HIV Prevention at AIDS 2018
22nd International AIDS Conference
Amsterdam, Netherlands 23-27 July 2018
Jared Baeten, MD PhD
Connie Celum, MD MPH
University of Washington
The large International AIDS Conference - held every other year (the last in Durban, South Africa in 2016) - was held in Amsterdam this year. As at all International AIDS Conferences, the meeting crossed all boundaries, with sessions on laboratory science, clinical care, epidemiology, policy and program work, advocacy, and social justice and contributions across the spectrum from basic, behavioral, social, and clinical sciences, as well as policy, law, economics, justice, and development. 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
). Like many recent International AIDS Society-sponsored conferences, the meeting was shaped by an overarching vision statement - this year, focused on the intersection of people, politics, and power and their need to come together to fulfill an equitable global response (sign on at http://www.aids2018.org/Get-Involved/Take-part/The-Amsterdam-Affirmation).
Prevention was front and center at this year's meeting. There was robust science, fantastic symposia, enlightening plenaries, and many, many enriching side programs. The use of antiretrovirals - as treatment (ART) for persons living with HIV, resulting in tremendous prevention benefits, and as pre-exposure prophylaxis (PrEP) as an option for HIV uninfected persons - consumed many sessions, creating great excitement. Other prevention interventions too are being delivered, in synergy with antiretroviral-based strategies, with opportunity still for more impact.
Plenary talks related to prevention included ones on the epidemiology of HIV among vulnerable populations and among persons who inject drugs (
http://programme.aids2018.org/Programme/Session/16) and models of differentiated prevention delivery in Kenya
(http://programme.aids2018.org/Programme/Session/36), among other great plenary sessions.
Great symposium sessions related to prevention included ones on: state-of-the-art methodologies for prevention science
drug policy and harm reduction (
), key populations (
http://programme.aids2018.org/Programme/Session/8), PrEP policies (
http://programme.aids2018.org/Programme/Session/50), combination prevention strategies
and next generation HIV prevention
ART for prevention
The prevention benefits of antiretroviral therapy (ART) are inarguable. Effective ART - i.e., sustained use that results in viral suppression - turns off HIV transmission. How to translate the scientific successes of ART into public health impact is the big question - i.e., how to deliver ART, at scale, to enough people, with sustained use and adherence, to result in population-wide reductions in new infections. There was fantastic new science at AIDS 2018 related to ART for prevention, detailed below.
In addition, AIDS 2018 arguably marked the culmination of success for the U=U (undetectable = untransmittable, https://www.preventionaccess.org/) campaign - i.e., a movement to state clearly that when persons living with HIV is on effective treatment, reducing levels of virus to undetectable, they are incapable of transmitting the virus to sexual partners. U=U is game changing - for reducing stigma, expanding and saving lives, and building enthusiasm for more treatment and prevention success. There was much U=U discussion throughout the meeting - and clear global agreement that U=U, truly.
The greatest U=U excitement came from the results of the PARTNER2 study (abstract WEAX0104LB, Rodger). The PARTNER1 study was reported in 2016; it demonstrated no HIV transmissions among HIV serodifferent couples (many heterosexual) in Europe in which the positive partner was on suppressive ART. PARTNER2 (as well as another study called Opposites Attract, published in Lancet HIV just before AIDS 2018) followed male-male couples. It was an observational, multicity study from 2014-2018; follow-up was 6 monthly. Eligible follow-up time was when the HIV infected partner was on ART and had a viral load <200 copies/mL, condomless sex was reported, as well as when the negative partner was not taking PrEP or PEP. HIV seroconversions in the initially-negative partners were assessed by phylogenetic analysis. 972 couples were recruited, of which 783 contributed 1596 eligible years of follow-up. Median age was 38 years for negative partners and 89% were white. Follow-up median was 1.6 years, 23% were diagnosed with an STI, 37% had condomless sex with other partners, median number of condom less sex acts was 43 in a year. Nearly all HIV positive partners remained on ART with viral suppression. 15 initially HIV negative partners became HIV positive - 11 self-reported recent condomless sex with others. All of the infections were not phylogenetically related to samples from the HIV infected partners. Thus, the risk of transmission was zero, with upper limit of confidence intervals that were very low: any sex 0.23 per 100 years of follow-up, anal sex 0.24, insertive anal sex 0.27, receptive anal sex 0.57. These results, plus Opposites Attract, affirm U=U for MSM couples. The authors estimated that given the upper limit of the confidence, a couple would need to have unprotected sex for >400 years to be at risk of transmission - a striking number.
