Antiretroviral Therapy as an HIV Prevention Strategy
I. Definition of the Prevention Area
Antiretroviral (ARV) drugs, used to improve the health of people infected with HIV, have also been shown to reduce HIV transmission to HIV-negative people. By reducing viral load in the infected partner, ARVs have been shown to lower the risk of HIV transmission to the uninfected partner. Similarly, some recent studies of pre-exposure prophylaxis have suggested that ARVs may reduce transmission, even when taken by HIV-negative persons at risk for HIV infection. The topic of pre-exposure prophylaxis is covered in another brief.
II. Summary of the Evidence
HIV circulates at high levels in blood, semen, vaginal fluids, and breast milk in people with HIV who are not taking ARVs. Effective ARV therapy (ART) lowers HIV levels in these fluids, often below the limit of detection of standard viral load assays. Antiretrovirals reduce HIV levels in pregnant and breastfeeding women enough to significantly lower the risk of mother-to-child transmission. Multiple observational studies have provided evidence that a lower viral load--and taking ART to lower one's viral load--decreases the chance of HIV transmission. These data informed the development of several recent clinical trials to determine the impact on HIV transmission that ARVs can have on both individual and population levels. One recently published trial, HIV Prevention Trials Network (HPTN) 052, reported the HIV transmission rates between serodiscordant partnerships randomized to have the HIV-positive partner start ART immediately or wait until their CD4 count fell below 250 cells/mm3. In this trial, authors report a 96 percent reduction in the risk of genetically linked (i.e., approximately 18 percent of new infections came from outside the partnership [see Eshelman et al. 2011]) HIV transmission among those who started treatment immediately. The effect was so strong that the trial's data safety and monitoring board recommended that all couples in the study be offered early treatment initiation.
III. Core Programmatic Components
The expanded use of ART in persons living with HIV as a tool to reduce HIV incidence has exciting potential, as ARVs also have the benefit of reducing morbidity and mortality in persons diagnosed with HIV. However, despite the encouraging results reported by HPTN 052, additional research and debate around policy and community implications are needed to identify the best approaches to implementation. To truly maximize the prevention benefit from HIV treatment, programs must ensure that:
1. People are tested for HIV and successfully linked to care and treatment programs
2. HIV-diagnosed individuals agree to start ART
3. ARV drugs are available and affordable
4. Patients started on ART are retained in care and remain adherent to treatment
5. Patients on ART do not engage in more risky behavior as a result of starting treatment.
While there have been impressive strides in expanding access to HIV treatment worldwide, improving the tolerability and acceptability of ARV drugs, and supporting ongoing patient retention and adherence, there still remains unmet need in each of these areas. A number of other issues must also be considered as additional data from ongoing trials become available:
- Who are the most appropriate target populations? Who should be prioritized if resources are limited?
- What are the best ways to educate health workers and target populations on the benefits of HIV treatment?
- How can we assure uninterrupted stocks of ARV drugs with increased demand?
- How can we message and counsel most effectively to improve uptake of testing, linkage to care, and lifelong adherence to ART?
IV. Current Status of Implementation Experience
HIV treatment programs have been contributing to HIV prevention efforts since their inception. However, many treatment programs still prioritize treating those with lowest CD4 counts first. As HIV viral load--and risk of transmission--is highest during acute infection and with lower CD4 counts, this is one strategy to maximize the prevention benefit of ART with limited resources. The additional prevention benefit from expanding ART to people living with HIV in discordant relationships and at higher CD4 counts is currently under study. Treatment-as-prevention programs are being evaluated in several ongoing trials around the world--most notably in Botswana, South Africa, Uganda, Malawi, Tanzania, Mozambique, Zambia, and the United States--either separately or as part of combination prevention packages. HPTN 065 in the United States is exploring the strategy in two cities: New York (specifically, the Bronx), NY, and Washington, DC. The HPTN 052 study trial sites in Brazil, India, Malawi, Thailand, the United States, and Zimbabwe are continuing to offer all study couples early initiation of treatment and will follow them until the planned end date of 2015.
This article reports expanded demographic and clinical information from the HIV Prevention Trials Network 052 trial, whose interim findings, released on April 28, revealed a 96 percent reduction in the risk of HIV transmission among participants with a CD4 count between 350 and 550 cells per cubic millimeter (cells/mm³) who received antiretroviral therapy (ART) immediately upon entering the trial. Over half of the participating serodiscordant couples (54 percent, or 954 couples) came from Africa, 531 couples from Asia, and 278 couples from the Americas; 50 percent of the infected partners were men. Enrollees were randomized to receive ART immediately upon testing positive or to delay therapy until their CD4 counts dropped below 250 cells/mm³ or they acquired an AIDS-related illness. Of the total of 39 HIV transmission events, 35 occurred among participants in the delayed arm (with 82 percent occurring among African couples). A total of 61 percent of the 28 "linked" events (in which HIV transmission was directly linked to the infected study partner) occurred among individuals whose partner had a CD4 count greater than 350 cells/mm³, and 64 percent of the HIV transmissions were from female to male partners. Individuals in the immediate treatment arm experienced a 41 percent lower risk of experiencing a clinical event, such as tuberculosis, compared to those in the delayed arm. According to the authors, the most likely mechanism in preventing HIV-1 transmission was sustained suppression of HIV-1 in genital secretions--the result of ART. The authors found that early initiation of ART has clinical benefits for both HIV-1-infected individuals and their uninfected partners, and therefore must be rolled out as a prevention strategy to reduce the spread of HIV-1 infection.
The study genetically links HIV cases of serodiscordant couples where the uninfected partner became HIV-positive in the course of the HIV Prevention Trials Network 052 trial. The 052 trial provided early initiation of antiretroviral therapy to the HIV-positive individual in a serodiscordant couple to determine if treatment had a prevention benefit to the uninfected partner. Thirty-eight couples seroconverted during the course of the trial. Phylogenetic analysis of HIV pol sequences and Bayesian analysis of genetic distances between pol sequences were utilized to determine linkages between the two partners' viral subtype. It was found that out of the 38 seroconversions, 29 (76.3 percent) were linked, 7 (18.4 percent) were not linked, and 2 (5.3 percent) could not be determined. The unlinked cases equates to a large minority acquiring HIV through someone other than the partner in the study. Linked cases were significantly associated with the HIV-positive partner being in the delayed treatment arm and were not receiving treatment at the time of transmission. In the unlinked cases, there was a significant association between having a higher number of sexual partners in the prior three months and HIV seroconversion. The authors underscore the importance of performing genetic linkage testing in similar trials.
