immediate access to antiretroviral therapy for all people upon diagnosis with HIV. The Vancouver Consensus Statement calls for immediate access to antiretrovirals and for access to pre-exposure prophylaxis (PrEP) for those at high risk of HIV exposure, and urges rapid progress towards the implementation of new scientific evidence. As the Statement itself indicates, IAS 2015 could also be termed the conference of PrEP; many presentations reported on PrEP studies, almost universally showing that it is the people at highest risk of HIV who demonstrate high rates of interest in, and use of, PrEP. It could also be seen as an important conference for people who inject drugs, since many presentations featured really effective prevention and treatment work in this still critically neglected population.
There was general agreement that this conference was a first. It was the first global HIV conference where the new possibilities afforded by intensified prevention efforts were discussed in a co-ordinated way. In time, if translated into a coherent programme, these could end the HIV epidemic.
START trial finally clears way for HIV treatment for all
START enrolled 4685 adults living with HIV in 35 countries. People taking part in the trial had a CD4 count above 500 cells/mm3 when the trial started and were randomly assigned to either start treatment immediately or to delay it until their CD4 count fell below 350 cells/mm3 or they developed symptoms. There were less than half as many AIDS-related events, serious non-AIDS events and deaths in people starting ART immediately as in the people who delayed it, with significant reductions in TB and cancers. START also throws the door open to the wider use of HIV treatment as prevention.
At a workshop on the use of treatment as prevention before the main conference, evidence from large studies in Africa was presented showing that the UNAIDS 90/90/90 target (90% of people with HIV diagnosed, 90% of those on treatment, and 90% of those having an undetectable viral load) could be achieved. Early data from the SEARCH trial in Kenya and Uganda showed that 90% of adults in stable housing have accepted the offer of HIV testing, 93% of those diagnosed have been retained in care for at least six months, and 92% of those in care have a viral load below 400 copies/ml. However, another speaker at the conference from Rwanda, a country that has been praised for its high rates of treatment, said that prevalence remains substantially higher in female sex workers, men who have sex with men and other key populations. If services are not appropriate and adapted to these groups, they and their sexual partners will not get the benefits of HIV treatment.
Comment: What was surprising about START was not the fact that starting ART early was beneficial – that had been widely expected – but how beneficial it was, with a 50% reduction in death and serious illness, and that it was in cancers and TB, rather than heart disease and strokes, that we saw the significant reductions. The real importance of START is that it provided the final evidence necessary to convince agencies such as the World Health Organization and the British HIV Association (BHIVA) that offering ART on diagnosis is the best way to treat HIV. It also has big implications for prevention as it finally removes any remaining doubts about the ethics of expanding testing and treatment as an essential component of ending the HIV epidemic.
Non-daily PrEP will work for some
Pre-exposure prophylaxis (PrEP) was one of the major topics of discussion at IAS 2015.
Three presentations from different sites of the ADAPT (HPTN 067) study showed that, for some people in some settings, taking PrEP just twice a week or only when sex was anticipated, and adding in a post-exposure dose after sex if it actually happened, was a feasible and effective PrEP option.
Adherence to the twice-weekly PrEP regimen among a relatively well-informed and educated group of gay men in Thailand was as high as it was to daily PrEP. Adherence to the ‘event-driven’ regimen was lower everywhere, however, and among more deprived populations of men who have sex with men in Harlem, New York and women in South Africa, only about half of possible exposures to HIV were covered by either intermittent PrEP regimen. This compares with daily PrEP covering two-thirds of exposures in Harlem and three-quarters in Cape Town.
Qualitative studies, exploring barriers to and facilitators of adherence to PrEP in ADAPT, unearthed a variety of factors. Two barriers to intermittent-PrEP adherence were common and both concerned the post-sex dose: firstly participants reported fear that the sexual partner might see them taking PrEP and assume they had HIV; and secondly, participants reported that falling asleep, drinking or being away from home were all practical barriers to taking the post-sex dose.
The French Ipergay study, reported in February, also showed that intermittent PrEP was effective but this study allowed its participants 24 hours to take the post-sex dose rather than two hours as in ADAPT (and allowed another 24 hours to take a final one). In Ipergay, the participants took a double (rather than single) dose of Truvada before anticipated sex, another difference from the ADAPT study. The principal investigator of Ipergay, Jean-Michel Molina, presented data showing that both its post-sex doses and the pre-sex double dose were crucial to achieving drug levels sufficient for protection. In people new to PrEP, only one of the component drugs in Truvada – emtricitabine – might reach protective levels during the first 24 hours.
