icon-    folder.gif   Conference Reports for NATAP  
 
  International AIDS Conference
Durban, South Africa
July 18-22 2016
Back grey_arrow_rt.gif
 
 
 
HIV Prevention at AIDS 2016 21st International AIDS Conference Durban, South Africa 18-22 July 2016
 
 
  Jared Baeten, MD PhD
Connie Celum, MD MPH
University of Washington
 
HIV prevention has played a central role in every international HIV conference, and this year's 21st International AIDS Conference continued that trend, with this year's theme "access equity rights now." As at all International AIDS Conferences, the meeting was tremendously multidisciplinary, with sessions on laboratory science, clinical care, epidemiology, policy and program work, advocacy, and social justice and scientific contributions across the spectrum from basic, behavioral, social, and clinical sciences. As has been the case at recent International AIDS Conferences, much of the scientific program is available online, including abstracts, copies of slides, and webcasts of some session (searchable program available at http://programme.aids2016.org/).
 
Prevention was front and center at this year's meeting. Just last fall, the World Health Organization issued updated guidance recommending antiretroviral treatment (ART) for all persons living with HIV infection, regardless of disease stage, as a prevention intervention, and pre-exposure prophylaxis (PrEP) as an option for HIV uninfected but at-risk persons. Several countries, including the US, Canada, France and the EU, Kenya, South Africa, and Peru, have made regulatory decisions that open the door for PrEP to be a major part of their prevention efforts, and UNAIDS has set an ambitious goal of 3 million persons on PrEP by 2020. Roll-out of PrEP has been slow, however, with substantial numbers of persons taking PrEP only in the US. For ART for HIV prevention, implementation has been piecemeal and far too many persons fall off the cascade of HIV care, either because of not testing in the first place, not initiating or sustaining engagement in care, or not taking ART to achieve viral suppression. Thus, there is much work to be done, and AIDS 2016 included a number of sessions focused on operational work and implementation science, aimed at disseminating best strategies for delivery of PrEP, ART, and other highly effective prevention interventions.
 
AIDS 2016 also marked the return of the International AIDS Conference to Durban. The last time the meeting was held here, in 2000, ART was virtually unavailable in Africa, the continent hardest hit by the epidemic, and in other low and middle income countries. As a result of tremendous activism, vision, and energy, AIDS 2000 marked the turning point in global availability of treatment; AIDS 2016 was cast by many as doing the same for HIV prevention. Plenary talks on the state of the epidemic (TUPL0102), HIV vaccines (WEPL0104), PMTCT (FRPL0103), and looking forward (FRPL0105) all had strong prevention components.
 
PrEP
 
PrEP (primarily oral tenofovir-based pills, as tenofovir disoproxil fumarate [TDF] alone or in combination with emtricitabine [FTC/TDF], the only approach to date to have received regulatory approval in any country) received considerable attention at this year's IAS meeting - arguably the most attention of any prevention intervention. A recurring theme was that HIV treatment will not be enough to end the global HIV epidemic, so potent interventions like PrEP must be a part of highly-effective combination prevention.
 
Delivery of PrEP. Updated results from the IPERGAY randomized trial were presented (Molina, WEAC0102). IPERGAY was a placebo-controlled trial of FTC/TDF PrEP, prescribed for "on-demand" use - i.e., 2 pills 2-24 hours before sex and 2 pills (one each day) for the 2 days after sex. Initial results for the trial were reported at CROI in 2015 and showed an 86% reduction in HIV risk (95% CI 50-98). After these results were demonstrated, the trial converted to an open-label extension phase, providing all men with active FTC/TDF PrEP. A total of 362 participants were eligible for the open-label phase, and 299 (83%) completed the 18 months of follow-up offered. Only one HIV infection occurred in the open-label phase, at an incidence of 0.19 per 100 person-years (95% CI 0.01-1.08) - a 97% reduction to the incidence seen in the placebo arm of the study previously. The single infection was not using PrEP. Condom use was low (∼30% with last receptive anal sex) and went down somewhat during the course of the open-label phase (see also Teyssier, WEPEC263); STI rates were very high throughout (35-40 per 100 person-years), but no higher in the open-label phase. These results emphasize how powerful PrEP can be as a prevention strategy in high-risk men. High STI rates need to be addressed but they also emphasize how much HIV prevention is being gained through PrEP use.
 
