HIV Prevention at AIDS 2014 - PrEP, TasP, PMTCT, Key Populations
20th International AIDS Conference
Melbourne, Australia 20 July - 25 July 2014
Jared Baeten, MD PhD
Kenneth Ngure, PhD, MPH
Connie Celum, MD MPH
University of Washington & Jomo Kenyatta University of Agriculture and Technology
HIV prevention now plays a central role in every international HIV conference, and this year's 20th International AIDS Conference continued that trend, with this year's theme "stepping up the pace." As at all International AIDS Conferences, the approach was tremendously multidisciplinary, from laboratory to policy and advocacy, and with contributions from research in the behavioral, social, and clinical sciences. It has become tremendously clear that all disciplines play a role in making prevention discovery and delivery a success, and the interdisciplinary nature of prevention research and implementation was abundantly clear in the presentations this year.
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 plenary and oral abstract sessions as well as a number of special session and symposia (searchable program available at http://pag.aids2014.org/). We have included key links below.
The loss of colleagues traveling to this year's AIDS conference on flight MH17 cast a shadow on the conference. Tributes to their memory occurred throughout the meeting, including touchingly at the opening session (http://pag.aids2014.org/session.aspx?s=1952).
The opening plenary of the conference was delivered by Salim Abdool Karim and presented a comprehensive overview of the epidemic from its beginnings to the present, including the opportunities of new science, delivery of antiretroviral treatment, and implementation of robust and integrated prevention, as well as the challenges, particularly of stigma (http://pag.aids2014.org/session.aspx?s=2013).
Tuesday's plenaries were forward-looking, focused on health system strengthening (Olive Shisana), gender (Jennifer Gatsi Mallet), and investments in HIV care and prevention (Mark Dybul) (http://pag.aids2014.org/session.aspx?s=2012).
Antiretroviral-based HIV prevention strategies - including antiretroviral treatment (ART, often called treatment as prevention [TasP]) to reduce the infectiousness of HIV infected persons and oral and topical pre-exposure prophylaxis (PrEP) for uninfected persons to prevent HIV acquisition - are powerful approaches for decreasing HIV spread. The last three years have been a whirlwind with studies demonstrating definitively that these strategies work for preventing HIV - and, when taken with high adherence, can work very well. Just prior to AIDS 2014, WHO released its 2014 Consolidated Guidelines on HIV Prevention, Diagnosis, Treatment and Care for Key Populations (http://www.who.int/hiv/pub/guidelines/keypopulations/en/). The guidelines include a new, strong recommendation encouraging PrEP for at-risk men who have sex with men (MSM) worldwide, a groundbreaking statement. The release of these guidelines received considerable press attention and commentary leading up to and at the meeting, and they signal a tremendous opportunity to have an impact on the global HIV epidemic in MSM. In the Thursday plenary session, Kenneth Mayer discussed stepping up the pace with development and delivery of effective prevention tools, including PrEP (http://pag.aids2014.org/session.aspx?s=2010#3). A Tuesday oral abstract session was devoted entirely to PrEP (http://pag.aids2014.org/session.aspx?s=1106). [from Jules-CDC Guidelines: http://www.natap.org/2014/HIV/051514_01.htm
In late 2010, the iPrEx study demonstrated that daily oral PrEP, using the combination of emtricitabine plus tenofovir (Truvada®) was efficacious for the prevention of HIV acquisition in MSM and transgender women (TGW). iPrEx was conducted among 2499 MSM/TGW in six countries (Brazil, Ecuador, Peru, South Africa, Thailand, US). In the iPrEx study, efficacy was 44% overall, but adherence, based on detection of tenofovir in blood samples, was only approximately 50%; among those with detectable tenofovir, particularly concentrations suggesting daily dosing, the investigators estimated HIV protection on the order of 90% or greater. Thus, iPrEx raised important questions about PrEP: when offered who will take it up, will those receiving PrEP maintain sufficient adherence, and will HIV protection in real world settings be similar to the 90+% estimated out of the clinical trial? After the iPrEx results were released, the investigators initiated a follow-on study, the iPrEx Open Label Extension (iPrEx OLE), which offered access to PrEP to former iPrEx participants (as well as participants in two smaller US PrEP trials that concluded at approximately the same time), in accordance with Good Participatory Practices. The primary iPrEx OLE results were presented in an oral abstract (Grant, abstract TUAC0105LB) and published simultaneously online in Lancet Infectious Diseases
(http://www.natap.org/2014/HIV/072314_02.htm)..... [Ole, Pre-Exposure Prohylaxis (PrEP) Initiative: Open Label Extension - "Adherence Has To Be Good, Not Perfect," http://natap.org/2014/IAC/IAC_71.htm]
iPrEx OLE offered enrollment into a prospective cohort to all men who had been in iPrEx and the related trials, regardless of HIV status or desire to use PrEP; follow-up was for up to 72 weeks and, at AIDS 2014, only the data from the HIV uninfected men were reported. For HIV uninfected men, PrEP was offered at enrollment and could be started anytime up through 48 weeks of follow-up; PrEP could also be stopped at any time. Thus, the goal of the iPrEx OLE investigators was to provide access to HIV prevention services, including PrEP but without a mandate that all participants use PrEP (or pretend to use PrEP but with low/no adherence). In total 62% of HIV uninfected men who had participated in the prior PrEP trials chose to be screened for iPrEx OLE (i.e., 28% declined study participation entirely) and 1601 enrolled. Importantly, 77 men acquired HIV between the end of the clinical trial and screening for iPrEx OLE, testifying to the continued need for prevention services in this population. Of the men who enrolled, 72% started PrEP at enrollment and an additional 6% started PrEP later. PrEP initiation was greater (81% vs. 75%) for those who reported receptive anal intercourse without a condom during the period just prior to enrollment (i.e., those at