The review explored the efficacy of using antiretroviral therapy (ART) as a prevention strategy to reduce HIV transmission in discordant couples in which the person living with HIV was actively taking ART. All randomized controlled trials, case-control, and cohort studies were included. Eight studies were included in the review: one was a randomized controlled trial and seven were observational studies. It was found that the rate ratio for the randomized controlled trail was 0.04 (95 percent confidence interval [CI] 0.00, 0.27), and 0.34 (95 percent CI 0.13, 0.92) for the seven observational studies in which CD4 counts were 350 to 550 cells/mm3. A separate analysis was performed on treated individuals who had a CD4 count of greater than 350 to determine if the rate of HIV transmission was affected. It was found that the rate ratio was 0.02 (95 percent CI 0.00, 2.87). The studies in the review confirmed the benefit of treating the HIV-positive person in discordant couples with ART to reduce HIV transmission to the uninfected partner. The authors note that implementation challenges remain, as do unanswered research questions such as the effect of long-term treatment, sustaining adherence, and drug resistance.
The two-page article heralds the potential of antiretrovirals (ARVs) being utilized as an HIV prevention measure in a combination approach. The author cautions against wide-scale implementation of ARVs as an HIV prevention method until challenges and research questions can be adequately addressed. The challenges include identifying individuals most in need, missing individuals in the acute phase of disease, increasing adherence, reducing drug resistance and toxicity, and limiting risk compensation. Current ARV programs have difficulty meeting the existing needs of populations, and to expand ARV as a prevention approach may place untold stress on already existing programs. ARV programs should prioritize those who are positive but are untreated, HIV-positive pregnant mothers, as well as those at higher risk of acquisition and transmission of HIV such as sex workers and people who inject drugs. There are limited funds for HIV programs; therefore, an incremental and balanced prevention approach should be implemented.
The paper outlines the current research findings in using antiretroviral therapy (ART) as an HIV prevention measure. They explore past studies and how science has evolved to the present climate. There are many benefits and positive results based on research in using ART for prevention such as in prevention of mother-to-child transmission, for viral suppression in infected individuals to lead to decreased transmission, and as a pre- or post-exposure prophylaxis. The authors also provide an overview of evidence on pre- and post-exposure prophylaxis. Unanswered questions remain such as how effective ART is in suppressing HIV in the genital tract, the possibility of increasing drug resistance in populations, and operational issues such as costs, targeting higher-risk individuals, adherence, and long-term effects of use. The authors view ART as one of many prevention strategies in mitigating the HIV epidemic.
The paper reviews the antiretroviral therapy (ART) as prevention for HIV and tuberculosis (TB) research literature to provide information for policies and future planning initiatives. The search included all research articles and current studies that evaluated or were evaluating how ART was used in prevention programs to impact HIV and TB morbidity, mortality, risk behaviors, and HIV incidence and transmission. A total of 50 studies covering 52 different countries were included in the review and geographical areas included North America (20), Africa (22), Asia (4), Europe (1), and studies having multiple country sites (3). Twenty-four studies were randomized controlled trials. It was found that there is considerable evidence to support ART as a prevention method. There are a number of current research studies being conducted as well as being planned in this area. The methodology, interventions, and geographical location across the studies vary greatly. The review found strong support for ART as a prevention method, but many unanswered questions remain such as when to start ART, the overall impact of ART on HIV and TB prevention at the population level, and how to improve coverage of ART.
The results of the HIV Prevention Trials Network 052 study were used in mathematical modeling to predict the effect of treating discordant couples with antiretroviral therapy (ART) to prevent transmission of HIV to reduce overall incidence levels. The number of infections prevented in relation to ART coverage levels was also explored. Country-level data on HIV prevalence, population size, and discordancy rates from Lesotho, Malawi, Rwanda, and Ghana were used in the model. It was found that reductions in annual incidence rates were highest in Lesotho, moderate in Malawi, and low in Rwanda and Ghana. The annual number of infections averted was highest in Malawi, high in Ghana, and low in Lesotho and Rwanda. In the model, the effectiveness of ART for prevention interventions in discordant couples in reducing annual incidence rates and number of infections averted was the greatest in Malawi. Ghana had a large number of infections averted, but the change on incidence was low. The annual number of infections prevented was low in Lesotho and Rwanda, but reductions in annual incidence rates were higher in Lesotho than Rwanda. Overall, treatment had a greater effect in reducing annual incidence rates in populations where there were a higher proportion of stable partnerships. The model displays the complexity in gauging which country would be most able to reduce annual infection and incidence rates based on expanding coverage of ART as a prevention method among discordant couples.
This article reviews the evidence supporting antiretroviral therapy (ART) as an HIV prevention strategy while critically examining the public health implications of such a strategy. Data clearly show that current ART suppresses but does not eliminate shedding of HIV in genital secretions. As a result, the magnitude of HIV transmission during ART is impossible to estimate. Current evidence for using ART as prevention includes retrospective and prospective observational studies involving couples, ecologic community studies, and an ongoing randomized trial. Such studies have data collection limitations and several kinds of bias. Furthermore, studies assessing the impact of ART on sexual behavior find both increases and decreases in risky sexual behaviors. While mathematical models using optimistic inputs suggest this strategy can eradicate HIV, others conclude that this is not the case. Use of ART for prevention will require much wider use of drugs in much healthier individuals. Clinical and community randomized trials (CRTs) are needed to determine the feasibility and benefit of this approach.
Data from San Francisco's comprehensive HIV surveillance system--which includes mandatory laboratory reporting of viral loads--formed the basis for comparing community viral loads (CVL) in San Francisco to incident HIV infections from 2002 to 2008. Mean CVL and total CVL were used to examine HIV infection trends. Both mean and total CVL decreased from 2002 to 2008. Decreases in new diagnoses of HIV cases and HIV incidence took place shortly thereafter. The trend in mean CVL was significantly associated with newly diagnosed HIV cases from 2004 to 2008 and HIV incidence from 2006 to 2007. Total CVL was also associated with newly diagnosed HIV cases and HIV incidence. Increased ART options and coverage may have led to these decreases in CVL in San Francisco and subsequent decreases in HIV diagnoses and new HIV infections. These findings support the hypothesis that wide-scale, early ART can have a preventive effect on HIV transmission at the population level.