Updated results were also presented from another open-label study. The Partners Demonstration Project was an open-label interventional study providing integrated delivery of PrEP and ART to heterosexual HIV serodiscordant couples in Kenya and Uganda (Baeten, WEAC0105). The study responds to calls for demonstration projects of PrEP to determine how best to deliver this new prevention strategy. A total of 1013 couples were enrolled in the project, beginning in November 2012 and couples were followed for 2 years each, until June of 2016. Couples were specifically recruited if they had characteristics indicating higher HIV risk - such as younger age, no children within the partnership (which may result in risk-taking behavior), limited condom use, and a higher viral load in the HIV infected partner. In the project, PrEP was offered to the HIV uninfected partner "as a bridge" to ART in the infected partner - i.e., during the period when the infected partner is not on ART (due to individual or systems delays in starting) and for the first six months after ART is started (to give time for ART to fully suppress viral replication). Thus, this population was recruited to be at considerably higher risk than previous studies of HIV serodiscordant couples. As reported at CROI 2015, PrEP uptake (97%) and adherence (>80% by multiple measures, including detection of tenofovir in plasma) were very high in the population and initial HIV incidence was low (0.2 per 100 person-years). At AIDS 2016, final results were reported. PrEP use remained high until the end of the study, and ART use was also high - more than 90% initiated ART by the end of follow-up and viral suppression was ∼90%. Only 4 HIV infections occurred overall - and all 4 were not using PrEP (either because they had declined or they had no PrEP detected in blood samples). Using statistical modeling to compare this rate to an anticipated HIV incidence based on a validated risk scoring tool for HIV serodiscordant couples, a total of 83 HIV transmissions would have been expected, at an incidence of 4.9 per 100 person-years (which is several times greater than the incidence seen in the control arm of HPTN 052, for example). This difference between 39.7 and 2 is a 96% reduction (95% CI 87-98, p<0.0001). These results suggest that this public health approach to delivering PrEP - offering it to the uninfected member of a couple until his or her partner has an undetectable viral load - results in essential elimination of HIV transmission in this key population. Qualitative work from the same study found (Wyatt, FRAE0102) that participants had confidence in PrEP, but some worries about stopping PrEP when they had crossed the "bridge," which seemed particularly influenced by incomplete confidence in the protective effects of ART for HIV prevention. These results emphasize the importance of continuing to develop clear messaging about how different prevention interventions can work together.
 
From San Francisco, a program delivering PrEP to men who have sex with men and transgender women (PrEP Health Program, San Francisco AIDS Foundation) has had high interest: nearly 1200 enrolled, retention ∼80% for a year or more, and 90+% self-reported adherence. No HIV infections have been observed (Crouch, FRAE0104). High uptake has also been seen in New South Wales, Australia, with a PrEP program launched earlier this year (Zablotska, FRAE0105). In another study from the US, a small pilot study was described that aims to simplify PrEP delivery by providing PrEP at home, accompanied by a kit containing counseling materials, self-testing for STIs and HIV, and request for refills (Siegler, FRAE0101). The study just completed a formative phase and early testing in a couple of dozen participants so further results will be highly anticipated.
 
The HPTN 073 study was a multisite demonstration project of PrEP for gay black men in the US. Two oral abstracts from HPTN 073 described recruitment and engagement strategies (Hucks-Ortiz, WEAC0103) and adherence (Wheeler, WEAC0104). Using a culturally-targeted strategy, 226 men enrolled, about 80% started PrEP, and about 60% had PrEP levels in blood showing consistent use. HIV incidence was 2.9 per 100 person-years in those receiving PrEP (5 infections, of which 2 were known to have discontinued PrEP and adherence for the others was not reported), compared to an incidence of 7.7 per 100 person-years for those who never accepted PrEP.
 
Young men, aged 15-17, from the US participating in a safety and adherence study of PrEP (ATN 113) had great interest in PrEP, ongoing need (measured HIV incidence of 6% per year), and few side effects (Hosek, TUAX0104LB). Adherence fell off after the first 12 weeks of follow-up, when visits went from monthly to quarterly, emphasizing that young persons may need additional attention to adhere to PrEP.
 
The US has had the greatest experience delivering PrEP to date and new data from Gilead Sciences, the manufacturer of FTC/TDF pills, showed how uptake continues to increase (McCallister, TUAX0105LB). The authors used pharmacy fill records, trying to exclude persons with known HIV infection, receipt of antiretrovirals for post-exposure prophylaxis, HIV treatment, or hepatitis B infection treatment, to narrow in on those likely to be receiving PrEP. Through the end of 2015, an estimated 79,684 persons have received PrEP, a 738% increase from 2012 to 2015. The highest use, on a population basis, is in the states of Massachusetts, New York, and Illinois, and men (76%) and those 25 years and up (85%) dominate prescriptions. Thus, PrEP use has increased substantially in the US, but work is needed to make sure PrEP is getting to those at greatest risk.
 