greater HIV risk). In addition, there was no increase in receptive anal intercourse without a condom during the follow-up period, suggesting PrEP did not lead to risk compensation. Those who declined PrEP cited a variety of reasons, most prominently concerns about side effects (50% of those declining), a desire not to take pills (24%), plans to avoid HIV in other ways (14%), and concerns about stigma (being thought of as HIV infected [7%] or identified as MSM/TGW [3%]). Among those receiving PrEP, HIV incidence was 1.8 per 100 person-years, compared to 2.6 per 100 person-years among those choosing not to receive PrEP. The difference was magnified when adherence to PrEP was taken into account (as measured by detection of tenofovir diphosphate in dried blood spots, currently a research assay not commercially available). HIV incidence was 4.7 per 100 person-years among those with no PrEP detected in blood samples, 2.3 per 100 person-years among those with evidence of PrEP use but less than 2 pills per week (a 44% reduction in risk, not statistically significant), 0.6 per 100 person-years with blood levels suggesting adherence of 2-3 pills per week (an 84% risk reduction, 95% CI 21-99%), and 0 per 100 person-years among those who blood levels consistent with taking ≥4 pills per week (100% risk reduction, 95% CI 86-100%). [Ole, Pre-Exposure Prohylaxis (PrEP) Initiative: Open Label Extension - "Adherence Has To Be Good, Not Perfect," http://natap.org/2014/IAC/IAC_71.htm]
Detection of tenofovir in blood samples was significantly associated with having receptive anal sex without a condom, having ≥5 sexual partners in the prior 3 months, older age (over 30 years, compared to <25 years) and having completed secondary schooling. Thus, in summary, the iPrEx OLE data demonstrate that MSM/TGW who had participated in the iPrEx study (and related smaller trials) were interested in using PrEP, once it was proven to be efficacious, and were able to use PrEP with sufficiently high adherence to garner HIV protection. Importantly, PrEP use was greater during periods of higher HIV risk (e.g., among men reporting receptive anal intercourse without a condom), and did not lead to greater risk-taking, both of which are tremendously encouraging. As an open-label extension to a randomized trial, the iPrEx OLE population was PrEP-experienced prior to starting the OLE study and thus their uptake and use may not fully reflect real-world delivery of PrEP. The results are highly informative nonetheless. Ongoing demonstration projects of PrEP - which are enrolling participants who had not been in prior PrEP clinical trials - will continue to provide critical information about who wants PrEP and how well it is used.
A second presentation from the iPrEx OLE investigators (Koester, abstract TUAC0102) reported on qualitative data from US sites participating in that study (Boston, Chicago, San Francisco). The investigators completed 60 in-depth interviews with MSM participants, with a goal of understanding sexual behavior while receiving PrEP (aiming specifically to understand whether PrEP resulted in risk compensation). Men reported a variety of condom use practices prior to starting PrEP, from never to routine, and did not significantly change their practices once initiating PrEP; the exception was young men, who increased their condom use. PrEP did lead to decreased stress, fear, and guilt related to sex. The investigators summarized their results as showing that PrEP was used as a supplement to existing HIV prevention strategies not a substitute.
An ongoing randomized, placebo-controlled trial (ANRS Ipergay) [http://www.natap.org/2014/IAC/IAC_27.htm ] is testing intermittent use of Truvada® for PrEP (dosed as 2 pills prior to sex and 2 after). The investigators reported initial adherence results from 153 participants at AIDS 2014 (Fonsart, abstract TUAC0103, presented by Molina). By self-report 53% used PrEP as scheduled during their last sex act and an additional 28% used PrEP but not exactly according to the recommended schedule. More than 80% of subjects had tenofovir detected in blood samples. Thus, these preliminary data suggest adherence to intermittent PrEP can be reasonable; further work in this ongoing study will provide greater information about the feasibility (and effectiveness) of intermittent PrEP for HIV prevention.
There were a limited number of presentations related to PrEP in non-MSM populations. The TDF2 study team - a randomized trial of FTC/TDF PrEP among heterosexual men and women in Botswana that demonstrated efficacy for HIV prevention in 2011 - presented initial information from their open-label extension study (Chirwa, abstract TUAC0104, presented by Taylor). Of 1219 TDF2 study participants 736 were contacted, 334 were screened, and 229 started PrEP; notably, the TDF2 open-label extension began in 2013, more than two years after the TDF2 itself concluded, which likely contributed to the large number unable to be contacted. An additional 38 subjects started the study but did not start PrEP; those who started PrEP felt themselves at greater risk for HIV (or had a partner at greater risk for HIV) than those who did not start PrEP. Of the 229 participants who started PrEP, 125 (54%) completed 12 months of follow-up; those who completed follow-up had felt themselves at higher risk for HIV. FTC/TDF PrEP was safe and well-tolerated. Thus, these preliminary data suggest interest in PrEP, particularly for those feeling themselves at risk for HIV. Adherence analyses and qualitative work are ongoing.
Another oral abstract (Curran, abstract TUAC0101, presented by Johnson) assessed the potential costs of incorporating HIV self-testing into PrEP implementation, using a model from Kenya as an example. The authors investigated three scenarios for HIV testing in a PrEP delivery program: 1) 3-monthly facility-based testing (the standard of care), 2) 3-monthly HIV self-testing at the facility ("supervised" self-testing, which would decrease counselor costs) and 3) alternating 3-monthly facility-based testing and HIV self-testing at home. Both of the self-testing options would be less expensive than facility-based only testing if self-tests were ≤$3 each, due to cost savings in counselor time. These preliminary findings should spur PrEP demonstration projects to evaluate how HIV self-testing can be incorporated into PrEP delivery.