This study broadly confirms that treatment has the potential to substantially reduce HIV transmission. However, how the treatment is best implemented and to what degree HIV infection is reduced depends on the epidemiological context. For instance, over 30 years, in populations with little variation in risk behavior and random mixing, HIV incidence can fall by 95 percent if 80 percent of the population is tested every three to four years. In other populations, testing once a year is most effective. Because ART as a treatment strategy depends crucially on the epidemiological context, it is likely that this approach is better suited to some populations than others. Although implementation of a test-and-treat strategy could reduce HIV incidence, failing to achieve sufficiently high coverage levels or failing to test frequently enough could lead to a dramatic spiraling of treatment costs. Furthermore, this model finds that the intervention does not interrupt HIV transmission, so the pool of those needing treatment would continue to grow. Failing to fully implement the test-and-treat strategy could lead to overall increased long-term ART costs.
There is growing evidence that initiation of ART for HIV affects the rate of transmission by reducing HIV1 ribonucleic acid (RNA) concentrations (viral load) to zero or undetectable levels. This article reports on the observation of a cohort of 3,381 serodiscordant couples across seven countries in Southern Africa. Participants who initiated ART during the two-year study period (n = 349) had a 92 percent reduction in the rate of transmission to a noninfected partner compared to those who had not started treatment. Among untreated participants, those with higher viral loads were about four times more likely than those with lower viral loads to have a partner that seroconverted, regardless of cluster of differentiation 4 (CD4) cell count. To maximize the preventive value of ARVs and reduce HIV incidence at the population level, the authors recommend better identifying individuals with high viral load as well as initiating treatment at higher CD4 counts than has been standard practice.
This study examines the hypothesis that the provision of highly active ART (HAART) during a 15-year period (1996-2008) in the province of British Columbia, Canada, had a population-level effect on the transmission of HIV by means of reducing the CVL. Researchers looked at the association between newly diagnosed cases of HIV and the expansion of HAART and found that the data signaled three distinct periods: two periods when HAART uptake was high and new cases declined significantly (1996-1999 and 2004-2008), hyphenated by a period of relatively stable treatment levels and numbers of new cases (2000-2003). A number of competing reasons for the strong correlation between rates of treatment and new infections were considered (i.e., decreases in sexually transmitted infections [STI], extent of HIV testing), and none moderated the relationship. The authors conclude that, although cause and effect cannot be established, their model and convergent evidence from other sources strongly suggest that ART is an important factor in mitigating HIV incidence.
Based on the findings of Dodd, Garnett, and Hallett (2010), this editorial discusses how modeling can be useful for "thought experiments." However, to be a useful tool for designing health policy, models must be based on realistic assumptions and include the most recent biomedical data. If assumptions made by Dodd, Garnett, and Hallett are compared to available clinical data on HIV, the reductions in incidence may be overly optimistic: infected individuals will live longer and be more infectious than the model assumes. Furthermore, the authors argue that inclusion of resistance is crucial for any model evaluating a test-and-treat strategy because clinical data indicate that resistance does occur. Lastly, deciding whether the primary purpose of antiretroviral (ARV) medication is therapeutic or for prevention purposes has significant implications for determining who gets treated, leading to an ethical dilemma: If the primary purpose of ARVs is prevention, Dodd, Garnett, and Hallett argue that behavioral "core" groups should be prioritized to receive treatment. If the primary purpose of ARVs is therapeutic, however, then the sickest should receive treatment.
The authors of this commentary assert that the assumptions used in the Granich et al. (2009) model require validation. Retrospective analyses of existing data and designing and executing new studies are needed to inform the model's inputs. Research gaps include the following: feasibility of universal testing, the relationship of the stage of HIV infection to the efficiency of transmission, the effectiveness of ART in preventing HIV transmission, drug resistance, behavioral disinhibition with ART-as-prevention, individual benefits of such treatment, and cost-effectiveness to society. The authors conclude that the mathematical model sets forth a testable strategy that potentially could curtail the global HIV pandemic. Evaluating the influence of each variable in this model can help prioritize the questions that must be answered to provide data needed to ultimately validate or refute this approach.
This commentary focuses on the public health and ethical issues related to the Granich et al. (2009) article on universal testing and treatment for HIV prevention. They suggest that a "treat early and treat hard" strategy is not likely to eradicate HIV, but can likely reduce new infections dramatically. Such a strategy, however, is based on large-scale public health benefits rather than the benefits to an individual patient. The authors suggest that such a strategy reflects public health at its best and its worst. At its best, it prevents morbidity and mortality for the population, both through better treatment of the individual and reduced spread of HIV. At its worst, the strategy will involve overtesting, overtreatment, side effects, resistance, and potentially reduced autonomy of the individual in their choices of care.
Current clinical guidelines recommend deferring initiation of ART until patients living with HIV reach certain immunologically or clinically defined stages of disease. The potential effect of ART on transmission of HIV is limited by the deferral of ART until later stages of disease, low coverage in many populations, and high viral loads during acute infection when individuals do not know they are infected. Moreover, the majority of individuals living with HIV in sub-Saharan Africa do not know they are infected and thus cannot receive ART. This mathematical modeling study attempted to define the optimal testing and treatment strategy if ART was to be used with the aim of eliminating HIV transmission. A strategy of yearly voluntary testing of all persons over 15 years of age combined with immediate ART initiation at the time of diagnosis could substantially decrease HIV transmission in sub-Saharan Africa and drive the epidemic into an elimination phase by 2020. Although the up-front costs would be substantial, cost-savings would be realized by 2050.
People working in HIV prevention worldwide wrote letters in response to Granich et al.'s (2009) modeling exercise. Comments about the model include the following: the need to include and cost out programs to reduce stigma and other barriers to counseling and testing (CT); the need to include sensitivity analysis of the assumptions to see how changes in the inputs affect the outcomes; the ethical considerations of shifting the treatment focus away from unhealthy individuals to healthy populations; the reality of trying to provide "universal" treatment in severely resource-poor countries; and more.
