|Home | About | Journals | Submit | Contact Us | Français|
Combination HIV prevention is a high priority for increasing the impact of partially efficacious HIV prevention interventions for specific populations and settings. Developing the package requires critical review of local epidemiology of HIV infection regarding populations most impacted and most at risk, drivers of HIV infection, and available interventions to address these risk factors. Interventions should be considered in terms of the evidence basis for efficacy, potential synergies, feasibility of delivery at scale, which is important in order to achieve high coverage and impact, coupled with high acceptability to populations, which will impact uptake, adherence, and retention. Evaluation requires process measures of uptake, adherence, retention, and outcome measures of reduction in HIV infectiousness and acquisition. Three examples of combination prevention concepts are summarized for men who have sex with men (MSM) in the Americas, young women in sub-Saharan Africa, and HIV-1 serodiscordant couples.
Control of HIV will require integrating a combination of evidence-based HIV prevention interventions, developed based on understanding of local epidemic patterns: local HIV prevalence, incidence, and epidemic factors and important risk factors among key populations. A growing number of interventions have shown partial efficacy against HIV, including knowledge of HIV serostatus, which leads to behavioral risk reduction particularly among those learning HIV-infected persons, condom use, medical male circumcision (MC) for HIV-uninfected men, and antiretrovirals when used as therapy (ART) by HIV-infected persons and pre-exposure prophylaxis (PrEP) by HIV-uninfected persons. However, no single HIV prevention strategy will have complete uptake or perfect adherence, thus will have less than 100% effectiveness, making the rational packaging of interventions into integrated programs one of the most critical research and implementation questions for HIV prevention for the near future.
Making combination prevention work requires careful selection of component interventions and objectivity in reviewing data about key risk factors, most affected populations, and efficacy of individual interventions. Parsimony in selecting possible interventions is important, as scale, coverage, affordability, and impact could be compromised with more complex combination packages. Pilot work is important to determine the acceptability and feasibility of scaling these interventions to achieve high coverage by prioritizing the subset of the population most at risk of HIV transmission or acquisition, and acceptability of interventions to those populations.
A key first principle for choosing the components of a combination intervention package is synergy – ideally that the effect of a combination of interventions is at least the sum of the parts, if not greater. Just like combination ART is most effective when the components are active against different parts of the viral life cycle, a combination of prevention interventions directed at different risk factors and avenues for HIV transmission may have the greatest combined impact. An example of targeting different paths for transmission among heterosexuals in a generalized heterosexual epidemic would be a combination strategy that includes: 1) MC for HIV- men, 2) ART rollout in HIV-infected persons (including partners of those men), and 3) behavior change interventions that reduce risk and increase uptake and adherence to these interventions. Interventions that seek to reduce infectivity in HIV-infected individuals are likely to be most synergistic with interventions that reduce susceptibility among HIV-uninfected individuals1. Mathematical models can be used to identify situations where interventions may have a ‘super-additive’ effect by reducing the basic reproductive number (Ro) of an infection below a critical threshold 2.
A second principle of combination HIV prevention is coverage, which is a function of access to the interventions and willingness of persons prioritized based on risk to utilize the interventions. A fundamental initial step towards achieving high coverage of HIV prevention interventions is HIV testing and knowledge of HIV serostatus, which is needed for targeting interventions to reduce HIV susceptibility or infectiousness. For HIV infected persons, prevention coverage entails breaking down the multiple steps in the cascade from HIV testing to linkage to care: clinic referral, ART eligibility assessment, pre-ART retention, ART initiation for those who are eligible, and sufficient adherence to achieve sustained viral suppression3, particularly among those most likely to transmit. Similarly, there is a cascade for ‘HIV prevention’ for persons who are HIV uninfected: learning one’s HIV status, uptake, and adherence to evidence-based prevention services, such as MC, as well as more user-dependent interventions such as pre-exposure prophylaxis (PrEP), coupled with prevention counseling. Achieving high coverage requires addressing the multiple steps in these ‘cascades’ beginning with knowledge of serostatus, demand stimulation (to increase awareness of HIV risk, benefits of and access to interventions), linkage, adherence and retention.