One qualitative study from Kenya and South Africa explored the potential for risk compensation among women if they were to be offered PrEP (importantly, the study did not provide PrEP to this population) (Corneli, MOPDD0104, http://pag.aids2014.org/session.aspx?s=1097). Some women reported that PrEP offered effective HIV protection that decreased the need for condom use. Additionally, women reported that condoms were often a source of conflict in relationships and sex without a condom was often necessary for financial assistance from partners; PrEP would provide an opportunity for some resolution to these challenges. The work emphasizes the need for empirical evidence to understand how women at risk for HIV will incorporate PrEP into their lives.
Treatment as prevention (TasP) is a potent antiretroviral based HIV prevention strategy. A Thursday bridging session was devoted entirely to TasP (http://pag.aids2014.org/session.aspx?s=1957). The session discussed TasP from the science, community and leadership perspective and offered insights into what implementing TasP entails including social, cultural and economic aspects. The changing relevance of the terms "treatment as prevention" and even "early treatment for prevention", given quickly evolving guidelines about the time to start treatment, was pointed out early in the session. The session offered an opportunity to better understand current treatment guidelines and their possible impact on various epidemic settings, have a better comprehension of community perspectives, and be able to identify implications, trade-offs and operational requirements of implementing TasP on a larger scale. The session also included a report back from the 2014 Treatment as Prevention Workshop. The science perspective of treatment as prevention showed the latest data on TasP as well as reviewed how implementation is playing out in practice and offered examples of best practices from South Africa. The community perspective analyzed the role of the community in strengthening TasP approaches and addressing bottlenecks, including questions of access and ethical aspects. A number of issues around human rights become a focus of concern in TasP implementation. The leadership perspective examined the challenges of large-scale implementation of TasP and focused on the Brazilian case to discuss the role of various factors (programmatic, community , policy) and particularly the role of commitment and leadership. Brazil's 2014 goal is to put at least 100,000 more people on treatment.
An oral abstract session on Wednesday also focused on treatment as prevention (http://pag.aids2014.org/session.aspx?s=11180). Encouraging data were presented from the ANRS TasP trial (Iwuji, abstract WEAC0105LB) which is being conducted in the Africa Centre in KwaZulu-Natal, South Africa in which HIV prevalence is 24%. The two-arm cluster randomized trial is evaluating treatment of HIV infected persons with ART at all CD4 counts versus at CD4 counts <350 cells/μL (the South Africa standard of care), both following community GPS mapping and home-based HIV testing. Data were presented from the pilot phase of the study, conducted starting in 2012, evaluating feasibility and acceptability. GPS was used to identify households, members underwent home-based HIV testing with HIV+s being referred immediately to HIV care and HIV-s undergoing semi-annual HIV testing. In control clusters, HIV+ persons were offered ART if their CD4 count was <350 and in intervention clusters, ART at all CD4 counts. Overall, 78% of 12,910 eligible persons were contacted, and HIV status was ascertained in 82%, of whom 31% were HIV+ based on self-report or newly diagnosed by rapid testing at that visit. Of the HIV+ persons, 47% were currently in care and approximately 40% were on ART. ART initiation was 34% within one year after referral - 39% in the control arm and 55% in the intervention arm. Among those not on ART at referral who were linked to TasP clinics, ART initiation after referral was 64% - 45% in the control clusters and 85% in the intervention clusters, including 80% of those with CD4 counts >350. The investigators presented their observed versus expected estimates, which showed that linkage to care amongst newly diagnosed as HIV was slower than expected (48% within 6 months compared to 70% expected uptake) but ART uptake was high among those referred, particularly among those linked to TasP clinics. Thus, there was high community acceptance of this TasP trial, with need for continued efforts at linkage to care and ART initiation; the next phase of the trial, evaluating efficacy, began in June 2014.
Qualitative research from the HPTN 071 (PopART) project, another randomized trial of TasP being conducted in South Africa and Zambia, explored community perceptions of the concept of TasP (Bond, abstract WEAC0103). While there was support in principle for this work, concerns about new acronyms (TasP), delinking treatment and prevention, and potentially increasing stigma against persons with HIV were raised. These concepts will require continued and detailed follow-up.
The real world effectiveness of ART for prevention among HIV serodiscordant couples was evaluated using programmatic data from Zambia (Wall, abstract WEAC010). Zambia scaled up couples voluntary counseling and testing (CVCT) over the past seven years. In the past four years, Zambia has tested 153,547 couples. The costs and effectiveness of both CVCT and ART were estimated, although the authors did not provide details of the analysis. Both CVCT and ART appeared to be associated with reduced HIV incidence; ART effectiveness, not unexpectedly, was lesser in this real-world implementation than in HPTN 052. Data on adherence, viral suppression, resistance and transmission linkages were not presented, which would be useful to understand the low effectiveness of therapeutic ART on HIV transmission. They recommend scale up of CVCT in government HIV clinics to increase ART adherence and retention in care among HIV serodiscordant couples.
Finally, a meta-analysis of 37 studies (30 from Africa, 4 Thailand, 2 Brazil, 1 India) assessed whether use of ART was associated with decreased condom use among persons in low- and middle-income countries (Kennedy, abstact WEAC0104). In the analysis, persons with HIV on ART were more likely to report using condoms than those not on ART, and results were consistent across multiple sensitivity and subgroup analyses. These results suggest the potential for a behavioral synergy with ART for reducing HIV transmission risk.