Another researcher, David Glidden, had established that levels of the other active drug – tenofovir – would reach higher levels in anal tissue than cervical tissue and maintain them longer. This could explain slightly lower efficacy rates in women and also indicate that women might be less likely to benefit from intermittent PrEP.
Comment: The results suggest some flexibility in the ways in which PrEP may be prescribed but need to be handled with caution. There needs to be further investigation into patterns of adherence in Ipergay to find out whether the high efficacy observed (86%) was because the regimen was sufficient for protection or because most participants were taking PrEP often enough for it to be equivalent to daily PrEP. And we need much better understanding of, and more studies of, PrEP in women. The TDF2 open-label study – see next story – is one.
Studies show that those most at risk of HIV have the highest adherence to PrEP
As well as the results of studies of intermittent dosing, IAS 2015 also heard results of demonstration studies in different parts of the world. The studies consistently found that people with the highest levels of risk were the most consistent users of PrEP.
The US Demo project reported on PrEP use in men who have sex with men (MSM) and transgender women in San Francisco, Miami and Washington DC. Average adherence was 85%, and the highest levels of adherence were reported by those who also reported condomless sex with two or more partners in the previous three months. Adherence was notably lower among participants from Miami (65%), who tended to be younger, were more likely to be black and to have somewhat lower levels of HIV-risk behaviour. The data showed that 97% of white participants had tenofovir levels in their blood samples indicating four or more doses a week, while 77% of Latinos and only 57% of black people did.
In Brazil, enrolment data from their demonstration project PrEP Brasil found that MSM who had had two or more condomless sexual partners were 80% more likely than men with lower levels of risk to apply to join the study. There were also higher than average levels of enrolment among men who have HIV-positive partners, among men who thought their risk of acquiring HIV over the next year was high, and among men who self-referred to the study (as opposed to being referred by doctors). In all, just over half of those eligible to join the study ended up enrolling but more than two-thirds of eligible transgender women enrolled. They were 64% more likely to enrol than cisgender MSM.
In Botswana, an open-label extension of the original TDF2 study, one of only two that have demonstrated efficacy in heterosexuals, found very high (near-100%) levels of adherence. They found no HIV infections where, based on the incidence seen in the original randomised study, four or six might have been expected. Adherence was somewhat lower in women (90%) and in people who experienced significant side-effects. Finally, as suggested by the Miami results above, young people may have particular difficulty adhering to PrEP.
The ATN (Adolescent Trials Network) 110 study recruited 200 young gay and bisexual men in twelve cities in the United States from an initial pool of 2000 who expressed interest. PrEP adherence of more than four doses a week started at 60% but had declined to 35% by the end of the study. There was a sharp racial divide in adherence: white and Latino participants maintained at least four doses a week throughout the study but black participants as a group never achieved the four-or-more doses a week level. By the end of the study the median drug level in black participants was scarcely above zero, indicating very little use of PrEP.
Comment: It is encouraging that people do seem to accurately judge their risk level and that those more at risk are more likely to take PrEP. This is especially important in countries with centralised health systems, as in Europe, where health ministries want reassurance that money spent on PrEP will not be wasted. It is less encouraging, though not a total surprise, that young people find it harder to take PrEP. But there needs to be more research into why adherence rates are so much lower among black MSM, given their very high rates of HIV. Is it to do with housing conditions? Health beliefs? Distrust of the medical profession? Fatalism? Only good qualitative research will find out.
Ukraine peer-educator project in people who use drugs almost halves HIV incidence
A study in Ukraine presented at the IAS conference showed that a peer education project reduced new HIV infections in people who inject drugs by 41%. For the study, people in recovery worked as outreach workers. They contacted and recruited 1205 HIV-negative people who inject drugs as peer trainers – even specifically ensuring that they did inject by checking for track marks and drug levels. They were trained to recruit and educate their peers on harm reduction practices. The training, led by the outreach workers, was scripted and involved role-play exercises. Each of the ‘peer leaders’ who had received the training was asked to bring two further people they knew who used drugs to the programme. By the end of one year, 18% of those receiving the peer group intervention had acquired HIV versus 32% in a control group.