Thinking about (and rethinking) PrEP safety. Because PrEP is for persons who do not have HIV (and indeed might never get HIV, even without PrEP, as being at risk does not necessarily guarantee infection), its safety must be high. PrEP was demonstrated to be safe in the clinical trials that proved its effectiveness for HIV prevention; thus, continuing studies of PrEP safety are addressing more nuanced aspects, especially those related to safe delivery of PrEP and to policy decisions. Towards this last point, a Tuesday abstract summarized a systematic review of data from randomized, placebo-controlled trials of PrEP related to antiretroviral resistance, which is an important consideration for many persons making PrEP policies (Grant, TUAC0104). The conclusions were powerful: resistance in persons acquiring HIV in PrEP trials was very rare, limited almost exclusively to persons with undetected acute HIV infection at the time of PrEP initiation and mostly related to FTC (the M184V mutation) rather than to TDF itself. This is because most individuals who acquired HIV while receiving PrEP had adherence too low to achieve HIV prevention (or to select for resistance), the exception was those already infected at initiation (who then received an insufficiently-suppressive antiretroviral regimen), and resistance to TDF has a higher bar than resistance to FTC. In the review of all data, PrEP prevented at least 8 HIV infections for every case of resistance that arose, and over 20 when acute infections at PrEP initiation were excluded - these are the kinds of numbers that help guide policy. One example of screening for acute infection (from the iPrEx OLE study) was presented - using symptoms could potentially identify those with acute infection but also had low positive predictive value; the authors concluded that symptom screening is not required to achieve a favorable risk/benefit ratio for PrEP delivery.
 
Renal safety is particularly important for TDF-based PrEP. In HIV infected persons, TDF has been associated with renal toxicity, relatively rarely and mostly mild but sometimes more severe. In clinical studies of PrEP, renal toxicity has been extremely uncommon, although small changes in glomerular filtration rate have been seen but not necessarily of clinical significance and all resolving with PrEP discontinuatino. Renal safety monitoring for PrEP is an open question - CDC guidelines in US recommend 6-monthly, although clinical trials tended to do 3-monthly, and implementation of PrEP in low-income settings might find even 6-monthly monitoring to be challenging. In a late-breaker abstract (Mugwanya, presented by Heffron, FRAE0106LB), data from two large PrEP studies, one that did 3-monthly monitoring and one that did 6-monthly monitoring, were compared. Both were done in Africa, among men and women, and both required normal renal function at baseline. The frequency of renal function decline - to <60 mL/min or a 1.5-fold change from baseline - was uncommon (<1% of persons over a year of PrEP use) and was no different with 3- vs. 6-monthly monitoring. Risk factors for creatinine clearance <60 mL/min included age ≥45 years, creatinine clearance between 60 and 90 mL/min, and weight ≤;55 kg, although renal decline was still rare even in those with these factors. In summary, these data suggest that 6-monthly monitoring of PrEP renal safety is a safe approach, that risk factors may identify those who need monitoring, and that consideration might be made for even less frequent (or no) monitoring in those without risk factors.
 
Two symposium sessions related to PrEP safety in pregnancy, breastfeeding, and in the periconception period were excellent and deserve watching online (MOSA35 and WESY06)
 
New PrEP agents. TDF-based pills are not likely to remain the only PrEP option for long. This is important - like for other prevention strategies (a strong analogy being contraception), choice is important and to have choice, one needs to have options. Earlier this year, two phase III trials among women in Africa demonstrated that an antiretroviral-containing vaginal ring (the dapivirine vaginal ring) was effective and safe for HIV prevention, and phase I-III studies of other potential options are ongoing. Several of these were discussed at AIDS 2016.
 
Maraviroc-based pills have been explored as an alternative to TDF-based pills, in the phase II study HPTN 069/ACTG A5305. Maraviroc (MVC) is a CCR5 antagonist that is used, although not tremendously commonly, in the treatment of HIV infection. The trial randomized men and women in the US to daily MVC alone, MVC + FTC, MVC + TDF, or TDF + FTC, for 48 weeks. The male data were presented at CROI earlier this year and found that all four medication approaches were safe, and side effects (including gastrointestinal and renal side effects) were comparable among them all; only 5 HIV infections occurred but two of them may have been breakthroughs of MVC. At AIDS 2016, data from women were presented (Gulick, TUAC0102). A total of 188 women enrolled and 160 (85%) completed the study, although some who completed the study discontinued treatment early (n=33). Adverse events did not differ across the study arms. Pharmacokinetic testing demonstrated detectable drug in plasma in 65% at week 24 and 60% at week 48, and this was similar across arms. There were no HIV infections in the study (incidence 0%, 95% CI 0-2.5%), although this is not surprising as other studies of women in the US have shown annual incidences well under 1%. A second MVC abstract (Broca-Coffano, TUAC0101) found that MVC did not successfully prevent oral acquisition of SIV in infant macaques (a model for breastmilk transmission of HIV). Whether MVC will really become an alternative to TDF-based pills for PrEP is to be seen - the initial data are supportive in terms of safety although potentially questionable for prevention potency, and more study is absolutely needed before MVC could be reasonably prescribed for PrEP.
 