ART is central to prevention of mother-to-child transmission of HIV (PMTCT) and a session on Thursday was devoted to implementation of Option B+ (i.e., provision of ART to all women in pregnancy with the intention of life-long therapy thereafter) (http://pag.aids2014.org/session.aspx?s=1132). Three presentations - two from Malawi and one from Zambia noted high loss to follow-up of women eligible for or initiating ART in pregnancy. In Malawi, women who were lost were traced, to understand reasons for loss to follow-up and clinical outcomes (Tweya, abstract THAX0101); of those who could be found (40%), half had stopped ART, with primary reasons being travel and lack of transport. The remainder had transferred care. In Zambia, women with a new HIV diagnosis in a PMTCT program were assessed for subsequent linkage to HIV care (Okawa, abstract THAX0103, presented by Chirwa). Only half were linked, of whom only half again initiated ART. Younger women (≤20 years of age) were less likely to initiate ART, as were those attending rural health centers not providing HIV treatment. These results emphasize that strengthening the connections between services is a priority. Finally, a second study from Malawi assessed facility type and linkage to care for pregnant women (van Lettow, abstract THAX0102, presented by van Oosterhout). In sum, 18-32% of pregnant women were not tested for HIV at antenatal clinics, influenced by client:staff ratios, stock outs of test kits, and models of care (i.e., higher testing rates at centers that initiated and followed women on ART rather than referring for ART initiation and/or follow-up). In addition, 7-20% of women initiating ART defaulted Option B+ by 6 months, associated with patient volume at the ART center and model of care delivery.
A preconference symposium on Sunday (http://pag.aids2014.org/session.aspx?s=1048) focused on the recently released WHO Consolidated Guidelines on HIV Prevention, Diagnosis, Treatment and Care for Key Populations, which focused on men who have sex with men (MSM); people in prisons and closed settings; people who inject drugs, and transgender people. Criteria for being defined as a key population for HIV included experiencing high degrees of risk, vulnerability, stigma and discrimination and compromised access to services. Populations, such as girls and young women, who face increased risk and vulnerability in particular geographical settings such as sub-Saharan Africa, are not included as 'key populations' in this document. The main new recommendations in the WHO guidelines were: 1) a strong recommendation for the use of oral HIV pre-exposure prophylaxis (PrEP) among MSM, based on the evidence shows that PrEP works and is safe if taken as prescribed; and (2) more clear directions for the use of harm reduction and opioid substitution therapy with injection drug users, and that evidence from the demonstration projects indicates that feasibility looks positive in this population. Strong emphasis was made on the need to address critical enablers among key populations, including decriminalization of these populations' practices regarding the transmission of HIV, improved access to prevention and care, reduction of stigma and discrimination, and community empowerment. The potential impact of combined biomedical interventions was presented from a modeler's perspective if health maximization, not just simple cost-effectiveness calculations.
A number of symposia at AIDS 2014 focused on key populations, based on risk behavior or mobility (e.g., migrant populations). One oral abstract session devoted entirely to key populations can be found at http://pag.aids2014.org/session.aspx?s=1124. Structural barriers include laws and lack of social protection, such as access to housing, migrants' access to HIV-related services, laws criminalizing sex work, sex between men, prostitution, and injection drug use need to be repealed. Structural interventions could have a particularly big role in reducing vulnerability in key populations, and creating enabling environments with greater access to stable living conditions and health care.
In a session devoted to health systems for HIV treatment and prevention, a presentation from the Kenya Ministry of Health described establishment of a Technical Support Unit to facilitate key population service delivery (Anthony, abstract MOAE0304, http://pag.aids2014.org/session.aspx?s=1129). Key population reporting increased from 52% to 92% over a 6-month period, and contact with at-risk female sex workers and MSM more than doubled, including HIV testing and STI screening. This example of enhancing government attention to key populations is a model for other settings.
Transgender persons face greater barriers to health care access, social acceptance, and employment, and a need for structural interventions, which were addressed in session on Wednesday (http://pag.aids2014.org/session.aspx?s=1112). In one abstract (Stanton, WEAD0301) a transgender transitional housing program in New York City was described that provided scattered site housing for transgender people living with HIV (among whom previous research indicates HIV prevalence is 29%), which included case management, group support, and access to HIV care services. They compared 90 transitional housing program residents to 90 homeless HIV positive clients on the same HIV medication, matched by age and race, and found 67% versus 32% were virally suppressed, respectively. The qualitative interviews indicated that stable housing provided less chaotic and risky environments, increased service utilization and increased ART adherence.
Transitional housing could be as an effective structural intervention for marginalized and stigmatized communities living with HIV
In a second abstract (Aristequi, WEAD0303) an example of an enabling structural intervention with passage of a Gender Identity Law in Argentina in 2012, which included, full development of individuals' gender identity, rectification of sex in birth certificates and national ID (without requiring judicial, medical or administrative authorization), and access to comprehensive healthcare, hormone treatments, and partial or total surgical interventions. Two focus groups were conducted following the enactment of the law, which indicated that transgender people described that legal recognition of gender identity provided recognition, a resource for when they face stigma and discrimination, empowerment, and greater engagement with health care, but that needs to be accompanied by professional training on transgender health issues and regional transgender clinics.
A Monday afternoon oral abstract session was devoted to epidemiology and prevention in young persons (http://pag.aids2014.org/session.aspx?s=1135), a priority population given epidemiological data indicating a 157% increase in new infections over the past decade. This age group also accounts for an increased proportion of new infections suggesting that while progress has been made in other age groups, this remains a neglected area.
Preliminary data were presented on an interactive video game (PlayForward: Elm City Stories) that was designed to foster skill development and increase HIV knowledge and awareness among adolescents (Fiellin, MOAC0103). The trial has enrolled 333 teens in the US; preliminary data were shown from 125 with up to 6 months follow-up. Adolescents ages 11-14 years were randomized either to 6-weeks exposure to a "serious" video game where they selected a character (aspirational avatar) where they were faced with a series of risks and consequences using message framing or other non-serious video content (e.g., "angry birds"). Significant improvements in HIV-related knowledge were observed in the video game intervention, were sustained for 3 and 6 months, and were correlated with the amount of time exposed to playing the game. The investigators will study whether the video game improves risk-taking behaviors.