Another Ukrainian study showed that HIV-positive people who inject drugs who received opiate substitution therapy (OST) were more likely to be engaged in HIV care and to receive antiretroviral therapy than people who did not. The two-thirds of a random sample of opiate users who were receiving OST were more likely to attend routine walk-in appointments and less likely to use emergency care; were more likely to have had a CD4 count in the last six months (82% versus 60%) and were more likely to be on antiretroviral therapy (37% versus 26%) despite having had HIV for the same average length of time (five years). Less than a third reported injecting drugs in the last month compared with 90% of non-OST receivers. Not surprisingly they spent vastly less on drugs in the previous month than those who did not receive OST (US $10 vs $397) with consequences for criminality and drug dealing – the average monthly wage in Ukraine is US $170.
A study from British Columbia, Canada, showed that providing OST to people who inject opiates reduced their risk of acquiring HIV compared to people who did not receive OST by 36%. And because OST is associated with lower rates of syringe borrowing and cocaine injecting, people who received OST were four times less likely to acquire HIV than people who did not: over four years, annual HIV incidence in people given OST was 0.6% compared with 2.25% in people not given it.
Finally, another study from British Columbia showed that both OST and HIV treatment were essential components of a programme to reduce mortality in people who inject drugs. Interestingly, OST worked rather counter-intuitively: drug deaths in people on OST but not HIV treatment were no lower, but HIV-related deaths were, probably because OST users had more medical contact. HIV treatment without OST reduced both HIV-related deaths and drug-related deaths: but offering both together had the biggest impact, reducing mortality by nearly 90%.
Comment: This summary only picks out four of an unusually large number of studies involving people who inject drugs and opiate users – previously a neglected key population. The pioneering work done in British Columbia has demonstrated that people who inject drugs can be engaged in HIV care and receive virally suppressive antiretroviral therapy as fully as any other group, and it is encouraging to see such models being extended to lower-income countries such as Ukraine. It is also encouraging to see that people who are still actively injecting drugs can be trained as peer educators for their own community. The fact, however, that a Donetsk centre involved in the OST study curtailed harm reduction services when it came under the control of pro-Russian separatists shows that this work faces a long struggle for acceptability.
World Health Organization urges more HIV testing for those who need it most
In order to achieve the ambitious 90-90-90 targets, improvements in HIV testing services will be needed. Many countries are lagging further behind on the target of diagnosing 90% of people living with HIV than on the other two targets.
In addition to recommending that routine HIV testing in medical settings be expanded to reach new groups of people, the World Health Organization’s (WHO) new guidance on HIV testing also recommends the delivery of HIV testing by non-medical ‘lay providers’, often in community settings. The guidance urges planners to make careful, strategic choices about which HIV testing interventions will be best able to reach individuals with undiagnosed HIV. New approaches may be needed to reach key populations.
Self-testing is not yet recommended. This is because the evidence base for its effectiveness and the best ways to provide it is still emerging. But WHO allowed substantial time for discussing self-testing at the launch of its testing guidelines at IAS 2015 and the approach may have an important place in future guidance. An important evidence gap relates to implementation in resource-limited settings with men who have sex with men, sex workers, people who inject drugs and other key populations. In places where pervasive social stigma and concerns about confidentiality make health services difficult to access, self-testing may have particular advantages for these groups in terms of privacy and autonomy.
A large study in Malawi, Zambia and Zimbabwe will distribute about 750,000 test kits. But though there are at least 20 self-testing studies among the general population in African countries, only six are among key populations. Studies of home testing among women in Kenya and transgender women in the US have found that people trade-off privacy and support, using self-testing to avoid being diagnosed within a hostile environment, yet often test in the presence of friends or partners for support. In a study of US gay men, those with access to self-testing tested for HIV more frequently, with 76% testing at least every three months (as the researchers had recommended), compared to 54% of those in the control group, with no differences in sexual behaviour.
Comment: Different ways of HIV testing work for different populations, and testing programmes probably always need to be conducted with an element of implementation research in mind, with a view to reducing poor diagnosis rates. Making sure that testing reaches high-incidence, key affected populations is key to its success as part of an HIV reduction strategy, especially in regions like central Europe where, as a recent study showed, testing rates in gay men remain unacceptably low. It is good that self-testing is now a possibility, but at present it is a technology without a clear indication for use: more research is needed to find out how best to use it and who would most benefit.
A third to a half of African people with HIV living in France acquired it there
Between a third and a half of African people living with HIV in France probably acquired HIV after they left Africa, according to a study presented to IAS 2015. Of 898 HIV-positive African migrants in France who had no record of a positive HIV test before entering France, 63 had had an HIV-negative test result while living in France and 28 only began having sex while in the country. Another 137 had been diagnosed more than eleven years after arriving in the country. Adding in people with HIV CD4 counts indicative of recent infection produced an estimate that at least 35% and possibly up to 49% acquired HIV in France. This finding accords with previous research in the UK that has calculated that around a third of African people with HIV in the UK acquired it after migration. Another study from Seattle in the US presented at the conference showed that 34 out of 112 HIV samples from African migrants could be linked to samples of other people living in the Seattle area, suggesting that transmission occurred locally. Some of these transmission clusters only included individuals of African descent living in the local area.