Long-acting injectable agents - that would work for a month, two, or more - offer substantial promise for becoming a PrEP option. Injectable agents have great appeal - they could be used discretely and could remove some adherence challenges (e.g., taking a daily pill, but not returning to a clinic to get HIV testing and a PrEP refill). Phase II and III studies are ongoing. One aspect of long-acting injectables is that they might remain in the body for an extended period after the last injection, which could be problematic if side effects occurred or if a person had continued HIV risk and thus had subtherapeutic concentrations of PrEP medication that could select for antiretroviral resistance. One abstract (McGowan, TUAC0103) studied the injectable TMC 278LA (long-acting injectable rilpivirine) and found that the medication was detectable in plasma and cervicovaginal fluid >18 months after injection of a single dose of TMC 278LA. These results are a strong reminder of important safety considerations with development of new PrEP agents.
 
The dapivirine vaginal ring is the PrEP agent that is furthest along towards becoming an alternative to oral TDF-based pills, after the results of the two trial earlier this year. At AIDS 2016, a new analysis from one of those trials (MTN-020/ASPIRE) was presented (Brown, TUAC0105LB). ASPIRE was a phase III, placebo-controlled trial of the dapivirine vaginal ring, changed monthly, among 2629 women from Malawi, South Africa, Uganda, and Zimbabwe. Overall, the ring reduced HIV infections by about one-third and by about one-half among women >21 years of age, who appeared to use the ring with greater adherence. Adherence, of course, is essential for PrEP and prior clinical trials among women (such as VOICE and FEM-PrEP) found very low PrEP adherence and no HIV protection, which is why the dapivirine ring trials are such a step forward. In the abstract presented at AIDS 2016 the authors did a deeper analysis of data from ASPIRE, relating HIV protection to adherence, determined by measuring the quantity of dapivirine left in the rings after they were returned each month (because less dapivirine left in the ring should mean more use). Several approaches were taken to interrogate the data, including looking at different absolute cut-offs for how much dapivirine remained and also scaling the amount remaining against the amount of time women had had the ring in their possession (as some women returned early or later for monthly visits and the amount of drug remaining should be scaled appropriately). Across all analyses, greater adherence to the ring was associated with significant HIV protection, ranging from 56% (with an adherence measure likely suggesting any use whatsoever, not necessarily high use) to 75% in two analyses and 92% in another for times with higher use. These results are very encouraging that the dapivirine vaginal ring might provide substantial HIV protection when used well. Two open-label extension studies (HOPE and DREAM) of the dapivirine vaginal ring initiated just before the opening of AIDS 2016 and their results will be eagerly anticipated.
 
ART for prevention
 
Antiretroviral treatment is a potent HIV prevention strategy. Like PrEP, the ultimate impact of TasP for HIV prevention depends on its coverage in a population (i.e., how many people who have HIV infection actually are receiving ART) and adherence (i.e., how well those receiving ART take it). In Durban in 2000, the IAS meeting changed HIV treatment by catalyzing activism to make ART available to all countries. For Durban 2016, ART is now recommended for all persons with HIV.
 
At the individual level, the HPTN 052 trial demonstrated that treatment essentially eliminates HIV transmission, a result that led to change in WHO and other  guidelines to initiate ART at higher CD4 counts.
 
The impact of rolling out ART on community-level HIV incidence is not known. There are several studies ongoing testing the effect of ART roll-out on community incidence. The first of these was reported at AIDS 2016. The ANRS 12249 study was a cluster-randomized trial of universal testing and ART versus South Africa standard of care (ART at CD4 <350), done in rural KwaZulu-Natal, South Africa. The trial was powered for a 34% reduction in HIV incidence, measured directly in HIV uninfected persons in a community cohort receiving periodic HIV testing. HIV prevalence was 31% and ∼34% were on ART at the beginning of the study. HIV ascertainment was approximately 75-80% throughout the study. Engagement in HIV care for HIV+ individuals with newly-identified HIV infection was only 47% at 1 year, in both arms, although ART initiation was much higher in the intervention arm (91% of those who engaged in care) than the control arm (52%) - putting this all together, ART coverage was 45% in the intervention arm and 43% in the control arm. 495 new HIV infections were measured: incidence 2.13 in the intervention arm and 2.27 in the control arm - a difference that was not statistically significant (adjusted RR 0.95, p=0.58). Thus, even with nearly complete population HIV testing, as a result of suboptimal and delayed linkage to HIV care, ART coverage was incomplete and much less than UNAIDS 90-90-90 goals and as a result HIV incidence was not reduced. These results are not surprising - coverage is key to ART for prevention - but they remind us strongly that social and structural factors must be addressed to achieve the benefits from ART for prevention.
 