Data were presented (Meinck, MOAD0101) from a community-based study in the Western Cape and Mpumalanga provinces in South Africa with door to door sampling of households with children ages 10-17 that began in 2010 and had a one year follow-up with 97% retention. Children completed a confidential self-report questionnaire about parental chronic illness, physical, emotional and sexual abuse. Structural factors were correlated with sexual risk behavior; 57% of girls with a parent who was sick with AIDS, had been abused or reported hunger reported a history of transactional sex, compared to 13% who reported a history of abuse or hunger, 7% of girls with a parent with AIDS, and 1% of girls with none of these factors. The investigators concluded that children in AIDS-affected families are at higher risk for child abuse victimization, and this risk is associated with higher levels of household poverty. Effective poverty reduction may be a key intervention to reduce child abuse in AIDS-affected families in South Africa.
In another study from South Africa (Cluver, MOAC0104) a national longitudinal survey of 6850 adolescents in 3 provinces in South Africa was reported, in which 6 sites had >30% HIV prevalence. During the study, South Africa scaled up child welfare grants ($20 per month divided across the family) from age 14 to up through age 17 as a poverty alleviation intervention. In the longitudinal study of 3515 teenagers in South Africa, unconditional cash transfers in the form of child grants (which households use to pay for school uniforms, food parcels, transportation) was compared with households who also received care (positive parenting, school counselors, free school meals if they attended school, teacher support, and/or participated in a food garden). For boys, cash alone had no impact on HIV risk behaviors, but cash plus care reduced them by 50%. For girls, cash alone reduced risk by 37%, but cash plus care reduced it by 45%. The investigators concluded that non-conditional child grants provided to the household, combined with care (teacher and positive parenting), mitigate structural risks of HIV, including reducing multiple partners.
A combined intervention was presented that aimed to economically empower women and reducing intimate partner violence, adapted from the Stepping Stones intervention (sessions about communication, behavior, sex and love, contraception, safer sex, gender-based violence) called Creating Futures (Jewkes, WEAD0503, presented by Gibbs,). The intervention was implemented in two informal settlements near Durban, South African among 232 young people average age of 22 years. There was no control group, but in comparing baseline and 12 month self-reported outcomes, both women and men reported significant improvements in monthly incomes and gender attitudes, but there was no impact in physical intimate partner violence but a modest decline among sexual intimate partner violence in the prior 3 months reported by women. The investigators recommended additional impact evaluation of structural interventions without financial incentives or cash transfers.
In another session (Wagman, THAC0103, presented by Nalugoda, http://pag.aids2014.org/session.aspx?s=1105), authors presented a cluster randomized trial of an intimate partner violence (IPV) and HIV prevention intervention on emotional, physical and sexual abuse, sexual risk and HIV incidence in Rakai, Uganda, which was embedded in the existing Rakai cohort study. The enhanced HIV service intervention was delivered to four community clusters (n=5,339), with outcomes (IPV-conflict tactic scale and reported risk behaviors such as spousal rape, total/extra-marital partners, condom use) compared with seven standard HIV service control clusters (n=6,112). Outcomes were measured across three follow-up rounds between 2005-06 and 2008-09. The intervention was associated with significantly lower proportions of women experiencing physical and sexual IPV in comparison to the control arm. There were no significant differences in the proportion of men reporting perpetrating emotional, physical or sexual IPV between intervention and control arms. The intervention was associated with significant reductions in spousal rape, increased rates of HIV status disclosure, and reduced overall HIV incidence and HIV incidence among men (and non-significant reductions in HIV incidence among women). The authors concluded that HIV prevention campaigns should ideally include interventions to reduce intimate partner violence.
From Lesotho (Nyqvist, THAC0101, presented by deWalque) authors reported follow-up of a short-term financial incentive study on HIV prevention in Lesotho in which 3,029 participants were randomized to either a control group or one of two intervention groups - a $100 or a $50 lottery ticket arm if remaining STI free. Outcome testing for syphilis and trichomonas was performed every four months over 2 years and HIV testing was at baseline and months 16, 20 and 24. Participants were eligible for lottery incentives it they tested STI negative. After two years, HIV incidence was significantly lower in the among intervention group participants (OR 0.79, 95% CI 0.62 - 0.99), with the effect greatest among women (OR 0.78, 95% CI 0.62 - 0.98) and in the higher prize lotteries (OR 0.70, 95% CI 0.51 - 0.95). The risk of HIV acquisition was sustained 12 months after the lottery incentive among women but not men. There is substantial interest in the potential role of economic incentives to increase uptake of STI and HIV prevention behaviors; further research is needed to understand mechanisms and sustainability of the effect after the incentives stop.
Men who have sex with men (MSM)
An oral abstract from Siberia (Efremov, abstract MOAE0204, http://pag.aids2014.org/session.aspx?s=1125) described a combination HIV prevention program from MSM and transgender persons which demonstrated a significant increase in HIV awareness (by 34%), with accompanying prevention behaviors (specifically, HIV testing, serosorting and monogamy) but the level of condom use decreased from 49.6% to 41.3%. However, the program also noted an increase in violence, discrimination and closeted behavior in the population since the adoption of Russia's anti-gay "propaganda law" in 2013, emphasizing the important role that stigma can play in harming effective prevention. Another oral abstract, from Nigeria, reported that MSM had increased fear of seeking health care and increased feeling of no safe place to socialize after recent anti-gay legislation was implemented (Schwartz, abstract TUAD0305LB, http://pag.aids2014.org/session.aspx?s=1111).