Five large studies are testing scaling up ART in populations to result in prevention benefits - by, acronym, they are called PopART, MaxART, SEARCH, TasP, BCCP. A bridging session (http://programme.aids2018.org/Programme/Session/101) laid out the basics of each of these studies - each done in a different place, with slightly different methods and goals. One (TasP, done in rural South Africa) presented its primary results two years ago - basically finding no prevention benefit, in the context of very low ART uptake. Three studies presented their primary results at AIDS 2018 (http://programme.aids2018.org/Programme/Session/164).
MaxART (abstract WEAX0102LB, Okello) is a randomized, stepped-wedge trial among 14 health facilities in the Kingdom of eSwatini, which offered ART regardless of CD4 count (termed Early Access to ART for All, or EAAA, by the study team), along with viral load monitoring, clinical mentoring, and community mobilization - essentially, a rolling program that improved and optimized ART delivery across the 14 clinics, done in a staggered (=stepped-wedge) fashion to both allow time for each program to be trained/improved and to observe the before-after impact of the program. The primary endpoints were retention in care and viral suppression. The work began in 2014, and final results were through 2017. Unlike other test and treat trials, MaxART did not measure HIV incidence - thus, there is not explicitly a TasP outcome of MaxART. In total, 3405 individuals were enrolled - 60% during the standard of care phase and 40% during the EAAA phase. Overall, there was a 60% increase in retention under the EAAA phase (95% CI 1.14-2.21, p=0.005), a difference that was emerging even at 12 months after enrollment (86% vs. 80%). The viral load results were still pending and were not presented in detail at AIDS 2018. Per-patient delivery costs were no greater under the EAAA phase. In summary, MaxART demonstrates that a universal ART strategy can be delivered successfully in a high-burden epidemic setting, resulting in better engagement in HIV care - a win-win for sure.
The Botswana Combination Prevention Project (BCPP, also known as YaTsie in Tswana) is a pair-matched community randomized trial in 30 communities in Botswana (abstract WEAX0105LB, Makema). The project evaluated a package of treatment and prevention interventions at the population level, although primarily better HIV testing and ART services, and measured population-level HIV incidence, done by baseline and annual surveys conducted among adults 16-64 years, in a random nested sample of 20% of households. The work started in 2013, and data were through June 2018. In the intervention communities, the intervention included: community mobilization, home-based and mobile and targeted HIV testing, linkage to care services including SMS reminders and active tracing, strengthened VMMC services, and universal ART starting in 2016 (from 2013 to 2016, the project provided ART to those with low CD4 count or with high viral loads if the CD4 count was high. In the standard of care communities, the intervention was ART for CD4 <350 until June 2016 when transitioned to ART for any CD4. The total population of BCCP was ∼180,000 (∼10% of entire Botswana population), and the nested cohorts had ∼4400 enrolled in each, with 95% of those completing follow-up testing at least once. Overall HIV prevalence was 28%, consistent with the Botswana epidemic. In total, 57 individuals in intervention arm acquired HIV (0.59%) versus 90 in the standard of care arm (0.92%), translating to a 31% reduction in incident HIV in the primary analysis (IRR 0.69, p=0.09) and a 30% reduction in models adjusting for community cofactors (IRR, 0.70, 95% CI 0.50-0.99, p=0.04). Process measures were consistent with these primary results - testing was high and the % of all HIV+ persons with viral suppression was high (88% in intervention arm). Of note, VMMC uptake was low. In conclusion, BCCP observed a 30% reduction in community HIV incidence with a combination of community-based HIV testing, linkage to care, and ART interventions, even in the background of high HIV treatment coverage. This is a reminder that ART service improvement is important, even when ART services are already very good, as they have been in Botswana for some time.