PMTCT
 
ART is central to prevention of mother-to-child transmission of HIV (PMTCT) but numerous challenges impede perfect implementation of PMTCT services. There were a number of excellent PMTCT-focused sessions at AIDS 2016, covering new science in reducing infant HIV infections.
 
In a session entitled, Healthy Mothers, Healthy Babies: The Path to eMTCT, the multipronged approach to elimination of MTCT in South Africa was presented (Ng'oma, TUAE0105). The approach is based on a decentralized M& E system with an innovative color-coded "robot dashboard" monitoring system to identify early warning to gaps / leaks in PMTCT cascade. They also have employed a simple tool for providers, community health workers and policy makers to track progress and prioritize actions to address the gaps and leaks. The M&E includes quarterly provincial reviews, annual stock take exercises with triangulation of data from different sources (DHIS, Lab, impact studies, surveys). They reported consistent increases in maternal ART coverage and early infant diagnosis in the past 6 years with an 84% reduction in new pediatric HIV diagnoses. They reported an impressive increase from 39% birth testing coverage in South Africa in July 2015 to 89% in 2016. They concluded that the successes toward elimination of MTCT in South Africa were due to political leadership at all levels; strong partnerships for a coordinated technical, financial and advocacy support (between the South African government, development partners of UN and PEPFAR, academia, NGOs, and civil society), and, decentralized, standardized simple but robust monitoring and evaluation systems for ownership and sustainability.
 
In the same session, another abstract assessed transmission risk in South Africa (Goga, TUAE0106). The relevance of this is that 8-24 months HIV-free survival is the gold standard measuring PMTCT impact but few low to middle-income high HIV prevalence countries produce these data nationally. In 2012-2014 South Africa conducted a national evaluation ot measure mother to child transmission of HIV and HIV-free survival, during a time when PMTCT policy transitioned from Option A to PMTCT Option B. The goal was a target sample size of 1620 infants for national estimates of 18-month HIV-free survival, assuming 5% MTCT and 5% death, recruited from primary health care clinics and community health centres. Of 2811 HIV exposed infants, they evaluated HIV transmission and mortality outcomes among 1797 (71%) who had interviews at 18 months post-delivery, among whom cumulative MTCT at 18 months was 4.3% (95% CI 3.7-5%). The most rapid increase in MTCT and infant mortality occurred during the first 6 months postpartum, followed by a gradual increase thereafter. Postnatal MTCT accounted for 39% of 18 month MTCT, whilst intrauterine and early postnatal MTCT accounted for 61% of 18 month MTCT. The encouraging result was that South Africa achieved their target of <5% MTCT and the data indicate the need to focus on continuing ART and infant follow-up, particularly in the first six months post-partum.
 
In an oral poster discussion (Barron, WEPDE0103), data were presented about MomConnect, which is a South African National Department of Health initiative that sends pregnant women messages appropriate to their stage of pregnancy to strengthen the demand for health services and empower women. It enables these women to interact with the health system, obtain further health information and to provide feedback on the quality of care that they receive to improve supply of services. Since its launch in August 2014 MomConnect has registered almost 600,000 pregnant women among the >95% of all facilities dealing with pregnant women which have recorded MomConnect registrations. Over 200,000 questions have been asked about ART, PMTCT, and safe delivery. Each question is answered on a daily basis and if serious women are directed to a health facility. MomConnect has empowered pregnant women and improved the demand for better quality as well as improving supply and the quality of health services. MomConnect data is being integrated with national data system.
 
In the same session, data from a prospective cohort of 471 pregnant women in Gugulethu, South Africa who had postpartum follow-up were assessed to determine whether self-reported antenatal ART adherence and viral suppression predicted postpartum viral rebound (Phillips, WEPDE0106LB). In the subset of 339 who were virally suppressed during antenatal care, 21% experienced a viral load >1000 copies/ml by 12 months postpartum. Women who reported ≥1 missed ART dose during a month of their pregnancy were 2 fold more likely to have a post-partum VL >1000 copies/mL. Identifying younger, single women and those who report missed doses during pregnancy may assist in targeting postpartum adherence and retention interventions in routine care low-resource settings.
 