This commentary about Granich et al.'s (2009) findings suggests that researchers are "clutching at straws" when they consider whether providing treatment with ART to everyone with HIV can help eliminate the epidemic. In addition to some flaws in the assumptions used in the model, the author points to resource-rich countries where HIV infections began to rise among certain populations as ART became more widespread. The author asks: If more people are living with HIV and these people are having unprotected sex, that causes more HIV infections in places where the population is well informed, HIV testing is actively promoted, and treatment is free and universally available; how can researchers realistically expect a different outcome in the African context?
In this analysis, the authors address each of the assumptions made in Granich et al.'s (2009) model. When they adjusted the assumptions to make them more realistic, their model concludes that a voluntary test-and-treat approach with ART is not likely to eliminate HIV. Based on their own assumptions, which they deem optimistic and unlikely, the researchers estimate it would take at least 70 years for HIV elimination to take place. More realistic assumptions (65 percent treatment rate among symptomatic individuals who cannot subsequently infect others with HIV) result in 34 percent HIV prevalence rate and 2 percent incidence annually. Treatment can act as an ancillary prevention tool, but the authors argue that ART should, first and foremost, remain as a therapeutic tool. The authors suggest that models be examined carefully, and their results interpreted with caution before being used as health policy tools.
This presentation estimates of the net cost and cost-effectiveness of ART as a prevention strategy in sub-Saharan Africa. Cost data were obtained from current country health systems and services, and the models included HIV epidemics in different countries. Interventions modeled include current ART guidelines based on CD4 cell counts and universal ART. Using these data, the researchers estimated outcomes on HIV trends, deaths, disability-adjusted life years (DALY), cost, and cost-effectiveness of ART as a prevention strategy from 2010 to 2050. This model found that expanded ART for prevention yields considerable benefits for control of the HIV epidemic and health in South Africa and Kenya. Expanded ART in South Africa could sharply reduce the cost burden of HIV, with a break-even point occurring in 2024. In Kenya, universal ART has a lower cost per DALY averted when providing ART for those with CD4 cell counts less than 200, but did not reduce costs over the 40-year model due to lower averted hospital costs.
Current prevention efforts are unlikely to rapidly reduce HIV incidence in countries with severe generalized HIV epidemics. As such, evaluating the role of ART for HIV prevention in these settings has become a priority. A mathematical model developed by WHO scientists suggests that universal HIV CT with immediate ART treatment for all those with HIV--regardless of clinical stage or CD4 cell count--can reduce HIV incidence by 95 percent over 10 years. Among other objectives, this meeting reviewed the evidence and experience on ART for HIV prevention and identified implications of such an approach.
Globally, an estimated 33 million people live with HIV, with nearly 3 million new infections occurring in 2007. Efforts to provide ART have reached over 4 million people in low- and middle-income countries in 2008, but many remain untreated. Treatment efforts have not been able to keep pace with infection and are not likely to unless a dramatic reduction in new HIV infections takes place. While current prevention approaches may reduce the number of new HIV infections, they are unlikely to eliminate the disease in settings with high prevalence. Thus, evaluating the role of ART for HIV prevention has emerged as a pressing issue. ART lowers the concentration of HIV in the bloodstream and in genital secretions. Because viral load is the single greatest risk factor for all modes of HIV transmission, ART use reduces the risk that HIV will be transmitted from one person to another. Currently, however, there is not enough evidence for the WHO to develop a policy or clinical guidance on the role of ART in HIV prevention.
Only about half of the eligible patients living with HIV in British Columbia, Canada, are on ART despite universal access to health care and ART. This mathematical modeling study examines the potential effect of expanding ART coverage among those eligible for initiation (based on CD4 cell count guidelines less than or equal to 200 cells/mm3 at the time of the study) over the next 25 years. The model includes the following: varying rates of emergence of resistance; a degree of adherence to ART; a shift in treatment guidelines (to CD4 cell count less than or equal to 350 cells/mm3); and a degree of ART coverage. Expansion of ART to cover 75 percent of the eligible population led to substantial decreases in both new HIV infections and total lifetime treatment costs. The potential number of averted HIV cases after 25 years was estimated to be 3,108 (29 percent of new positive test results), 4,776 (44 percent), and 5,710 (53 percent) with coverage rate increases to 75 percent, 95 percent, and 100 percent, respectively. This study indicated that expansion of ART programs in Canada to cover more of the treatment-eligible population could have a substantial impact on the population-level incidence of HIV infection.
In 2008, the Swiss Federal Commission for HIV/AIDS released a document stating that people living with HIV with undetectable plasma HIV RNA levels, who are on HAART, and have no other genital infections are not sexually infectious. This statement generated international controversy and concern about increases in risky sexual behaviors in response to the statement. This simulation study used a mathematical model to estimate the risk of HIV transmission from people living with HIV with undetectable viral loads over many sexual exposures. A key assumption of this model was that a person living with HIV could never be completely noninfectious, no matter how low his or her viral load. The model suggested that if condoms were completely abandoned by people with undetectable viral loads and who are on ART, the risk of HIV transmission would accumulate over time, and the population's HIV incidence could increase four-fold.
This systematic literature review summarizes the data on ART in preventing HIV transmission. The concentration of ART in genital secretions may be important in decreasing the infectiousness of people living with HIV on ART. Protease inhibitors have a limited ability to concentrate in the genital tract because they are highly protein-bound in the blood. Nucleoside and nucleotide reverse transcriptase inhibitors achieve much higher levels in the genital tract. Lamivudine, emtricitabine, and tenofovir achieve levels in the genital tract over four times those in the blood.
This model assesses how ART can work as a treatment strategy using "best case" and "worst case" scenarios. Scenarios include treating all those infected and only those with AIDS. If only AIDS patients were treated, HIV prevalence would increase because of this population's increase in life expectancy. If both AIDS and pre-AIDS cases were treated, the increase in prevalence would be initially counteracted by averting more infections. Over the long-term, however, these gains are negated by an increase in risk-taking behavior coupled with an increase in life expectancy. Treating both AIDS and pre-AIDS patients saves fewer life-years per year of treatment than only treating AIDS patients. Furthermore, the higher the ARV coverage, the more viral resistance occurs and the wider it spreads. The authors conclude that HIV epidemics in sub-Saharan Africa cannot be controlled through treatment, regardless to what degree ART is available. Thus, any ART strategies must be integrated with prevention efforts.