An oral abstract session focused on MSM (http://pag.aids2014.org/session.aspx?s=1119). One presentation from that session (Holt, THAD0101) addressed risk reduction strategies used by Australian gay and bisexual men who had had anal sex without condoms with casual partners. Three-quarters of men consistently used at least one strategy for casual sex, with the most common strategies being serosorting by HIV status and condoms. HIV-positive men were much more likely to rely on serosorting than HIV-negative men (60% versus 44 %). HIV-negative men were more likely to use a combination of strategies. A second (Ifekandu, THAD0103) presented qualitative research with MSM in Abuja, Nigeria, among whom an estimated 36% are HIV positive. MSM described multiple sexual partners, inconsistent condom use, transactional sex, poor access to services and harassment from the police, and societal pressures to maintain the appearance of being heterosexual. Most men used social networking websites to meet partners. This is one of several presentations of MSM in Africa, India and Jamaica, which indicate that in countries with harsh laws against homosexuality and entrenched homophobia, MSM have less access to HIV prevention, and are at significantly increased risk of HIV.
In another oral session on MSM (http://pag.aids2014.org/session.aspx?s=1128) [http://www.natap.org/2014/IAC/IAC_45.htm] a comparison of London and San Francisco highlighted the differences in the HIV testing care cascade (Brown, THAC0205LB). Using 2010 data from CDC and Public Health Evaluation data from London, whereas linkage to care within 3 months and retention in care were higher in London, an estimated 21% of HIV positive MSM are undiagnosed in London, compared to about 10% in San Francisco. London and San Francisco had an estimated 70% and 60% of diagnosed HIV infected MSM who were virally suppressed in 2010, which has gradually increased since 2006. The key challenges are increasing HIV testing in London, including among MSM who are in-migrants to London, earlier ART initiation (given ART guidelines are based on CD4 ≤350) and retention in care in San Francisco to achieve higher rates of viral suppression. Recreational drug use is an important issue, as suggested by rising hepatitis C prevalence in London, and high levels of recreational drug use in a recent London Shigella outbreak and based on research in high-risk venues (through the Chemsex study). The investigators highlighted the importance of increasing HIV testing, awareness, disclosure, and the importance of challenging the safety of serosorting.
From Thailand (Thienkrua, THAC0202 presented by Pattanasin,) HIV incidence among young MSM enrolled in the Bangkok MSM Cohort Study was estimated. Men were recruited at venues, Internet, male sexual health clinics, and social or friendship networks of outreach workers between 2006 and 2010. Of the 1744 young MSM, HIV-prevalence was 21% and incidence was 7.5 per 100 person years. Multivariate risk factors for HIV-incidence were unprotected receptive anal intercourse AHR, 2.8), being paid for sex (AHR 2.2), having a casual sexual encounter at a sauna or at home (AHR 1.9 and 1.6, respectively), and living alone or with a roommate (AHR 1.5). This high HIV incidence among young MSM in Bangkok highlights the urgent need for combination HIV prevention interventions.
Injection drug users
In an oral abstract (Cook, WEAE0201, presented by McLean http://pag.aids2014.org/session.aspx?s=1124), the big gap in need versus funding for harm reduction for people who inject drugs was discussed. The core harm reduction measures include not only needle and syringe programs, opioid substitution therapy, and targeted education and outreach, but also HIV testing and counselling, antiretroviral therapy, condom programming and prevention, diagnosis and treatment of sexually transmitted infections, viral hepatitis and tuberculosis. Only 7% of the need is being met with current funding. In the same session, the Global Drug Commissioner, Sir Richard Branson joined the conference by video link, and said the global war on drugs had failed both in terms of controlling drug trafficking and consumption and of public health outcomes, particularly in relation to HIV and hepatitis C.
Female sex workers
A session (http://pag.aids2014.org/session.aspx?s=1120) focused on different interventions for the heterogeneity in sex workers are heterogeneous and some populations are particularly hard to reach, including adolescent bonded sex workers (Bangladesh), street-involved girls/women (Ukraine), adolescent female sex workers (China), male sex workers (Sydney) and regular male partners of female sex workers (Kenya). Approaches included social media, promoting human rights and engaging brothel keepers, and addressing needs beyond HIV, including housing needs through provision of half-way house access for street-involved girls and women in the Ukraine. Another session (http://pag.aids2014.org/session.aspx?s=1001) described gaps in prevention and care for sex workers.
From Kenya (Akolo, MOAD0204), baseline data on the forgotten population of regular partners of female sex workers were presented; men were recruited to 7 clinics in Nairobi. Of 732 regular male partners, 2/3 were married and about 1/3 had other partners, only 2.2% demonstrated correct condom use, 4.9% tested HIV positive, and STI prevalence was high.
From Mexico (Robertson, TUPDC0201, presented by Sthrathdee), correlates of HIV incidence and HIV/STIs were explored in a study of 214 female sex workers and their intimate partners recruited from the border of Tijuana, Mexico and San Diego, US. Couples did not use condoms except with paying customers and there was a high level of drug use/injection. Incidence of any non-HIV STIs was very high for women and higher than among men (~22 vs ~8 per 100/p-yr), with the most common STIs detected chlamydia>syphilis>GC. They observed a 2 fold higher incidence of STIs among FSWs who reported stimulant use (crack, cocaine, ATS) and a 60% lower incidence among FSWs with regular sex work partners.
Time-varying predictors of any STI for women were negatively correlated with "hitting" back if their male partner assaulted them. Reduced STI risk for the male partners was associated with the FSW having "regular" clients. The study highlight the importance of FSWs intimate and commercial relationships in influencing HIV/STI transmission dynamics. Having access to self-empowerment and evidence-based drug treatment, as well as potential benefits of couples-based interventions for FSWs with regular partners to address stimulant use, conflict resolution, and HIV/STI prevention is key to HIV/STI prevention.