The SEARCH trial (abstract WEAX0106LB, Havlir) is the third community-level ART trial presented at AIDS 2018. SEARCH tested HIV test and treat with universal ART using a multi-disease, community health care model in 32 pair-matched communities in rural Kenya and Uganda. Communities were randomized to either the full intervention, which included baseline and annual health fairs, universal ART eligibility, implementation of a chronic care model of care of HIV (plus hypertension and diabetes), rapid ART initiation and viral load monitoring and counseling, mobile phone triage and reminders, and flexible clinics; control communities had HIV care according to national standards. The primary study endpoint was cumulative HIV incidence at Year 3 among persons HIV negative at baseline, measured in the full community; the trial was powered to detect a 25-40% reduction. The study began in 2013, and data were presented until 2017. The study followed >150,000 persons; HIV prevalence was 19% in Kenya and 7%/4% in the two Uganda regions. At baseline just over half had ever tested for HIV - and this increased to 90% at baseline in both types of communities (and then further increased to 94% in the intervention arm communities after subsequent successive waves of testing). In the intervention arm, ART start was more rapid than in the control arm for those newly diagnosed (80% versus 45% at 1 year, p<0.001). Viral suppression among all HIV infected persons was 79% by year 3 for the intervention arm (compared to 68% for the control arm), and compared to 42% at baseline. Suppression was slightly lower in men (74%) and youth (55%); in-migrants had lower suppression. There was no difference in HIV incidence - 0.8% overall; results were consistent when stratified across the two countries. Of note, mortality was lower in the intervention arm - 21% lower (RR 0.79, 0.65-0.96, p=0.02), and TB incidence was also lower, by 59%, p=0.02 as was hypertension and diabetes control. The investigators speculated that the lack of difference in HIV incidence might be because the control arm was very strong in this study (90% aware of HIV status), because new ART guidelines went into place when the study was ongoing, and potentially because of new infections coming in from outside the community. HIV incidence did decline overall in the study - 32% overall, 45% in Kenya, 49% among men, 19% among women, although these are all still above elimination levels.
Taken together the results of the population-scale ART trials are showing somewhat encouraging but somewhat less positive than anticipated results. They emphasize that population coverage is important (particularly shown in the contrast between the TasP trial and BCCP/SEARCH), that treatment provides tremendous individual benefits (evidenced in SEARCH), but that treatment alone, even at the fairly high levels, does not provide high community-level prevention benefits. There is potential for pushing testing and treatment coverage higher (although SEARCH and BCCP got to very high numbers), and additional prevention benefits may come from synergizing with primary prevention interventions (e.g., VMMC, PrEP).
Pre-exposure prophylaxis (or PrEP), in which an HIV uninfected person uses an antiretroviral as chemoprophylaxis against HIV received considerable attention at this year's AIDS 2018 meeting. Some sessions were unbelievably standing room only. PrEP remains primarily oral tenofovir-based pills (mostly tenfovir disoproxil fumarate in combination with emtricitabine [FTC/TDF]), the only approach to date to have received regulatory approval in any country, although new PrEP agents are marching through clinical trials.
A Tuesday oral abstract session addressed diversity of delivery of PrEP in many settings (http://programme.aids2018.org/Programme/Session/103), one on Wednesday focused on economics and implementation science (http://programme.aids2018.org/Programme/Session/135), and one on Thursday addressed sexual health in the era of PrEP (http://programme.aids2018.org/Programme/Session/1457). An oral poster discussion session on Tuesday was overflowing the room (http://programme.aids2018.org/Programme/Session/149).
From France, results of an open-label PrEP study were reported (abstract WEAE0406LB, Molina). 3000 adults who have inconsistent condom use are to be enrolled and offered daily or on demand PrEP. The work began in 2017, and follow-up will continue until 2020 - the larger goal is to expand PrEP use at the community level enough to result in a 15% reduction in new HIV infections in the city of Paris. Follow-up is quarterly, including HIV testing and STI testing. As of July 2018, >1600 have been enrolled - 99% MSM, 85% white, median age 36 years, 16% reporting chemsex, average of 10 partners in past 3 months. 55% chose on demand PrEP, 45% daily PrEP - those choosing daily had a higher average number of partners. The daily/on-demand ratio has remained consistent but about 15% switch between approaches each quarter. 1100 have reached at least 3 months of follow-up - 89% report PREP use with last sex (81% on demand, 98% daily), 20% report condoms. No HIV infections have been observed so far and the upper bound of the 95% confidence interval on HIV incidence is thus <1%; if HIV incidence would have been that of the previous IPERGAY trial, 85 infections should have occurred already. Behavioral surveys show a slight increase on condomless sex but maybe slight reduction in number of partners; safety is very high and no discontinuations due to adverse events have been observed. These results confirm high efficacy of both daily and on-demand PrEP in MSM.