Comment: There are obvious implications for HIV prevention work with migrant and especially African communities throughout the US and Europe if more than a third are acquiring HIV in the country they migrate to. A Dutch study from as long ago as 2010 suggested that in some migrant African communities, the risk of HIV might actually be higher in their host country than their home country. This is partly because first-generation migrants may preferentially choose partners from their own population, thus concentrating HIV within that group. This is also one explanation also for the continuing high rates of HIV among black American gay men.
Eastern Europe has lower viral suppression rates than Africa
Some of the world’s richest countries are far short of achieving the UNAIDS 90-90-90 target, research from the UK shows. Progress is worst in Eastern Europe, where most countries lag behind average performance in sub-Saharan Africa on every indicator. The 90-90-90 target set by UNAIDS translates into undetectable viral load in 72.9% of all people living with HIV. How far are countries from achieving this target?
The research shows that Switzerland, Australia and the United Kingdom have the highest proportion of people living with HIV with undetectable viral load, with over 60% of the estimated population of people with HIV having an undetectable viral load.
Worldwide, the estimated proportion of people living with HIV who have an undetectable viral load is 32%. The US actually falls short of this, with an estimated 30%, even though it has one of the highest proportions of people diagnosed, at 86%. But some eastern European countries do even worse. In Estonia, while close to 90% of people with HIV are diagnosed, only 29% are on antiretroviral therapy (ART) and in Russia, only 11% are. There are some surprises, too. In France, a country often seen as having good healthcare, 52% of people living with HIV have an undetectable viral load – the same proportion as in Rwanda – and in British Columbia, proud of its work in getting HIV prevention and care to people who inject drugs, only 35% have an undetectable viral load.
Comment: Analysing the ‘treatment cascade’ in different countries is vital in order to uncover ‘break points’ where weaknesses in that particular country’s healthcare system prevent a properly co-ordinated and efficient response to HIV. Each country has its own barriers and, as this survey shows, they may produce surprising results. In Russia and Estonia, both the inefficient delivery of treatment and punitive attitudes towards people who use drugs may contribute to low treatment rates. In the US, the problem seems to be to do with distrust of, and difficulties in navigating, the healthcare system. In some African countries, HIV stigma and fear of testing, outdated perceptions of HIV treatment, and structural barriers such as distance to clinics may all play a part.
Other recent news headlines
Final follow up of the HPTN 052 study of treatment as prevention shows no evidence of HIV transmission from people with fully suppressed viral load to their partners, four years after the first results from the study demonstrated that early HIV treatment reduced the risk of HIV transmission by 96%. The final analysis reduces that estimate down to 93% after there were two more transmissions, one from a partner who had recently started treatment and did not yet have an undetectable viral load, and one from someone who had stopped treatment.
Younger gay men have much lower risk of acquiring HIV than previous generations
An innovative analysis of HIV trends in the Seattle area has found that gay men who were born in the early 1960s had the highest lifetime risk of acquiring HIV, but that this risk has declined dramatically in subsequent generations. While black gay men born in the 1970s and 1980s have a much lower risk of acquiring HIV than their forebears, racial disparities remain stark.
Same-day start to antiretroviral treatment leads to faster HIV suppression in San Francisco
A programme at San Francisco General Hospital that offers antiretroviral therapy (ART) on the same day as HIV diagnosis led to a high rate of treatment uptake and more rapid viral load suppression compared to standard practices. Participants given immediate ART achieved viral suppression in 56 days, compared with 119 days for those offered ART on the following appointment after lab results. After six months on ART, 95% of these participants achieved viral suppression, compared with 70% with the usual standard of care.
Intensification of antiretroviral therapy reduces risk of mother-to-child HIV transmission
A study from Thailand showed that maternal and infant antiretroviral therapy (ART) intensification – essentially, additional antiretrovirals for the mother and PrEP for the infant – is very effective in preventing HIV transmission during labour and birth in pregnant women who present late for care (with less than eight weeks of standard ART). Overall 88 mother/infant pairs took extra antiretrovirals in addition to the standard ART regimen and no transmissions were reported.