This study involved 393 heterosexual couples in Spain. Index partners living with HIV were enrolled in the study between 1991 and 2003. The non-index partners--whose only known HIV risk was sexual contact with the index partners--were tested for HIV at study entry. Because the study spanned the era before and after the introduction and uptake of HAART, the investigators could evaluate the relationship between the availability of HAART and the prevalence of HIV in the non-index partners. In the pre-HAART era (1991-1995), 10.3 percent of the non-index partners tested positive for HIV, compared to only 1.9 percent of partners in the late HAART era (1999-2003). The rates of protected sex increased and the rates of sexually transmitted coinfections decreased during this time, but the availability of HAART was associated with an 80 percent decrease in heterosexual transmission even after accounting for those factors. Among 313 partners of subjects who were not taking HAART, 27 (8.6 percent) tested positive for HIV, whereas none of the 60 partners of subjects who were taking HAART tested positive for HIV. This study showed that the introduction of HAART into a population can be associated with less HIV transmission in heterosexual couples.
To help inform assumptions being used in mathematical models, this cross-sectional study randomly sampled the blood and urine of 1,000 people of reproductive age in South Africa. Researchers also obtained a questionnaire about participants' sexual partnerships and condom use. HIV prevalence was found to be 21.8 percent. Using WHO guidelines for initiating ART, 9.5 percent of the subjects living with HIV (equaling 2.1 percent of people of reproductive age) needed to start ART immediately. Such treatment, they calculate, would reduce the risk of HIV infection annually by 11.9 percent. The U.S. Department of Health and Human Services has different guidelines on initiating ART. Using these guidelines on the same population, 56.3 percent of participants living with HIV would require immediate ART, resulting in an annual HIV transmission risk reduction of 71.8 percent. Thus, under WHO guidelines, HAART will not substantially reduce heterosexual spread of HIV in this population. The authors conclude that funding for treatment should not compete with funding for prevention.
According to the authors, as treatment as prevention strategies are scaled up, one of the most important threats to this population-level prevention strategy may be antiretroviral-resistant HIV. Based on results from this study, they find that interruptions of antiretroviral therapy (ART) of 48 hours or more may cause the emergence of resistant strains in female genital tract secretions. They stress that adherence to ART was a key determinant of genital tract resistance among the 102 female participants. During a 12-month period following initiation of non-nucleoside reverse transcriptase inhibitor (NNRTI) medication, drug-resistant virus was detected in the genital tracts of five women. The authors believe that this number could rise if the women who left the study due to withdrawal and/or loss to follow-up were also measured for antiretroviral resistance in the genital tract. Both unavoidable interruptions due to drug toxicity or systemic illness and avoidable interruptions due to late refills led to increased interruptions on average of four days or more. They stress that structural barriers to adherence, including transportation difficulties and pharmacy stock-outs, may also contribute to treatment interruptions and an increased risk of transmission of drug-resistant HIV-1 virus. They call on governments and program managers to increase efforts to prevent treatment interruptions by improving program effectiveness, promoting consistent and timely adherence and refills, and using less toxic ART.
The current findings on the impact of antiretroviral therapy (ART) on preventing HIV are analyzed and the benefits and implementation challenges are discussed in this article. The authors state that ART as a prevention method will be key to future programs, but more needs to be understood before it is universally rolled out. For example, more research needs to be conducted exploring how the male and female genital tracts and rectal mucosal tissue are affected by ART. HIV-1 replication continues in these areas despite viral suppression in the blood, which affects transmission. In addition, drug resistance is another concern if more individuals are to be on ART. Observational studies in serodiscordant couples, community-level ecological studies, and mathematical modeling demonstrate solid findings in favor of ART as a prevention method, but limitations still need to be addressed. As programs are implemented, questions pertaining to how to increase HIV and CD4 testing to identify those most in need, as well providing optimal treatment during the early stages of HIV, will need to be clarified.
A review of existing literature was conducted to explore the attrition rates in the continuum of care between when individuals test positive for HIV to linking to care and treatment services. People living with HIV (PLHIV) are lost at various points along this continuum and understanding the reasons behind this loss is vital in improving health outcomes of PLHIV. The systematic review included peer-reviewed literature as well as conference abstracts that studied patient retention between HIV testing and antiretroviral therapy (ART) initiation in sub-Saharan Africa. The authors defined three different stages: the first being from HIV testing to obtaining CD4 count results, the second from not eligible for ART to accessing pre-ART care until eligible, the third is ART eligible to initiating ART. Twenty-eight articles and abstracts were included in the review: twenty studied only one stage, six addressed two stages, and two addressed all three stages. The median proportion of patients retained in stage 1 was 59 percent (35 to 88 percent), stage 2 was 46 percent (31 to 95 percent), and stage three was 68 percent (14 to 86 percent). Therefore, based on this review, there are great challenges in following healthy PLHIV to ensure initiation to care and treatment services that could span a long period of time. A limitation to the review was the wide variation in how studies defined and measured entry points to care and treatment. It is recommended that researchers standardize language to allow the public health community to generalize across studies.
The review explores the link between antiretroviral therapy (ART) initiation and risky sexual behaviors. Eight cross-sectional and nine observational cohort studies were included in the review, all of which were conducted in developing countries. All of the studies, except one, found that there was a decrease in sexual risk-taking after ART initiation. The authors note that the majority of patients on ART decrease their reported risky sexual behaviors after treatment, but a sizable minority still engage in risky behaviors. The patients that engage in risky behaviors have common characteristics, including their type of partner (causal, spouse, seroconcordant), desire for children, psychosocial factors (depression, drug use), and higher-risk populations (men who have sex with men, people who inject drugs), which need to be addressed in behavior change programs within the treatment setting. A limitation in these studies is that patients on treatment have access to resources that others may not such as risk-reduction counseling, support groups, and ongoing care, which encourages them to make more healthful sexual decisions. Using ART as a prevention method is a viable option, but limitations and challenges should not be ignored.