From Uganda (Hladik, TUPDC0202), the high HIV transmission potential and late diagnosis among a respondent-driven sampling of female sex workers in Kampala was discussed. A total of 1481 FSWs were enrolled based on self-report of exchanging sex for money in the past 6 months, among whom HIV prevalence was 31.5%, which is 3 times higher than the general female population despite a relatively short duration of sex work (median of 2 years). Among the HIV positive FSWs, only 50% reported ever having tested and knowledge of their HIV infection. The median CD4 was 440 and viral load was 3160 copies/ml with 63% unsuppressed. Overall, 18% of the HIV-positive women reported ART use, but half had unsuppressed viral load; the median viral load was 1040 copies/ml. This study shows the value of expanded biomarkers in HIV surveys of key populations.
From Zimbabwe (Cowan, THAC0305LB), baseline data from the SAPPHIRE study were shown. The SAPPHIRE study will evaluate impact of community mobilization education, behavioral, and clinical services (contraception, STI and HIV services), and peer support for ART adherence in FSWs. Among 2722 who were sampled through respondent-driven sampling, the overall weighted HIV prevalence was 56%, of whom 61% reported they knew they were HIV positive and 50% were virally suppressed, indicating that improvement is still needed in HIV testing and linkage to HIV care and ART. Consistent condom use with clients was reported by 59% and intimate partner violence was reported by 37%. Although treatment among FSW has increased since 2011, HIV prevalence remains high, the authors recommend further progress is required to increase uptake of testing, treatment and prevention.
The prevention benefit of voluntary medical male circumcision (VMMC) in reducing female-to-male HIV transmission risk was demonstrated in three landmark randomized trials completed in 2006. Implementation of circumcision services has become a clear public health priority.
A poster discussion session (http://pag.aids2014.org/session.aspx?s=1104) presented new research on epidemiology and delivery of VMMC.
In one prospective study of 4716 HIV serodiscordant couples from Kenya and Uganda, the relationship between male circumcision status and syphilis acquisition (in both men and women) was explored (Pintye, abstract MOPDC0103). A total of 221 incident syphilis cases were identified, based on a >4-fold change in RPR titer with a positive confirmatory result during study follow up. Male circumcision status was determined by physical examination. Male circumcision was associated with decreased risk of incident syphilis, ranging from 40-70%, in men and women. The biggest reduction in incident syphilis among men was among HIV-infected circumcised men (62% reduction). Among women with a circumcised male partner, the risk reduction was 48% among HIV-infected women and 75% among HIV-uninfected women. Thus, decreased risk of incident syphilis among both men and women is a further benefit of male circumcision, in addition to the well-documented reduction in HIV incidence among men.
A randomized trial of conditional economic incentives on uptake of VMMC was presented (Thirumurthy, abstract MOPDC0104, presented by Agot). The study was conducted in Nyanza, Kenya among 1504 HIV negative men ages 25-49 years old, who were randomized to no incentive or incentives of approximately US$2.50, $8.75, or $15 (provided as food vouchers at the time of VMMC). The $8.75 and $15 equivalent incentives were associated with a 4.3 and 6.2-fold increased uptake of VMMC, respectively, with an absolute increase in uptake that was relatively modest (from 2% in the control group to 9% in the US$15 group). However, given the relatively low prevalence of MC prevalence in the 25-49 year old age in this region, even after 5 years after roll-out of VMMC in Kenya, this increase in VMMC uptake warrants further evaluation. (These data were also published concurrently online in JAMA http://jama.jamanetwork.com/article.aspx?articleID=1890398&utm_source=Silverchair%20Information%20Systems&utm_medium=email&utm_campaign=JAMA%3AOnlineFirst07%2F21%2F2014.)
An assessment of post VMMC-risk compensation in the Zambian national VMMC program in which 725,000 circumcisions have been conducted since 2008 was presented (Hewett, abstract MOPDC0105). They conducted three rounds of data collection in a population cohort, and evaluated 5 indicators for risk compensation: sex with ≥2 partners in the past year, unprotected sex, sex after alcohol use, STI symptoms and paid for sex in the past year. Between 2998 and early 2013, 21% of the sample were recently circumcised, 16% between the first and third round of data collection. The authors found no consistent evidence of risk compensation in two years; some indicators suggested that circumcised men were less risky than uncircumcised men.
In another study (Kaufman, MOPDC0106 presented by DeCelles, abstract) a sport-based intervention (Make the Cut, or MCuts) to increase VMMC in Zimbabwe was presented. The intervention was developed by Grassroot Soccer in 2012, which included a 60 minute session facilitated by circumcised men, delivered with soccer teams along with referral cards, poster, biweekly SMS reminders, and transport. 64 soccer teams (1152 men) were randomized to receive the intervention from soccer pros or non-pros (these arms were combined in the analysis), or the control condition. VMMC uptake was measured over 4 months, and was increased 9.8 times in the intervention compared to the control group (from 0.4% to 3.5%). There was lower uptake of VMMC among men older than 30 years. Although the overall effect was modest and the study had less statistical power than planned (17 teams dropped out), it suggests a potential strategy which should be evaluated for VMMC demand creation among males of various ages.
In a session about scaling up VMMC (http://pag.aids2014.org/session.aspx?s=1121), the Kenyan national program described their rapid results initiative, which achieved ambitious targets in HIV testing and counseling and VMMC (Koros FRAE0101). During two months in 2013, almost 1.5 million Kenyans were tested (25% of the program total in 2013), and 76,902 men were provided with VMMC services (comprising 45% of the 2013 annual program total), with a low adverse event rate of 0.6%. The Kenyan program demonstrates that national campaigns can rapidly deliver HTC and VMMC services with a low rate of adverse events with VMMC.