Economic analyses are important for estimating the cost-benefit of intervention packages in terms of HIV infections averted and lives saved. As an example, economic analyses of the impact of scaling up ART showed that at high ART coverage, over time, the intervention costs incurred are balanced by reduced costs of HIV-associated morbidity, mortality and incident HIV cases averted 4, 5. Economic analyses can estimate the initial cost outlay and time to ‘break even’ in costs based on infections prevented with a universal ‘test and treat” scenario 4, or scaling up ART coverage at different CD4 levels5. In assessing the health economic impact of combination prevention, the principles of synergy and coverage also apply. Interventions that reduce susceptibility, such as PrEP and MC, are cost-effective additions to ART for prevention through the synergistic effect of reducing HIV incidence at costs that, relative to life-long ART, are less6, 7. Nonetheless, costs for prevention can plateau over time. If strategies to reduce infectiousness and susceptibility are not adequately scaled up, then the number of new infections may remain constant rather than decreasing, thus increasing the overall costs in the long term4. One challenge for health economic analyses is to incorporate heterogeneity in costs, in addition to heterogeneity in disease modeling, because as programs are scaled up, smaller programs may have increased costs8. For HIV prevention interventions to be cost-effective over time, economic analyses support combining highly effective strategies, widespread coverage for ART, synergistic interventions that reduce susceptibility and identification of efficiencies in delivery of services throughout the cascade from testing, linkages to and retention in care6.
We describe three populations and relevant considerations in developing and testing possible strategies for combination HIV prevention in these populations to illustrate the differences in the populations to be reached, risk factors for new HIV infections, and interventions to be considered for combination HIV prevention packages.
MSM account for the majority of new HIV infections throughout North and South America. In the US, MSM comprised nearly two-thirds of new HIV infections in 2010; they are the only group in whom new infections are rising. Black MSM aged 13–24 years had more than three-fold the number of new HIV infections as white and Latino MSM, and increased the most from 2007–1010. Transgender women (people assigned ‘male’ at birth but identify as female and/or transgender) have extremely elevated infection rates in both North and South America9–12. Although HIV infection rates are high in Asia and Africa as well, less is known about drivers of infection in these populations; research is ongoing to determine the feasibility of reaching MSM in Asia and Africa13, which will guide future HIV prevention efforts.
Recent modeling on the epidemiology of new infections in MSM in the Americas suggests that more than one-third of new infections occur within main partnerships, and approximately two-thirds of those infections occur within partnerships that are not known to be serodiscordant14. To address the major risk factors for HIV infection in MSM populations in the Americas, several approaches are being currently piloted individually; those with high uptake and adherence will be combined to test their ability to achieve synergistic reductions in infections (Table 1).
Synergies for prevention in this population could be achieved by 1) addressing multiple components of the “transmission chain” (e.g., reducing per-act risk transmission through PrEP and reducing partner change rate through behavioral interventions); and 2) targeting populations with minimal overlap (e.g., PrEP for HIV negative men, HIV testing and linkage to ART to suppress viral load in HIV positive partners). The individual components of the proposed package are being piloted (Table 1) to determine which are desirable, scalable, culturally appropriate, potentially cost-effective, and have plausibility for having an impact on HIV seroincidence in both North and South America. Once the package is finalized, the integrated strategy will be tested in a two–stage process: a “vanguard” pilot study of the entire package, and if the combination impacts intermediate measures (e.g., uptake of and adherence to components and synergistic combinations), an efficacy trial.
One of the highest priorities for delivery and evaluation of integrated strategies for HIV prevention is young women in sub-Saharan Africa36. The magnitude of the HIV epidemic in heavily impacted areas such as KwaZulu-Natal province in South Africa is staggering, where a study of women attending family planning and STD clinics from 2004–7 found an HIV prevalence of 35.7% among young women whose median age was 22 years and an HIV incidence of 6.5/100 person-years37. Data from the Africa Center in KwaZulu-Natal showed highest HIV incidence (6.6/100 person-years) among women at age 24 and a peak HIV incidence of 4.1% among men five years later (i.e., age 29)38.
Drivers of HIV risk among young African women include unprotected sex, sexually transmitted infections in some populations, and age differences. Older partners have higher HIV prevalence and gender power disparities are greater, making it harder for women to negotiate safer sex39–41. Gender-based violence may be an important driver in settings, as indicated by the alarming rape statistics from parts of Africa. However, evidence-based interventions are not available for some of the social and behavioral drivers of infection among young women in Africa. Young African women also are at risk for unwanted pregnancy42, 43 among HIV-positive and HIV negative women, and require addressing cultural underpinnings and system access44, 45. Social norms often dictate that young women should not engage in sex and thus influence provider attitudes about contraception and impact young women’s access to prevention services, condoms, contraceptive services, sexual and reproductive health services, STI treatment, and HIV testing46, increasing their risk when they are sexually active. There is a tremendous unmet need for contraception worldwide47; the incidence of unplanned pregnancies among young women in Africa ranges from 4–16% in microbicide and HIV prevention trials48–50 and the risk of HIV acquisition and transmission has been found to be twofold higher during pregnancy51–53. A complicating factor is that some observational data indicate a 1.4–2.0 fold increased risk of HIV acquisition among women who use injectable hormonal contraceptives, particularly among depot medroxyprogesterone acetate (DMPA) users54, 55. However, the observational data are inconsistent and must be balanced by the safety, reversibility, contraceptive effectiveness and widespread availability and acceptance of DMPA among providers and women56, 57.
