CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

1033 Durable Viral Suppression Among HIV-Diagnosed Persons United States, 2012-2013 Nicole Crepaz 1 ;TianTang 2 ; Gary Marks 1 ; Michael J. Mugavero 3 ; Lorena Espinoza 1 ; H. Irene Hall 1 1 CDC, Atlanta, GA, USA; 2 ICF Intl, Atlanta, GA, USA; 3 Univ of Alabama at Birmingham, Birmingham, AL, USA

Background: Viral suppression is associated with reduced morbidity, mortality, and risk of transmitting HIV. Estimating the percentage of persons with HIV who have durably suppressed viral loads (DSVL), plasma HIV burden, and time without viral load (VL) suppression can help in monitoring disease burden and HIV transmission risk potential and guide interventions. Methods: We used data from the National HIV Surveillance System reported from 17 jurisdictions to determine viral burden among persons aged > 13 years who received a diagnosis of HIV infection before 2011 and were alive through 2013. DSVL was defined as all viral loads < 200 copies/mL during 2012-2013. Viremia copy-years were estimated to determine plasma HIV burden. HIV transmission risk potential was estimated by time viral load was above 200 copies/mL over the 2-year period. Results: Of 264,865 persons engaged in care (with at least 1 VL) in 2011, 251,649 (95%) had VL tests between 2012 and 2013. The median number of VL tests during the 2-year period was 5. Of the 251,649 persons, 62% had DSVL. The percentages of persons with DSVL were lower among females (vs. males: 55% & 64%), black/African American (vs. white: 53% & 74%), and persons aged 13-24 years old (vs. > 55 years: 39% & 73%). The geometric mean viremia copy-years in the 2-year period among those without DSVL was higher than the geometric mean averaged across all persons (7,730 vs. 356 copies/mL). Among those without DSVL, the average number of days a person spent above 200 copies/mL was 452 days, corresponding to 62% of the 2-year observation time. Among those without DSVL, female, black/African American, and persons aged 13-24 also had substantially higher geometric means of viremia copy-years and higher numbers of days with VL above 200 copies/mL compared to their respective counterparts (female vs. male: 8,360 & 7,494 copy- years; 478 & 442 days; black/African American vs. white: 9,341 & 5,851 copy-years; 483 & 412 days; aged 13-24 vs. >55 years: 13,971 vs. 3,356 copy-years; 550 & 377 days). Conclusions: It is encouraging that about two-thirds of HIV-diagnosed persons in care had suppressed VL over a 2-year period. The remaining one-third had high plasma burden and spent substantial time without VL suppression, which increases the risk of HIV transmission. Greater disparities in disease burden and transmission risk potential were seen in several subgroups. Targeted care and treatment efforts are needed to address the disparities.

Viral Load Burden, 2012-­‐2013

Persons living with HIV N=251,649

Persons living with HIV without DSVL N=95,186

Percentage with DSVL

62.2%

37.8% 7,730 b

356 a 179 a

Geometric mean viremia copy-­‐years Mean number of days a person spent above 200 copies/mL

452 b

a Due to some missing data, results are based on 251,611 persons living with HIV b Due to some missing data, results are based on 95,148 persons living with HIV

1034 Changes in Viral Load Across US Clinics Over Time Jane Simoni 1 ; Robin Nance 1 ; Joseph A. Delaney 2 ; Ira B.Wilson 3 ; Frances Aunon 1 ; Steven A. Safren 4 ; Michael J. Mugavero 5 ; Kenneth H. Mayer 6 ; Mari M. Kitahata 1 ; Heidi M. Crane 1 1 Univ of Washington, Seattle, WA, USA; 2 Univ of Washington Sch of PH and Community Med, Seattle, WA, USA; 3 Brown Univ Sch of PH, Providence, RI, USA; 4 Massachusetts General Hosp, Boston, MA, USA; 5 Univ of Alabama at Birmingham, Birmingham, AL, USA; 6 The Fenway Inst, Fenway Hlth, Boston, MA, USA Background: Individual HIV clinics have reported increasing viral suppression rates. We evaluated changes in viral load (VL) over time and identified associated factors in HIV care settings. We were particularly interested in the impact of changing demographic and clinical characteristics such as shifting substance use patterns, adherence, and use of integrase inhibitors. Methods: Data are from 29,467 participants at 8 HIV clinics in the national CFAR Network of Integrated Clinical Systems (CNICS) cohort who had VL values between 1997-2014. We determined the annual percentage of participants with an undetectable VL (<400 copies/mL) each year including all tests and limiting them to 1/person/year. We then accounted for differential loss to follow-up by inverse probability of censoring weights based on prior VL and demographic factors. We used logistic mixed models to estimate associations with undetectable VL. Models were restricted to 2010-2014 and individuals on antiretroviral therapy (ART) to minimize impact of changes in treatment initiation guidelines. Results: The number of VL tests per year ranged from 9,180-39,540. Participants with undetectable VL values increased from 30% in 1997 to 87% in 2014 (see Figure) and results were similar when correcting for loss to follow-up. Undetectable VL percentages were higher for each decade of older age and in men vs. women (p values <0.05). For example, while the undetectable VL percentage increased over time for both men and women, men consistently averaged ~3% higher than women. Neither mean adherence increased nor current substance use decreased over time, and some drugs (e.g. marijuana, methamphetamines) increased. In multivariate models of individuals on ART after 2010, demographic factors including older age, white race, male sex and better medication adherence remained associated with undetectable VL (p values <0.05), as were integrase inhibitor use (OR 2.4, 95% CI 2.2-2.7, p<0.001). Conclusions: VL suppression rates at clinics across the US have improved dramatically in recent years. Adherence and substance use remained key predictors of absolute VL values but did not explain the improved percentage of viral suppression over time. Increased use of integrase inhibitors is likely one contributing factor. Findings suggest the need for increased evaluation and support for at-risk subgroups such as women and younger patients, and the need for a better understanding of factors that are driving changes in VL.

Poster Abstracts

442

CROI 2016