From Australia, the incidence of sexually transmitted infections (STIs) was compared before versus after PrEP roll-out in Melbourne (abstract THAC0502, Traeger). Men enrolling in a study (PrEPX), a population-level, multi-site, PrEP implementation project, were followed for incident STIs. Australia's PrEP use has expanded massively in the last few years - more than 15,000 individuals are estimated to be taking PrEP, and PrEP is subsidized by the national medication program. Follow-up was between 2016 and 2018; the population was nearly 3000 individuals (98% gay or bisexual men). During follow-up, 48% were diagnosed with at least one STI, 26% with two, and 13% with three (and this 13% accounted for 53% of diagnoses). STIs were associated with younger age, larger number of partners, group sex, and less condom use. STI rates were 42% higher after PrEP initiation (p<0.001); this was in part (but not in full) explained by increased STI testing frequency on PrEP. These results push for more STI prevention strategies, particularly for those getting HIV prevention benefits from PrEP.
From South Africa, PrEP programming for sex workers and men who have sex with men were initiated (in 2016 and 2017, respectively); an abstract explored the national program (abstract WEAE0401, Pillay) by surveying providers and clients. Both sex workers and MSM identified their sexual behavior and perception of HIV risk as reasons to initiate oral PrEP. Side effects were raised as a challenge for users but were underestimated by providers. There was also a notable number of at-risk persons who were not offered PrEP - highlighting potential missed opportunities. Counseling can somewhat mitigate readiness for side effects although not necessarily to deal with side effects when they happen.
From Kenya, an early PrEP program among female sex workers, men who have sex with men, and young women reported (abstract WEAE0403, Kyongo). The work was done between 2015 and 2016 and enrolled 1585 individuals, approximately equally split between the three risk groups. Attrition from PrEP was high, for all groups but especially young women - retention at 3 and 6 months was on the order of 15-25% overall, but ∼10% only for young women at 6 months. Lack of community awareness of PrEP was called out as a potential factor.
Another set of abstracts from Kenya illustrated high PrEP uptake integrated into new care settings (abstract WEAE0402, Kinuthia and abstract TUAC0304, Pintye). In a large program offering PrEP in antenatal clinics, there appeared to be high PrEP interest. Among 9171 women (52% antenatal, 48% postpartum) offered PrEP, 1993 (22%; 45% pregnant, 55% postpartum) took it up. The average age was just 24 years and 34% had a partner of unknown status. Factor associated with PrEP uptake were age <24, being unmarried, recent intimate partner violence, recent STI, and partner of HIV infected or unknown status. Reasons declared for not wanting PrEP focused on feeling not at risk or needing to consult with the partner; few mentioned fears of use of PrEP and effects on the unborn baby. In a related program operating in family planning clinics, 1122 women were offered and 232 (22%) initiated. In that setting factors associated with PrEP initiation included age older than 24 years, use of oral contraceptive pills, and, like in the antenatal/postnatal setting, recent partner violence, recent STI, and partner status positive or unknown.
From Senegal, a program offering PrEP in Ministry of Health Clinics showed good retention so far (abstract TUAC0301, Sarr). A total of 267 initiated PrEP and approximately 70% were retained at 6 months; only age (<25 years) was associated with not being retained. Medication use appeared reasonably good; women reported not using PrEP when they did not feel at risk. This work is spurring discussions with the Senegal Ministry of Health to scale this program.
From Thailand, an overview of PrEP services across the country identified key population-led services as a particularly valuable niche (abstract TUAC0302, Vannakit). Retention in PrEP services was generally high for men who have sex with men.
Two abstracts (abstract TUPDX0106, Cottrell & abstract TUPDX0107LB, Hiransuthikul) reported on the potential drug-drug interaction of feminizing hormone treatments (used by some transgender women) with tenofovir/emtricitabine as PrEP. The first measured tenofovir and emtricitabine (and their active metabolites) levels in blood and in rectal tissues among a small number of transgender and cisgender women taking tenofovir/emtricitabine as part of HIV combination treatment. The second enrolled transgender women initiating hormone therapy, and the starting PrEP; blood levels of PrEP and of exogenous estrogen were measured. Notably, PrEP did not have any effects on estrogen levels. However, in the presence of exogenous estrogens, both studies found lower PrEP levels - the second study quantified this as 13% lower. It is completely unknown whether these lower levels are actually a risk for lower HIV protection, or whether they simply indicate just lower numbers. More work is clearly needed for this important population.