South Africa's treatment goal in the national strategic plan is to provide antiretroviral therapy (ART) to at least 80 percent of treatment-eligible clients. It is necessary to understand the challenges in expanding treatment to those who are eligible, including those who refuse treatment, to reach this goal. Included in the study were adults who tested positive for HIV at the Perinatal HIV Research Unit in Soweto and were eligible for treatment based on CD4 counts. There were 7,287 clients who presented for voluntary counseling and testing, and 2,562 clients (35 percent) tested positive. Of those who tested positive, 743 (29 percent) were eligible to start treatment but 148 (20 percent) refused. It was found through multiple logistic regression that those who refused were more likely to be single and to have active tuberculosis compared to those who accepted treatment. Social workers recorded the reasons for refusal and the most sited "feeling healthy" followed by unable to disclose status, drug side effects, unable to adhere to drugs, cultural beliefs, and stigma. Increasing voluntary counseling and testing, drug availability, and reducing cost of drugs are not the only factors in expanding treatment. Promoting ART as a life-saving and safe method will be necessary to encourage all eligible individuals to accept treatment.
This news article discusses a new initiative being piloted in Canada's British Columbia province called "Seek and Treat." Using health workers working on the streets, this ambitious project will focus on reaching populations that have been traditionally difficult to reach, such as sex workers and injecting drug users. The 48-million dollar project will provide these populations with HAART.
With no HIV prevention interventions producing dramatic declines to the HIV epidemic in sub-Saharan Africa, the authors wonder whether Granich et al.'s (2009) test-and-treat theoretic model provides a "ray of hope" in curbing the epidemic in this part of the world. As such, if proof of concept and validation studies find that this model is indeed plausible, the authors present four key operational challenges to universal testing and treatment in sub-Saharan Africa, and possible ways to address them. These include 1) community acceptability and protection of human rights--that is, weighing public health benefits versus health benefits to the individual; 2) availability of safe, effective, first-line ARV regimen for asymptomatic patients; 3) delivering ARV on a mass scale in sub-Saharan Africa; and 4) ensuring a viable, accurate monitoring and reporting system. Their insights include discussion of conceptual, financial, ethical, and programmatic issues related to a universal test-and-treat strategy.
This presentation includes data from an ongoing clinical trial in Côte d'Ivoire that randomizes participants to ART before they reach the point when the WHO recommends initiation. Data collected from the study were then used to inform a mathematical model. The model concluded that combining annual HIV screening with providing ART when diagnosed with HIV would increase survival and decrease morbidity. However, it could also decrease secondary transmission in the short-term while increasing secondary transmission in the long-term. For short-term decreases in secondary transmission to be maintained over time, decreases in the rate of ART failure and HIV transmission risk must take place.
This presentation outlines the PopART pilot study, which will assess whether a CRT of universal testing and treatment is feasible, practical, acceptable, and potentially effective in drastically reducing HIV transmission. Taking place in Uganda and Zambia from 2010-2013, this study will collect information on acceptability; refine and explore mathematical models that can inform the design of the CRT; undertake qualitative research on trust, barriers, and experiences in service delivery; and assess five components related to human rights and ethics. The study will also pilot test 100 individuals who are living with HIV with immediate treatment with ART. If results indicate that the intervention is safe, acceptable, and has a greater than 50 percent effect on HIV incidence, then the researchers will begin a large-scale CRT using a universal test-and-treat strategy.
This presentation summarizes the research questions that need to be answered to understand whether a universal test-and-treat strategy for preventing HIV transmission is acceptable, feasible, effective, safe, and cost-effective. The various issues are summarized under the following categories: HIV testing strategies, HIV treatment strategies, health service issues, effects of universal test-and-treat strategy on HIV transmission, and others. Because preliminary studies may find that such a strategy is not feasible or acceptable, more limited approaches to ART for HIV prevention must also be a part of the research agenda.
After reviewing studies that find reduced HIV transmission among people using ART, this presentation reviews ways that studies could show the "proof of concept" for a universal test-and-treat strategy. It includes discussion of using a cluster of individuals (for example, an entire village) as the unit of randomization and two trials to assess whether a test-and-treat strategy can work.
This news article discusses the interest in implementing a test-and-treat strategy for HIV. U.S.. health officials have designed a three-year study in the Bronx borough of New York City and the District of Columbia--places with some of the highest HIV rates in the country--to determine the feasibility of such an intervention. The article highlights some of the challenges in getting people tested for HIV and getting medical care to those who have tested positive for HIV.
Thirty representatives from human rights organizations in seven sub-Saharan African countries met to discuss the Granich et al. (2009) model and the practical application of this model to HIV programming. A two-page statement of opposition to the universal test-and-treat strategy outlined in the Granich et al. paper lays out their concerns, including flawed assumptions, human rights shortcomings, and a lack of attention to marginalized and/or vulnerable groups. They state that the model fails to adequately capture modes of transmission outside of heterosexual sex, barriers to testing and treatment among vulnerable groups in particular, and ideological opposition to many proven HIV prevention methods. They state, "The model is fundamentally at odds with a human rights based response to HIV." The group recommends discussions, research, and analysis of a list of key issues before any action is taken to implementing such a universal test-and-treat model.
This study followed a cohort of 926 heterosexual Ugandans living with HIV who were receiving ART to see whether theories of behavioral disinhibition among ART users were borne out. Participants had ART delivered weekly to their homes by lay field officers who referred participants to the study clinic for health care and counseling when needed. Counselors undertook private, home-based interviews with study participants at enrollment and every three months on sexual behavior. Compared to baseline, risky sexual behaviors declined after about six months of ART use, despite participants reporting an increase in sexual desire in the three months prior to being interviewed. The study found a 70 percent decrease in the number of unprotected sex acts with partners of unknown HIV status. Furthermore, the authors calculated a 98 percent reduction in estimated HIV transmission as a result of lowered viral load and changes in sexual behavior. Limitations of the study included self-reported sexual behavior, a majority of study participants being abstinent, a relatively short follow-up period (five to seven months), and that the same counselors providing risk reduction counseling also conducted the interviews with study participants.
This commentary proposes that the "positive prevention" approach to HIV prevention being used in industrialized nations--focusing preventive interventions only on those infected with HIV--be adopted in Africa and other high-prevalence regions. The authors propose 10 approaches for such a program in Africa and prioritize the interventions based on those with the greatest potential to reduce HIV transmission. These include disclosing HIV status to partners, partner testing, providing ART, preventing unintended pregnancies and maternal-to-child transmission of HIV, among others. Some of these approaches are proven and some have just been piloted, while others need exploration and the development of evidence-based policy.