Given the overlapping risks for HIV acquisition, pregnancy and the need for both HIV prevention and reproductive health services, combination prevention for young African women could be delivered through integrated reproductive health and HIV prevention services which include behavioral, biomedical and structural interventions (Table 2). It will be important to pilot different program models for providing youth-friendly integrated reproductive health and HIV prevention interventions to achieve high uptake of HIV testing, PrEP and long-acting, reversible contraception options, as described below.
Population data from Africa suggest that a substantial fraction of new infections may occur within stable serodiscordant marital or cohabiting heterosexual relationships, with the majority of transmissions from the HIV-infected partner in the serodiscordant partnership and a substantial minority from an outside partner69–73. Understanding HIV prevention choices and targeting prevention strategies to this group are public health priorities. A number of countries have identified HIV serodiscordant couples as a priority population for implementation of novel HIV prevention strategies, given their high risk, smaller number for targeting relative to the general population, ability to be targeted for prevention efforts through promotion of couples HIV counseling and testing, and clear advantage to the partnership to avert HIV transmission. Importantly, during the past two years, two pivotal novel prevention interventions – ART (through HPTN 052) and PrEP (through the Partners PrEP Study) – demonstrated high efficacy for HIV protection in clinical trials conducted among HIV serodiscordant couples and are the core of a potential integrated strategy for combination prevention in couples. WHO has released guidelines for counseling and HIV-1 prevention for HIV serodiscordant couples, which emphasize the centrality of ART and PrEP, along with attention to other HIV prevention interventions including male circumcision74, 75. Determining how these efficacious interventions can effectively be delivered in real-world settings is the priority for combination prevention for this population.
HPTN 052 and Partners PrEP delivered their intervention strategies in the context of a combination package of HIV preventions, including frequent HIV counseling and testing, risk-reduction counseling (including as a couple) and access to condoms, male circumcision, ART according to national guidelines, and other prevention strategies (e.g., screening and treatment for sexually transmitted infections). The impact of these integrated services is reflected in substantially diminished HIV risk even in the delayed treatment or placebo arms of those trials. Thus, these clinical trials offer a model of combination HIV prevention for couples in the unique context of randomized clinical trials with intensive interventions and follow-up.
Recent WHO recommendations for earlier initiation of ART for HIV infected members of HIV serodiscordant couples require translation into programmatic contexts75. WHO guidelines are evolving in some settings to include lifelong ART for HIV-infected mothers regardless of CD4 count for the prevention of mother to child transmission76. Similarly, optimal strategies for PrEP delivery are yet to be defined and require demonstration projects and use of implementation science methods including demonstration of effectiveness among couples in which an HIV infected partner is not yet on ART, due to refusal or other reasons (i.e., PrEP as a bridge until ART is started and viral suppression achieved)77, 78. Indeed, neither ART nor PrEP use were associated with 100% protective efficacy, indicating a need for examining the effect of strategic integration of these two efficacious interventions, against a background of an effective prevention package. Thus, implementation science is needed to define:
A potential package of combination services to evaluate using implementation science is presented in Table 3.
Combination HIV prevention requires rigorous review of the epidemiology of HIV infection to identify populations most impacted and most at risk, drivers of HIV infection, and efficacy of available interventions to address these risk factors. Interventions should be considered in terms of potential synergies, feasibility of delivery at scale, and acceptability to populations. Evaluation of combination prevention packages requires a staged approach to evaluate acceptability, feasibility of delivery, and integration with other services, which should be followed by an evaluation of impact with outcome measurements, ideally based on HIV viral suppression in HIV infected persons and HIV incidence in uninfected persons. Economic evaluation is important for costing delivery components and to estimate the cost per HIV infection averted and lives saved.
Funding: National Institutes of Health R01 AI083034, R01 AI083034-02S2, R01AI083060-04, Directors Award RC4 AI092552, and HPTN UM1 AI068619.
Conflicts of Interest: None