Strategies to reach men for HIV testing
The release of more Population Health HIV and AIDS surveys (PHIAs) and results from the Universal Test and Treat studies, which included results of the SEARCH and BCCP trials indicate that epidemic control will not occur unless much higher rates of HIV testing, linkage to care, retention on ART and viral suppression are achieved (likely in conjunction with primary prevention). It will be particularly important to achieve higher HIV testing and linkage rates among men and youth, which were highlighted in a number of presentations.
In an effort to de‐stigmatize HIV testing and reach more men, a strategy of providing HIV testing services as part of screening for multiple diseases in a district hospital in western Kenya was evaluated (abstract WEAE0101, Ndede). In March 2017, a multiple disease screening program was established, offering a package of wellness services including screening for obesity, hypertension, and symptomatic TB and HIV testing. The service was advertised by radio and billboard as a free health check. From July to December 2017, 5645 adults were screened, 54% of whom were male, and 2.2% of women and 1.8% of men tested HIV‐positive. Approximately 15% of men tested during extended evening and weekend hours.
Another strategy to increase men's uptake of HIV testing is through providing convenient and confidential outreach HIV testing services through mobile clinics, which was evaluated in Malawi (abstract WEAE0102, Geoffroy). The mobile clinic was located at markets, tea estates, and religious facilities and included a gender-segregated space and male counselors. Through December 2017, there have been 106 male-targeted testing events, testing 6166 people, of whom 71% were male. This is significantly higher than the 26% males tested at weekday mobile clinics over the same period. Across all ages, 30% of males had not been previously tested and 42% of men aged 15 to 24; and 3% of men tested HIV+ (5% of men > 25 years old). Convenience of location (38%), publicity surrounding the event (35%) and convenience of time (23%) were cited as primary reasons for attending.
Home‐based testing was found to identify more previously undiagnosed older men than mobile testing in the BCCP (abstract WEAE0104, Roland). In Botswana, an estimated 85% of persons in know their status. In this cluster randomized trial of universal test and treat, almost 50,000 participants who did not know or have documentation of their negative HIV status in the prior three months, were tested for HIV with home (39%) or mobile (61%) testing, and 3.8% were newly diagnosed with HIV. Older males yielded the highest proportion of new diagnoses among all categories (p<0.001), and testing yield for new HIV diagnoses was significantly higher in the home than with mobile testing (4.5% versus 3.3%). These data indicate that home-based HIV testing strategies are important for identifying HIV‐infected individuals in a country with high coverage of knowledge of HIV status.
A presentation from the PEPFAR-supported DREAMS districts reported on increased funding to ensure saturation of HIV services for young adult men in DREAMS districts, including HIV testing, ART, and voluntary medical male circumcision (VMMC) (abstract TUAC0203, Cooney). To analyze the impact of this initiative, they compared DREAMS and similar non‐DREAMS districts in five countries to examine changes in HTS, VMMC, and treatment results after two years of DREAMS implementation (2015 to 2017). PEPFAR indicators were used to analyze the total number of VMMCs completed, clients tested, and new clients on treatment for males 15 to 49 years in DREAMS and non‐DREAMS districts. Results showed significant positive differences between DREAMS compared to non‐DREAMS districts in three countries among men ages 15-49 in terms of an increase in total number of men tested, newly on ART, and two countries for an increase in VMMCs. The authors concluded that focusing on treatment as prevention for male partners during DREAMS implementation may be associated with an increase in uptake of these services among men.
Strategies to increase uptake of HIV care and prevention services among men are also needed. A study in Lesotho evaluated comprehensive health services for men in health facilities as part of the DREAMS initiative to increase uptake of HIV services among men (abstract WEAE0105, Lekhotsa). The Men's Clinics intervention consisted of dedicated clinic space for male clients only to receive services by male nurses and counsellors to help men feel more comfortable accessing services, and included STI screening and treatment, HIV testing services, care and treatment, PrEP and PEP, TB Services, treatment for other co-infections, Condom distribution, and counselling for HIV prevention, Education on PMTCT and index family testing, and VMMC, were offered in morning, evenings, and weekends with appointments to reduce waiting times and improve client satisfaction. After the Men's Clinics were introduced, there was a 49% increase in men tested for HIV, 29% increase in men diagnosed as HIV-positive, and 63% increase in men initiated on ART.