This two-page editorial introduces nine articles in this issue that discuss possibilities and concerns surrounding the test-and-treat method. Cohen et al. start the series with a review of the HPTN 052 trial. Dieffenbach presents an implementation research agenda. The next papers examine ways of measuring viral levels and detecting acute infection, and the implications of antiretroviral therapy (ART) for prevention. The last article explores questions about the actual implementation of the test-and-treat strategy. The series provides the latest thinking and science on this emerging HIV prevention strategy.
According to the author, HIV testing must become the entry point for all individually focused HIV combination strategies. He advocates focusing on the psychosocial and structural drivers that impede HIV testing, such as lack of access to care if the diagnosis is positive. He identifies linkage and retention in care as imperative to move treatment as prevention from proof of concept in a clinical trial to a public health intervention. He points out that linkage to care continues to remain a challenge in many settings; for example, 50 percent of HIV-positive people in the United States are currently not receiving care services, including antiretroviral therapy (ART). He stresses that easy-to-use point-of-care testing kits administered by peer counselors may provide access to those hard-to-reach populations that may not otherwise seek care. He also stresses that linkage to care is not enough and that there must be a focus on retention, particularly following initiation of ART treatment. According to the author, other issues related to treatment as prevention include the effect of the route of exposure to HIV on the efficacy of ART as prevention, the most effective way to initiate early treatment, and a possible role for therapeutic vaccines and a functional cure. Other issues include how to sustain the prevention effect and how to improve viral load testing and HIV therapy. He also advocates for increasing implementation research on the factors that impede access to testing and care services.
The authors of this article sought to measure the health system cost, health benefit, and cost-effectiveness of antiretroviral therapy (ART) for HIV prevention programs in South Africa if ART were more readily available to those in need. They modeled four ART options: 1) the current practice of providing ART for individuals who have CD4 counts of <200 cells/mm3; 2) starting ART for those with CD4 counts of <350 cells/mm3, the current WHO recommendation, which South Africa started implementing in 2010; 3) starting art at CD4 counts of <500 cells/mm3; and 4) universal access to ART. The inputs include information on the epidemic, health care, and costs, and the outcome measured were new HIV infections, deaths, disability adjusted life years (DALYs), and cost. The model will included costs from a best practice program that includes HIV counseling and testing, ART provision, and other support. It was found that all of the enhanced models were beneficial in terms of reducing HIV incidence and prevalence, decreasing deaths, increasing lives saved, and decreasing DALYs and costs by the forty-year end point. The rates of change for each outcome improve based on the increased availability of ART. Costs for the enhanced scenarios are almost all the same, or are at a reduced cost, by 2015. The authors demonstrate through their economic models that increasing investments in ART will provide long-term benefits to both the people of South Africa and the health care system.
The research presented in this article compared the costs and impact for South Africa of providing universal access to HIV treatment--antiretroviral therapy (ART) for those who test positive for HIV and have a CD4 count under a certain level--versus a "test and treat" (T&T) strategy where all individuals are tested in a certain area, and all who test positive receive treatment, regardless of CD4 count. This study followed an earlier World Health Organization (WHO) study that found that universal T&T would eliminate HIV in South Africa within 10 years and cost approximately U.S.$10 billion less than universal access to treatment over 40 years. Using the same mathematical model as the WHO study but with additional variables to reflect epidemiological realities, the authors got very different results: universal T&T would take 40 years to eliminate HIV in South Africa and cost about U.S.$12 billion more than universal access. The authors attribute the difference between these two sets of results to limitations in the earlier WHO mathematical model, which underestimated survival length and did not account for the development of resistance to HIV drugs. The authors predict that, after 20 years of universal T&T implementation, about 2 million individuals in South Africa would require expensive second-line ART, compared to about 1.5 million for a universal access strategy.
According to the authors, HIV-related risk behaviors, treatment access, and treatment success ultimately affect the outcomes of treatment as prevention (TASP). Before fully implementing TASP, the authors recommend that behavioral issues be examined at each stage within the cascade of care: awareness of serostatus, linkage to care, retention in care, access to antiretroviral therapy, adherence to therapy, and successful suppression of viral load. To identify gaps, they encourage prioritizing who will receive TASP, identifying some of the unmet needs for TASP, and measuring the correlation between HIV-related risk behaviors and TASP. According to the authors, TASP is most beneficial for people living with HIV who are aware of their serostatus and are engaged in risky behavior. The authors encourage public health and clinical providers to develop tools to identify clients that engage in either sexual or drug injection transmission risk behaviors, address the social and individual determinants of these risk behaviors, and employ TASP when these behaviors are present. They anticipate that understanding who is most likely to transmit the virus can make TASP more effective at meeting the clinical needs of these individuals and the larger public health imperative of averting new infections. TASP received wide coverage at the International AIDS Conference because of its potential as a "game changer" in the global epidemic. Related research presented at the gathering included designing TASP programming and implementing TASP in countries with high levels of antiretroviral coverage.
View Related Conference Abstracts
This introductory article provides a framework for a series of articles commissioned by the HIV Modelling Consortium on the future research agenda and implementation approaches of treatment as prevention (TASP). According to the authors, the role of antiretroviral therapy in reducing HIV incidence will be one of the most important tools in HIV prevention in the near future. They stress that implementation requires an interdisciplinary approach that involves epidemiology, economics, demography, statistics, biology, and mathematical modeling. Examining 12 mathematical models, the authors report that HIV incidence can be reduced by 35 to 54 percent after eight years if 80 percent of people living with HIV were treated when their CD4 count is 350 cells/µl. They report that once more years are added to these models, it becomes difficult to pinpoint which of many sexual risk behaviors lead to transmission. The series also examines how early infections before diagnosis--when viral concentration spikes and individuals are more infectious--may or may not lessen the impact of TASP programs and modeling estimates. Other topics included in this series include costs associated with TASP, the future direction and priorities of modeling, and best practices for model presentation and interpretation. A related presentation at the International AIDS Conference discussed the use of modeling to show efficacy of TASP for the South African epidemic.
Taiwan implemented comprehensive surveillance of HIV cases in 1989 and a program of universal free access to HAART in 1997. The government encouraged early, intensive treatment of HIV infection, with a recommendation to withhold HAART only from patients who had a HIV RNA viral load of less than 5000 copies/mL and a CD4 cell count in the normal range. Using data from surveillance and statistical modeling to estimate the transmission rate of HIV in the population, the investigators examined the effect of providing free, universal HAART on HIV transmission. The prevalence of HIV in Taiwan at the end of 2002 was 0.019 percent of the total population. The HIV transmission rate decreased 53 percent after the introduction of the HAART program. This change occurred without any change in syphilis rates during the same period, suggesting that HAART introduction was not associated with a change in sexual risk behavior. This study suggested that widespread use of HAART could be an important tool for controlling the HIV epidemic, especially in countries with low prevalence of HIV infection.