Data from the PROMISE study were presented in two late-breaker abstracts (Currier, THAB0103LB and Stranix-Chibanda, THAB0106LB). Currier presented the randomized comparison of stopping or continuing ART among postpartum non-breastfeeding women with pre-ART CD4 > 400 cells/mm3 (PROMISE 1077HS), which was conducted in 52 clinical research sites in 8 countries. The primary composite endpoint was time to AIDS event (WHO Stage 4 Condition), serious cardiovascular, renal, hepatic event or death. Of the 1653 women who were enrolled, 825 were randomized to discontinue ART, and 827 were randomized to continue ART (90% of whom were on a PI-based regimen) with a median follow-up of 2.3 years. Women who were randomized to continue ART had a 32% lower risk of reaching the primary composite endpoint of time to an AIDS event, and a 44% lower risk of reaching an AIDS event or a WHO stage 2 or 3 event in a secondary analysis. Importantly, among those who were randomized to continue ART, 23% had virologic failure, reflecting adherence challenges in women who continued ART post-partum. The investigators concluded that interventions to improve adherence as well as newer regimens with a high genetic barrier to resistance are needed to insure maximal long-term benefit.
 
In the same session, Stranix-Chibanda presented about the low acceptance of early ART among postpartum women in PROMISE studies globally (studies 1077 BF/FF/HS, which were conducted in 70 sites in 15 countries). During follow-up, the results of the START trial about the benefits of early ART were provided to women in the PROMISE studies and all women in PROMISE had an initial counseling session in which they were recommended to take ART. Of 1483 not on ART at the time of START results, 66% accepted ART. The major reasons for accepting ART are concerned about their health and understanding that treatment is now recommended in about 75% of women. The main reasons for not accepting ART were wanting more time to consider ART and feels well/knows CD4 count cited by about half of women. Thus, despite intense ART education and HIV monitoring in a well-resourced clinical trial setting, a substantial minority of women required additional time and counseling to start early ART for their own health.
 
Testing and Linkage to Care
 
As HIV treatment guidelines are moving towards universal treatment regardless of CD4 and with implementation of effective biomedical prevention strategies such as medical male circumcision and PrEP, HIV testing becomes even more critical. With the ambitious new 90-90-90 goals set by UNAIDS, there is momentum to increase prevention efforts, and HIV self-testing provides a number of benefits that could help expand HIV testing. The confidentiality it offers maybe preferred by hard to reach groups, such as MSM. In an oral abstract session called "Innovations in HIV testing: The first 90" (WEAE01), data were presented about the benefits and adverse outcomes of HIV self-testing among high-risk MSM in China (Qin, WEAE0102). In China, there is widespread use of HIVST among MSM in a permissive regulatory and legal environment, without national guidelines. Much existing evidence about HIVST comes from research settings, where supervision and counseling resources minimize the likelihood of adverse outcomes and increase the likelihood of linkage to care, They conducted a nationwide survey of MSM in China recruited from popular social media and networking or dating websites, and analyzed benefits (e.g., first-time testing, increased testing frequency, post-testing counseling), adverse outcomes (e.g., coercion, violence, suicidality), and correlates of first-time HIV testing. Among 1189 MSM who completed the survey, 647 had ever taken an HIV test and 341 men (28.7%) had ever taken a self-test, of whom 200 reported their first HIV test was a self-test. About half of the men who reported self-testing, confirmed their results and received post-test counseling, and about 10% reported adverse outcomes, including coercion to test, suicidality reported by 14 men and 2 reported violence. Young men who were not 'out' to their doctors were more likely to report that they used HIVST for their first testing. They found a high proportion of those with positive HIV tests reported receiving confirmatory tests and counseling, which is encouraging for linkage to HIV care.
 
In the same session, another abstract (Mothibi, WEAE0104) presented about index client tracing in South Africa as a home-based HIV counselling and testing strategy to identify and link PLWHIV to treatment, including both HIV, STI and TB testing to individuals and couples within index households. They conducted HBCT in 3 high HIV prevalence settings based on antenatal HIV prevalence in 2013: eThekwini and Umgungundlovu in KwaZulu Natal and Buffalo City in the eastern Cape. They counseled 66,766 household members of 14,779 HIV positive persons for a 4.5:1 ratio, of whom 89% received HIV testing and 97% received TB symptom screening. A total of 9219 (15.5%) tested HIV positive and 21% had TB symptoms. Remarkably, 93.7% of the HIV positive persons were linked to HIV care and treatment. Although a lower proportion of household males than females were counseled (19% vs 81%), the uptake of HIV testing among those counseled was similar (81% M and 91%F), as was HIV positivity rate (13.9% M and 15.9% F). They concluded that index client tracing for HIV counseling and testing of household members in high HIV prevalence settings was highly acceptable, had a high yield of new HIV diagnoses, among whom a substantial proportion had concomitant TB, and a high proportion of whom were successfully linked to treatment facilities.
 