A number of presentations about the 90:90:90 cascade indicate that the biggest challenge is achieving 90% of persons with HIV knowing their status. A cluster RCT of facility based HIV self-testing in outpatient waiting rooms was conducted among 15 Malawian health facilities in 2017-2018 (abstract TUAE0105, Dovel). Health facilities were randomized to 1) routine provider initiated testing and counseling (PITC); 2) 'Optimized PITC' (additional provider trainings and job‐aids); and 3) HIVST (HIV self-testng) in outpatient waiting‐spaces, private spaces for interpretation, and optional post‐test counseling. The primary outcome was HIV‐testing among outpatients, which was significantly higher (52%) in the HIVST arm than in Optimized PITC (14%) and PITC (12%). The proportion of outpatients who had not tested in the past 12 months was also significantly higher (60%) in the HIVST arm than 'Optimized PITC' (18%) and PITC (16%). Although HIV positivity rates did not differ by arm, HIVST had a 3-4 fold higher absolute number of new positives identified compared to 'Optimized PITC' and PITC. Acceptability of HIVST was also high; participants who were tested by HIVST were more likely to want to test again using the same method and more likely to recommend testing to others compared to those tested by 'Optimized PITC' or PITC.
Another strategy to increase knowledge of HIV status and demand for ART is through community‐based HIV self‐testing, which was evaluated in a cluster randomised trial in Malawi rural communities (abstract THPDC0103, Inravurdh). The intervention was door‐to‐door distribution of HIVST kits by resident community‐based distributors (CBD), who provided continuous HIVST access and option of post‐test support and assisted referral to routine confirmatory testing and ART services, which was compared to the standard of care. A total of 83 CBDs delivered 79,349 HIVST kits over 12 to 15 months, with only 3 reported social harms. Of 5504 adults in the post‐intervention survey, 43% were men and 15% were adolescents. Coverage was significantly higher in the HIVST than SOC clusters for both recent testing (64% vs. 46%, adjusted risk ratio (aRR) 1.4; and differences between arms were more pronounced for adolescents (aRR 2.0) and men (aRR 1.6). There was also a 1.4 increase in ART initiations (per 1000 adult clinic population) in the HIVST versus SOC arm during the intervention period).
To evaluate linkage to HIV care following HIV self‐testing, the HPTN071/POPART study, a cluster randomized trial of universal test and treat, conducted an evaluation of community‐based distribution of oral HIV self‐test kits in 4 communities in Zambia (abstract THPDC0102, Floyd). Among the almost 400 persons diagnosed with HIV, 94% (184/195) in HIVST and 98% (199/204) in non‐HIVST zones were referred to care.
Another way to reach men is through secondary distribution of HIV self‐tests as a way to promote HIV testing among male partners, which was evaluated in a subgroup analysis of young women in a randomized trial is western Kenya (abstract THPDC0104, Agot). Among 367 young women aged 18 to 24 years, male partner testing uptake was higher in the HIVST group: 92% in the HIVST group and 56% in the comparison group. Couples testing was also significantly almost more likely in the HIVST group than the comparison group (78% versus 38%).
Another strategy for increasing knowledge of HIV status is through community HIV index testing through tracing sexual contacts of HIV-positive index cases identified at health facilities, which was implemented and evaluated in Zimbabwe (abstract TUAE0103, Muchedzi). Nurses provide HIV testing and other health services to children and primary sexual contacts of the index-case at the household. Among almost 15,000 HIV+ index-cases identified from 330 health facilities, almost 13,000 households were followed for community index-testing and over 22,000 people tested for HIV, among whom 8,367 (38%) were HIV+, and 5,771 (69%) were linked to care and treatment. This community index testing had a 10-fold higher HIV yield than the health facility yield rate of 4%. This strategy reached almost 12,000 men with HIV testing, of whom 72% were in the priority ages of 25-49 years.