The U.S. President's Emergency Plan for AIDS Relief (PEPFAR) guidance is aimed at PEPFAR country teams to provide them with the latest scientific data on prevention programs to increase the impact of their country portfolios. Prevention programs should be tailored to the epidemiological and social context of the country, as well as optimization of current programs implemented by other partners to help fill gaps. The guidance highlights the importance of knowing the country's epidemic, context, response, and costs as well as ensuring that HIV prevention is part of the overall country continuum for the response. It also outlines current evidence and program activities that PEPFAR funds will support. Post-exposure prophylaxis, treatment as prevention, and pre-exposure prophylaxis are contained in the antiretroviral for prevention section. The evidence section outlines the findings from past observational studies that demonstrated the reduction of HIV transmission from the treated HIV-positive partner to the negative partner. These past observational studies were confirmed by the results of the randomized controlled trail, HIV Prevention Trial Network 052, in July 2011. PEPFAR recommends that country programs identify discordant couples as soon as possible to link them to prevention, care, and treatment services. Programs should consider the early treatment of people living with HIV whose CD4 count is 350 cells/μL or lower who are in discordant couples. Additional Office of the U.S. Global AIDS Coordinator guidance will be forthcoming when World Health Organization guidance is available.
This document briefly reviews seminal articles related to two emerging research areas: community-based HIV testing and treatment as prevention. The author presents strategies and evidence for voluntary community-based HIV testing and provider-initiated CT. The key knowledge gaps and research needs for the test-and-treat strategy are also reviewed. The author welcomes comments on the test-and-treat approach, particularly from individuals living with HIV.
This manual was developed to help fill the growing need for family planning (FP) and reproductive health services among people on ARVs. The manual provides different levels of integrating on-site FP information, counseling, and services into HIV services. It also discusses the role of facility-based and community outreach personnel for integrated services. The goal of the training manual is to provide personnel with "knowledge and skills, and to promote the accepting attitudes needed to provide FP as a component of comprehensive HIV care and treatment."
According to the World Health Organization (WHO), treatment as prevention (TasP) should be implemented as a key component of combination HIV prevention. The WHO June 2012 Antiretroviral Treatment as Prevention (TasP) of HIV and TB Programmatic Update recommends that as countries scale up treatment to all who are eligible, the focus should be on key populations such as serodiscordant couples and pregnant women, where the prevention impact may be greatest. The document emphasizes the benefits of increasing access to antiretroviral therapy (e.g., lowering tuberculosis infections by 63 percent), but cautions that putting aside resources solely for TasP efforts may still be difficult for many resource-limited countries. It also notes that communities must be intricately involved in the planning of TasP, particularly in deciding who will get access to it and how it will be made available. Along with the release of this guidance, WHO plans to release consolidated guidance in mid-2013 on TasP for a variety of populations.
Debate Six: Treatment as Prevention
World Bank & U.S. Agency for International Development. (2012).
This report outlines the sixth in the Emerging Issues in Today's HIV Response debate series sponsored by the World Bank and the U.S. Agency for International Development. In November 2011, a panel of four renowned public health experts--two each assigned to argue for and against--addressed the proposition: countries should spend a majority of what is likely to be a flat or even declining HIV prevention budget on treatment as prevention. The topic was inspired by the stunning findings of the recent HIV Prevention Trial Network (HPTN) 052 clinical trial of serodiscordant couples showing that early treatment reduces the risk of HIV transmission to an uninfected partner by at least 96 percent. Among other arguments, the two panelists in favor said that because of the overwhelming evidence for the effectiveness of treatment as prevention, ethical principles require that this strategy be implemented quickly as part of comprehensive HIV prevention and care programming. The two panelists in opposition countered that, despite HPTN 052's powerful results, a blanket recommendation to launch treatment as prevention is premature, because there are still too many unanswered questions about its long-term effects in different populations. They also argued that each country's response to HIV should not be driven by a single programming imperative but instead reflect the specific characteristics and prevention needs of the local epidemic. More than 800 people worldwide registered to attend the debate either in person in Washington, via the Internet, or by video conferencing.
The Impact of ART on HIV and TB: South African Centre for Epidemiological Modelling and Analysis
Williams, B. (2009).
This complex, technical presentation shows the formulae and models used for assessing the impact of ART on HIV and tuberculosis (TB). It includes models and studies looking at the following various outcomes: increases in TB with declines in CD4 cell counts, decreases in disease duration as incidence increased, reduction in TB when on ART, and predicted versus observed CD4 cell counts. The presentation ends with research needs, such as models that include age, gender, and better data to inform the assumptions.
Informal Working Group Meeting: Modelling the Impact of ART on TB and HIV
These notes from the informal working group meeting discussing modeling simulations provide a flavor of the presentations and discussion taking place among participants. As a result of the discussions, recommendations on moving forward were made for designers of trials to provide proof of concept, for considerations for developers of models, and for the WHO as a convening body.
The Case for Expanding Access to Highly Active Antiretroviral Therapy to Curb the Growth of the HIV Epidemic
Montaner, J. S., Hogg, R., Wood, E., et al. The Lancet (2006), Vol. 368 No. 9543, pp. 531-6.
This commentary makes the case that the HIV pandemic presents "an exceptional threat to humanity" and that "similarly exceptional interventions" are needed to reverse the pandemic. Given that prevention efforts are only partially successful and underused, a vaccine is not on the horizon, and current treatment cannot eradicate infection, the authors argue that there is potential for HAART to help stem the HIV epidemic. The authors present a brief theoretical model of their approach.
AVAC, Global Advocacy for HIV Prevention
The website offers the latest news on the past, present, or ongoing status of biomedical HIV prevention research studies. Readers can review summary tables from various HIV prevention clinical trails, search information on prevention trails, see what is new on the site this month, and review the user's guide for help in using the site. The site offers information about the following biomedical prevention trails: Microbicides, pre-exposure prophylaxis (PrEP), Treatment as Prevention, and Vaccines.