In that session, a cluster RCT of financial incentives to increase couples HIV testing and counseling (CHCT) in 68 communities in rural Zimbabwe was presented (Sibanda, WEAE0105). Almost 15,000 people were tested in the communities randomized to receive incentives for CHCT (choice of laundry soap, cooking oil or petroleum jelly) and almost 11,000 in the communities not randomized to incentives for CHCT. The effect of incentives was strong: 1062 (10.0%) tested with their partner in the control arm and 7852 (55.7%) tested with partner in incentives arm for an adjusted OR of 13.5 (95% CI 10.5-17.4). The authors concluded that incentives were effective in increasing CHCT, with the main motivation for testing being the desire to know their joint status, and that although social harms were infrequent, couples learning that one or both partners were HIV positive need support.
 
In the oral abstract session, "All in the Family" (WEAD01), a randomized controlled trial of the Uthando Lwethu (Our Love) intervention on rates of couples HIV testing in rural South Africa was presented (Darbes, WEAD0102). 332 couples were randomized from a rural community in KwaZulu-Natal South Africa to a couples-based behavioral intervention comprising six sessions (two group sessions/four couple counseling sessions) or the control condition of one group session. The intervention explored barriers to HIV testing and promoted improved communication skills and positive relationship dynamics. The primary outcomes were participation in CHTC and number of reported unprotected sex acts in the last 90 days with primary partner. Using intent-to-treat analysis, at final 9-month follow-up, a higher proportion of intervention couples had participated in CHTC than control couples (42% and 12% respectively, p≤;0.001), with a shorter time to CHTC than control group couples who participated in CHTC (p≤;0.0001). For sexual behavior, there was a significant reduction in proportion of unprotected sex acts for intervention couples at 3-month follow-up (IRR = 0.74, p≤;0.022). Addressing relationship factors among African heterosexual couples can significantly improve rates of CHTC. The proportion of unprotected sex acts was reduced at first follow-up but not sustained over time.
 
In an oral abstract session, "Testing Times-Interventions to Improve Rates of HIV Testing" (FRAC01), data about community-based HIV counseling and testing in Tugelo Ferry in rural KwaZulu-Natal, South Africa were presented (Shenoi, FRAC0103), in which HIV testing with phlebotomy for CD4 cell count, TB symptom screen with sputum collection, and linkage to care was conducted at community congregation sites (eg., pension pay points, taxi ranks, prisons, secondary schools). Through 880 community events, almost 14,000 persons were screened with 86% HIV tested, 43% of whom were men. HIV positivity was 10%, among whom 56% had phlebotomy for CD4 with a median CD4 of 429. They calculated the number needed to screen to identify an HIV+ person, which ranged from 5.5 at prisons to 59 at secondary schools. Overall, 42% of HIV+ women were identified at municipality events and 30% of HIV+ men were identified at taxi ranks, indicating that Community-based HIV testing with CD4 count services may remove obstacles to HIV care.
 
In the same session, another RCT was presented, which promoted male partner and couples testing through distribution of HIV self-tests by pregnant and postpartum women (Agot, FRAC0104). HIVST is included in the National Policy on HIV Testing Services in Kenya, and may help increase partner and couples testing. Women were eligible if they were ages 18-39 years and in ANC from 20 weeks gestation, with a partner of unknown HIV status or known negative, and did not believe violence would result from distributing self-test to their partner. Women were given two oral fluid based HIV self-tests to take home with them and shown how to correctly use the self-tests, and provide written and pictorial instructions with each self-test with encouragement to distribute self-tests to their male partner. The comparison group was women who were given referral vouchers that invited their partner to obtain HIV testing at VCT clinics, alone or as a couple. The primary outcome was whether their partner had sought HIV testing in past 3 months, and secondary outcomes included sexual behavior and decision-making about condom use and intimate partner violence. Of the 600 women who were enrolled, partner testing was 91% in the HIVST arm and 51% in the control arm. Couples testing was 42% more likely and disclosure was 39% more likely in the HIVST group than the comparison group.
 
Lastly, given that testing is most impactful if it achieves high rates of HIV care linkage,In the session, "Target 90-90-90: The Ups and the Downs" (WEAE02), the findings from LINK4HEALTH: A combination strategy for linkage to and retention in HIV care in Swaziland were presented (McNairy, WEAE0206LB). In a cluster randomized controlled trial, 10 study units were randomized to the combination intervention strategy (point of care CD4, accelerated ART for those with CD4<350, SMS appointment reminders, basic care and prevention package, and non-cash financial incentives) versus standard of care. The co-primary outcome was linkage (HIV clinic visit within 1 month after HIV testing) and retention (HIV clinic visit within 12 months). The units randomized to the combination intervention strategy had approximately 50% prompt linkage to care and 12-month retention in care.
 