To evaluate the feasibility of using geo-spatial mapping to target HIV preventions, the Rakai and PANGEA-HIV consortium evaluated whether high prevalence fishing communities are a major source of new HIV infections to the inland populations in Rakai District, Uganda (abstract THAC0102, Ratmann), based on interviews and deep sequencing of HIV samples from 2011-14. Overall, 4309 (69%) were antiretroviral naive, of whom 2803 (65%) were sequenced and 359 phylogenetically likely transmission events involving 676 individuals were reconstructed. The direction of transmission could be inferred in 241 likely transmission events. Only 3/241 transmission events occurred from fishing sites to agrarian/trading communities. Adjusting for differences in participation and sequence sampling by age and community, an estimated 34% of transmissions occurred within fishing sites, 58% within agrarian/trading communities. Only 3% of transmissions from fishing sites were estimated to agrarian/trading communities, and 4% from agrarian/trading communities to fishing sites, indicating that HIV is infrequently transmitted from high‐prevalence fishing sites to the inland population.
The UN Children's Fund presented sobering data about the challenges ahead in terms of the impact of future demographic transitions on HIV epidemic control among adolescents and youth in sub‐Saharan Africa (abstract THAC0105, Khalifa). The context is that new HIV infections among adolescents and youth (AY) aged 15 to 24 in Africa are only reducing by an average 3%/year since 2010 and the number of AY is estimated to increase by 85%. They estimate that by 2050, new HIV infections will decrease by over 70% in Eastern and Southern Africa, 2% in West and Central Africa, and 53% overall in the Africa region. However, none of 46 sub-Saharan African countries will achieve epidemic control among AY by 2030. Between 2017 and 2050, 9.6 million AY will become newly infected with HIV in Africa, 67% of which will occur among girls and young women. These estimates underscore the need to increase efforts to deliver effective HIV prevention strategies at scale to adolescents and youth.
Given the need to improve delivery of HIV prevention and reproductive health services to adolescent girls and young women (AGYW), models for integrated youth-friendly health services are needed. The Girl Power study evaluated whether integrated youth‐friendly health services (YFHS) for AGYW led to increased uptake of condoms, HIV testing, and hormonal contraception, compared to standard of care (abstract HPDE0101, Rosenberg). Four comparable public sector health centers were selected in Lilongwe, Malawi and randomly assigned to either the SOC (vertical HIV testing, STI management, and family planning in three separate areas with providers who received no additional training) or YFHS (same services delivered in an integrated fashion in youth‐dedicated spaces, with peers and providers trained in youth‐friendly approaches). Among 1000 AGYW enrolled in the 4 health centers, the median age was 19 years and AGYW in the YFHS models accessed HIV testing 2.4 (CI: 1.9 to 2.9) times more often, condoms 7.9 (CI: 6.0 to 10.5) times more often, and hormonal contraception 6.0 (CI: 4.2 to 8.7) times more often. Each of the three YFHS health centers performed better than the SOC on each indicator. This project demonstrates that in public sector health centers an integrated model of YFHS that included brief provider training and modest clinical modifications lead to considerably higher SRH and HIV service utilization for AGYW.
Multiple structural factors, including gender-based based violence and poverty, impact African AGYW's risk of HIV, and to address this, a pragmatic cluster randomized trial of a parenting support and violence reduction intervention (Cash+Care) was conducted in the Eastern Cape (abstract THPDE0103, Cluver). A pragmatic cluster randomized trial (1100 participants, 40 clusters) in South Africa's rural and urban Eastern Cape. Adolescents and caregivers participated in 14 evidence‐based sessions about conflict reduction, protection from sexual abuse in the community, and family budgeting, which were implemented by local community members. Retention was 97% at 5-9 months post‐intervention. The intervention had a significant impact on key HIV risk reduction factors: lower family violence, improved involved parenting, and less poor parental supervision, improved family economic welfare, including sustained food availability, lower alcohol and drug use amongst adolescents and amongst caregivers, and with improved planning for protection against sexual coercion. This cluster RCT in South Africa shows that an intervention to support families and strengthen parenting can reduce direct risks for adolescent HIV‐acquisition in terms of lowering violence, low supervision, food insecurity and substance use.
Several innovative programs in Asia evaluated whether community-based and peer-delivered HIV testing could reach key populations (KPs). Vietnam was the first Asian country to adopt global 90‐90‐90 targets, despite ˜30% uptake of HIV testing among KPs. Studies indicated KP preference for community testing services, leading to a pilot of KP‐led community‐based organizations in urban areas and village health workers in rural areas using blood‐ and oral fluid‐based rapid tests in Vietnam (abstract THAC0202, Ngo). Of 59,333 clients who were HIV‐tested, 4% were newly diagnosed HIV‐positive and 90% enrolled in treatment. The key lessons were that CBOs/VHWs were capable and highly effective in offering HIV testing, reaching first-time testers, facilitating ART enrollment among those HIV‐diagnosed, and fostering greater engagement between CBOs and the government.
