CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

1114 Global Fund Cost Projections for ImplementingWHO 2013 Guidelines Obinna Onyekwena ; Ade Fakoya; Michael Johnson; Michael Olszak-Olszewski; Mark Dybul The Global Fund to Fight AIDS, Tuberculosis and Malaria, Geneva, Switzerland

Background: Although recent global cost estimates indicate overall investments needed for implementation of the World Health Organization 2013 consolidated ART guidelines, detailed financial estimates for individual countries are limited. The aim of this study was to estimate additional costs for the transition of Global Fund grants to implement ART eligibility recommendations of the new guidelines. Methods: The thirty-two countries which represent 83% of current Global Fund country allocations for HIV were included in the review. Data on treatment targets, ARV costs, and financing contributions to ARV were extracted from the Global Fund reporting database, grant documents and National AIDS Spending Assessment reports. Global projections for additional numbers of persons eligible for treatment, reported by WHO, were applied to country treatment targets to derive country-level number projections for 2015 to 2017. A weighted average ARV cost was used to determine associated ARV cost projections. Due to inter-program variability, facility and adherence support costs were not included. Treatment numbers and cost projections were disaggregated by new eligibility criteria and compared to current total allocations for HIV. Results: ARV medicine cost for 2014 Global Fund commitments in 32 countries was estimated to be $628 million for 5 million patients on antiretroviral treatment. Additional cost of ARV medicines expected from ART eligibility recommendations was projected to be US$695 million to the end of 2017, for an additional 1.7 million persons on treatment. Costs for implementation of Option B+ and treating all HIV positive children below the age of five were US$53 million and US$106 million respectively; while initiating HIV positive persons with a CD4 count between 350 and 500 cells per mm 3 and in serodiscordant relationships had estimated additional costs of US$294 million and US$242 million respectively. The total ARV medicine cost projections represented approximately 41% (US$2.6 billion) of the total HIV allocations projected from 2015 to 2017 for the 32 countries analysed. Conclusions: ARV medicine scale up costs alone will account for a significant portion of HIV resources allocated by the Global Fund to national HIV programs. This does not take into account additional facility and adherence support costs needed for quality service delivery. Understanding differential cost data in the implementation of treatment guidelines should strengthen strategic investments and portfolio optimisation.

WEDNESDAY, FEBRUARY 25, 2015 Session P-Z2 Poster Session

Poster Hall

2:30 pm– 4:00 pm Modeling HIV Epidemiology

1115 Estimating the Number and Characteristics of Male-Male HIV Transmissions in the USA Eli S. Rosenberg 1 ; Jeremy Grey 1 ; Gabriela Paz-Bailey 2 ; H. Irene Hall 2 ; Amy Lansky 2 ; Jonathan Mermin 2 ; Jacek Skarbinski 2 1 Emory University Rollins School of Public Health, Atlanta, GA, US; 2 Centers for Disease Control and Prevention, Atlanta, GA, US; 3 CDC, Atlanta, GA, US

Background: HIV transmission risk is primarily dependent on sexual and injection drug use behaviors and HIV viral load. Data on behaviors and viral suppression among HIV- infected men who have sex with men (MSM) can be used to estimate transmissions to uninfected male partners. These estimates are important for informing HIV prevention efforts, particularly the balance between the population-impact benefits of antiretroviral therapy (ART) for HIV-infected MSM versus pre-exposure prophylaxis (PrEP) for at-risk HIV-uninfected MSM. Methods: Using weighted respondent-level data on risk behaviors and viral load of HIV-infected persons in the United States from the CDC National HIV Behavioral Surveillance System and Medical Monitoring Project, and population-size estimates from the National HIV Surveillance System, we developed a static, deterministic model to estimate the number of HIV transmissions in 2009 attributable to individuals at each step of the HIV care continuum, by attributes of those individuals and their partners. We estimated transmissions from HIV-infected MSM to male sexual partners, stratified by partner type (main versus casual). Results: An estimated 592,100 HIV-infected MSM had 4,009,405 male partners in 2009, with 24,069 transmissions to 1,132,800 HIV-negative or unknown status anal intercourse (AI) partners. Overall, 78% of transmissions were in main AI partnerships. Per serodiscordant AI partnership, HIV acquisition risk was higher for main than for casual AI partners (9.80 vs 0.57 per 100, respectively). Furthermore, 71% of all transmissions were in main AI partnerships in which the infected partner was not receiving ART, even though they represented only 13% of serodiscordant AI partnerships. In contrast, casual serodiscordant AI partnerships of MSM not receiving ART represented 21% of transmissions and 73% of serodiscordant AI partnerships.

Poster Abstracts

Conclusions: To reduce HIV incidence among MSM, focused efforts are needed to increase the percentage of HIV-infected MSM who are on ART and achieve viral suppression and the percentage of HIV-uninfected MSM in discordant relationships who receive PrEP. We estimate over 1.1 million serodiscordant male-male AI partnerships in the United States in 2009, defining an upper-bound for the number of partnerships involving HIV transmission risk. Targeting the relatively small number of main serodiscordant relationships for ART and PrEP may have a particularly high prevention yield.

647

CROI 2015