Medical Male Circumcision
 
VMMC progress in 14 countries, in which >11 million VMMCs had been conducted through 2015, was presented (Njeuhmeli, THAE0303). 11 million represents 56% of the estimated 20.9 million VMMCs required to reach 80% coverage by end 2015. The median estimated cost per HIV infection averted was $3,800. Across all countries modeled, 50% of the projected HIV infections averted were attributable to circumcising the 10-19 year-olds highlighted remaining challenges of low site utilization/inefficient use of limited resources, reaching most at-risk men ages 15-29 years old, constantly keep safety and quality of services at high level, linkages between services (HIV+ men to care and treatment, HTC to VMMC), and planning for sustainability. The authors recommended that countries should monitor MC coverage by five-year age group to better plan their programs.
 
In an oral abstract session, "Circumcision: Where to, How to, Who to?" (WEPDC01), data on adolescent girls' support of male peers and sexual partners receiving voluntary medical male circumcision (VMMC) services and implications for demand creation were presented (Dam, WEPDC0106). They conducted 12 focus group discussions (FGDs) conducted with female adolescents (ages 16-19) in South Africa, Tanzania, and Zimbabwe, which focused on girls' opinions and perceptions of VMMC and 92 interviews conducted with male adolescent VMMC clients 6-8 weeks post-procedure. Girls discussed preferring circumcised male sexual partners over uncircumcised ones, citing the former's sexual appeal, hygiene, better sexual performance, and reduced chances of passing on infections (including HIV). Girls discussed being supportive of boys' decision to be circumcised and both overtly and covertly influencing their peers/partners to undergo VMMC. Girls discussed not necessarily offering tangible support during the healing process, but rather emotional support in making the decision to get circumcised. They concluded that adolescent girls are involved in VMMC decision-making, especially with older adolescent boys, and that demand creation initiatives should continue to engage females in promoting VMMC to their male counterparts. In another abstract (Grund, WEPDC0104) data on barriers to VMMC among uncircumcised men aged 25-39 years in Nyanza region, Kenya were presented. Reduction of HIV risk was the primary reason why men aged 25-39 years get circumcised. Concern about lost wages and pain/fear were the most common reasons why uncircumcised men were not circumcised prior to the study, followed by dislike of the VMMC facility or providers is also an important barrier to uptake for these men. Interestingly, concerns about 6 weeks of sexual abstinence was not a common primary barrier. Innovative demand creation strategies are needed to address structural and financial barriers among older uncircumcised men.
 
Epidemiology
 
A number of studies used phylogenetics and other methods to understand transmission dynamics in heavily-burdened populations. From Peru, an interesting study has enrolled >3000 men who have sex with men for immediate ART (if HIV+) or frequent HIV testing (if HIV-), finding high incidence (8.6% per year), geospacial evidence of transmission hotspots, and phylogenetic clusters associated with similar hotspot venues (Brezak, THAX0102). From the US, phylogenetic analyses showed that bisexual identity, depressive symptoms, and marijuana use were associated with larger transmission clusters among black men who have sex with men, suggesting areas for targeting prevention interventions (Morgan, THAX0103). From South Africa, similar analyses suggested that young women (aged <25 years) in rural KwaZulu-Natal were linked in transmission clusters to men not on ART with high viral loads who were older, although by just a few years (mean 28 years) (de Olivera, THAX0104).
 
Vaccines
 
An effective, safe, and deliverable HIV vaccine is an important dream for HIV prevention. The first suggestion that a vaccine could be developed that would protect against HIV was the RV-144 trial from Thailand. The HVTN100 study built off of RV-144, taking a ALVAC vector, modified to reflect subtype C (vCP2438), in combination with subtype C gp120/MF59 (Bekker, TUAX0102LB). This phase I/II randomized, double-blind, placebo-controlled trial assessed the immunogenicity of this vaccine combination among 185 persons in South Africa; the vaccine required 5 doses over the course of a year. The endpoints were immunologic: antibody and CD4 responses against HIV Env protein, with a total of 4 criteria. All four were met - defining this vaccine as sufficiently immunogenic to move forward into a phase III trial, which will be starting next year. RV-144 provided ∼30% protection against HIV overall, although a suggestion of greater protection early after vaccine provision (maybe as a result of higher immunity soon after a dose), and so phase III studies are not expected to provide perfect protection, if successful; nevertheless, a partially-effective, safe, and deliverable vaccine would be an incredibly important part of combination HIV prevention.