Given the shortage of physicians to provide surgical VMMC services, non-physicians were successfully trained to deliver VMMC in a rural area in northeastern Uganda (Tumuheki, FRAE0102); 2439 men were circumcised (almost 10% of the TASO total VMMCs throughout Uganda), and the adverse event rate was 0.1%.
Given the human resource constraints for performing surgical VMMC in many African settings, there has been an interest in VMMC devices. An analysis of the cost-effectiveness compared the PrePex device to surgical VMMC in two Lusaka VMMC clinics (Vandament, FRAE0104). Since the PrePex device is approved for VMMC for men ≥18 yrs, which comprise 60% of the men being circumcised in Zambia, the current cost of $20 per PrePex device, cost parity is only achieved if there was 81% uptake among adult men. Increased commodity costs offset human resource cost savings. The results did not change substantially with a lower device cost (i.e., $12 per PrePex device). One limitation was that staff and resources for the PrePex study were separate from routine surgical staff and supplies; the sites were providing 2 circumcision methods in parallel, which is likely less efficient than a "mixed model" site where staff and resources could be flexibly allocated across methods in response to demand. The authors concluded that further studies are needed to evaluate the economics of PrePex-only sites, and demand creation models targeted to PrePex.
HIV incidence among women was assessed in Orange Farm, South Africa (Jean, FRAE0105LB). VMMC has been scaled up in Orange Farm since 2007, as reflected in MC prevalence increasing from 11% to 53% between 2007-2011, and during which time HIV prevalence decreased by 50% among circumcised men. In this study, 3 independent surveys were conducted in 2007, 2010, and 2012 among 4538 women ages 15-49 years who ever had sex. HIV prevalence was 22.4% among the 1363 women who only had circumcised partners and 36.6% among women had uncircumcised partners (a 15% reduction in the adjusted prevalence rate ratio). Using mathematical modeling, women having only circumcised partners were estimated to have a lower risk of incident HIV infection; the estimated reduction in HIV incidence was 16.9% among all women and 20.3% among 15-29 yr old women. This study indicates a significant prevention benefit of VMMC for women as well as men.
Reproductive health and HIV
Hormonal contraception is used widely and plays an important role in preventing unintended pregnancies and reducing maternal morbidity and mortality. However, some, but not all, prospective observational studies have found an increased risk for women to acquire HIV infection when they are using hormonal contraception, especially injectable depot medroxy-progesterone acetate (DMPA, otherwise known as branded Depo-Provera). This injectable contraceptive is a popular contraceptive method used worldwide, including in southern and East Africa where HIV is highly prevalent; it is easy to use, fast to administer, and can be used discretely. For women in regions where HIV rates are high and injectable contraceptive use is common, understanding the possible link between injectable contraceptive use and HIV infection is a public health priority. The challenge with studies on this issue is that observational data are potentially open to bias, and hormonal contraceptive users could be less likely to use condoms (and, even more importantly, less likely to use condoms but more likely to falsely report using them), which could result in confounding in the analysis. At AIDS 2014, an oral abstract session was devoted to the challenging data relating contraception and HIV (http://pag.aids2014.org/session.aspx?s=1115).
A large individual patient meta-analysis, which included 18 studies following >37,000 women (experiencing 1830 incident HIV infection), was reported (Morrison, abstract THAC0503). Compared to women not using contraception, the adjusted hazard ratio for HIV acquisition was 1.50 (95% CI 1.24-1.83) for DMPA, 1.24 (95% CI 0.84-1.82) for norethisterone enanthate (NET-EN) another injectable progestin, and 1.03 (95% CI 0.88-1.20) combined oral contraceptive pills. There was no modification of the contraceptive-HIV relationship by either age or HSV-2 infection status. Interestingly, the authors created a scale to assess the potential risk of bias in the contributing studies and found that studies at lower risk of bias also had lower risk estimates for DMPA. Thus, in this very large compilation of data, DMPA, but not NET-EN or oral contraceptives, was associated with increased HIV risk. The authors suggested that a randomized trial of different contraceptives may be the only way to provide more clear data for this critical question.
In contrast, data from a 17-year prospective cohort of HIV serodiscordant couples in Zambia found no relationship between injectable contraceptive use and HIV risk (Wall, abstract THAC0504). In that study, 1393 couples in which the male partner was HIV infected were followed in a prospective observational cohort. HIV incidence was high, nearly 9% per year, and 207 incident transmissions were observed. In multivariate analysis, use of injectables (HR=1.0, 95%CI 0.6-1.5), oral contraceptive pills (HR=1.2, 95% CI 0.8-2.0), or implants (HR=0.9, 95% CI 0.3-2.6) in the past three months was not associated with HIV acquisition.
Finally, an abstract from the World Health Organization detailed WHO recommendations on medical eligibility considerations for different contraceptives for women at high risk for HIV (as well as for women living with HIV) (Gaffield, abstract THAC0505LB). In brief, recommendations did not change after a Q1 2014 consultation, compared to recommendations made in 2012. New WHO guidance recommends no restrictions on the use of combined hormonal contraceptives (pills, patch, vaginal ring, or injectable) or progestogen-only contraceptives (pills, injectable, or implants) for women with or vulnerable to HIV. Because risk of HIV acquisition associated with progestogen-only injectables is uncertain, women and couples at high risk of HIV infection should be informed about HIV preventative measures, including male and female condoms. WHO continues to actively review its recommendations in light of the accumulating evidence and strongly supports the need for further research to identify definitive answers that address concerns around increased biological vulnerability to HIV among women using progestogen-only injectables.
Thus, in summary, there remains uncertainty in whether hormonal contraception, particularly injectable DMPA, truly increases HIV risk in women. This will continue to be a priority topic for future studies and meetings.