To determine whether geosocial networking mobile applications (apps) used by men who have sex with men could be utilized to target HIV and STI interventions in China the T2T study assessed app use to provide reference for app‐based HIV interventions (abstract WEAC0102, Zou). Men aged ≥18 years old with ≥2 male sex partners in 3 metropolitan cities were studied: Guangzhou, Shenzhen and Wuxi. Of 603 subjects with a mean age of 28 years, 80% had ever used an app, 77% of whom had been using an app for more than 12 months, and 42% spent >30 minutes per day on apps. The proportion of app use increased significantly with increasing age, longer time of stay in the city, higher educational level, more frequent alcohol use, and receptive anal intercourse. Prevalence of HIV and bacterial STIs were similar in the two groups except app users had significantly more rectal chlamydia (15.6% vs. 6.6%). GSN app should be used as a platform to carry out interventions aimed at reducing HIV and rectal sexually transmitted infections.
To increase engagement of transgender women, integrated gender-affirmative hormone treatment (GAHT) services were piloted in Thailand to increase uptake of HIV testing, syphilis testing and PrEP (THAC0204, abstract Janamnuaysook). Of 972 transgender women who attended the clinic, 91% received HIV testing, HIV prevalence was 12%, and 80% initiated ART. Compared to clients not accessing GAHT‐services, GAHT‐service clients were significantly more likely to re‐visit the clinic (50% vs. 34%), had higher rates of repeat HIV testing (32% vs. 25%), repeat syphilis testing (14% vs. 9%), PrEP uptake (10% vs. 6%), and use of other sexual health services, including hepatitis B testing/vaccination and STI treatment (50% vs. 34%).
To understand how HIV prevention programs can better support female sex workers (FSWs), a prevention cascade analysis was conducted among 611 FSWs who participated in the intervention arm in a cluster‐randomized trial in Zimbabwe (abstract THAC0503, Fearon). Approximately half of these women (54.7%) reported adherence to condoms and/or to PrEP. While women knew that condoms prevented HIV and reported good access (both 94%), only 46% reported no episodes of condomless sex in the previous month. Alcohol use of women and of their clients was associated with lower condom adherence, and newer entrants to sex work, and younger women were less likely to report taking PrEP every day. Integrated interventions to address alcohol use could increase condom and PrEP use among FSWs.
New HIV prevention strategies in development
Broadly neutralizing antibodies (bNabs) could provide an alternative approach to traditional vaccine approaches. bNAbs are capable of neutralizing most circulating strains, targeting different non‐overlapping epitopes on the HIV‐1 envelope spike. Two highly potent and complimentary bNabs are 3BNC117 and 10‐1074 , which have longer half-lives through introducing two amino acid mutations (M428L and N434S (referred to as "LS") into the Fc domains. Their efficacy in blocking infections was assessed in a repeat low dose mucosal challenge model with rhesus macaques with SHIVAD8‐EO; following a single intravenous infusion of each mAb (20 mg/kg body weight) to six animals (abstract THAA0105, Gautam). The most striking result was the long period of protective efficacy conferred by a single injection of Fc modified human anti‐HIV‐1 neutralizing antibodies in macaques. A single intravenous infusion of 10‐1074‐LS dmAb markedly delayed virus acquisition for 18 to 37 weeks (median = 27 weeks) whereas the protective effect of the 3BNC117‐LS bNAb was more modest (protection for 11 to 23 weeks; median = 17 weeks). To model immunoprophylaxis against genetically diverse and/or neutralization resistant HIV‐1 strains, a combination of the 3BNC117‐LS plus 10‐1074‐LS mAbs was injected into macaques by the subcutaneous route. With almost 3‐fold less of each bNAb in the mixture administered, compared to the amount of single mAb injected in the intravenous infusions, the mAb combination still protected macaques for a median of 20 weeks. This could translate into an effective semi‐annual or annual immunoprophylaxis regimen for preventing HIV‐1 infections in humans.