CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

990 IMPACT OF TESTING DELAY ON LOW-LEVEL VIRAEMIA IN SOUTH AFRICA: A PROGRAMME-WIDE VIEW Nei-Yuan M. Hsiao 1 , Elton Mukonda 1 , Maia Lesosky 1 , Jean Maritz 2 , Wolfgang Preiser 2 , Landon Myer 1 1 University of Cape Town, Cape Town, South Africa, 2 Stellenbosch University, Cape Town, South Africa Background: Low-level viraemia (LLV) can be common in HIV-infected individuals receiving antiretroviral therapy (ART) and had been shown to be associated with risk of virological failure following a period of sustained virological suppression. In large ART programmes where viral load (VL) testing is centralised, delay in sample processing could affect the detection of these LLV due to sample degradation. Methods: We examined the frequency of LLV in patients on ART in the public health care system of the Western Cape province, South Africa from 2009- 2015 using routine laboratory VL data from the South African National Health Laboratory Service. We analysed the proportion of LLV test results by delay in sample processing time (<48h – 10days in 24h increments) over the sample period. The delay in processing is measured by turnaround time (TAT) which is calculated from the estimated time of collection till the time of test completion. LLV is defined as any detectable VL <400 copies/mL, we also calculated the proportion of VL test results falling between 400-1000 copies/mL. Results were stratified by calendar year and healthcare facility size. Results: We analysed 1,067,153 VL episodes from 509 facilities and identified 209,312 samples with LLV (20%). While the median TAT had significantly reduced from 173 hours in 2009 to 71 hours in 2015, with the exception of 2009, the frequency of LLV detection had remained fairly constant over that period (14% in 2009, 19% in 2010, 22% in 2011, 22% in 2012, 22% in 2013, 22% in 2014 and 20% in 2015). Samples with increased TAT are associated with less frequent detection of LLV. The decrease in detection is observed in sample with TAT>48 hours. While 23.8% of samples tested within first 48 hours had LLV, the proportions of LLV were 21.4%, 20.1%, 19.6% and 17.8% for samples tested on day 3, 4, 5 and 6 respectively. Further delay in testing did not result in significant change in the frequency of LLV. The decrease in LLV detection rate remained despite taking size of facility and calendar years into consideration. These trends were not observed for VL results between 400-1000 copies/mL, with 3% detected in the first 48h compared to 4% on day 6. Conclusion: The proportion of LLV decreases as TAT increase beyond the first 48 hours, pointing to potential early viral RNA decay due to compromised sample stability prior to plasma separation. As detection of LLV becomes more clinically relevant, the delay in testing should be minimised in large programmes.

Background: HIV and Hepatitis C virus (HCV) share a common mode of transmission, injection drug use. Some jurisdictions have replaced point- of-care rapid testing (RT) with laboratory-based testing (LBT) to enable use of combined antibody-antigen assays that identify acute HIV cases. While identification of acute HIV cases allows for early intervention, there are potential benefits associated with RT, including immediate test result delivery. To examine differences in HIV and HCV testing approaches in a high-risk population, we conducted a prospective randomized trial of HIV and HCV RT versus LBT at the largest residential detoxification center in the Boston area. Methods: Eligible participants (not aware of co-infection with HIV and HCV) completed a baseline questionnaire and were randomized to either HIV and HCV RT (Orasure) or LBT (HIV Combo Ag/Ab EIA, HCV EIA). For both groups, RT and specimen collection for LBT were performed by a trained counselor. RT was provided within 20 minutes; LBT results in 2-3 days. The primary outcome was the proportion of test results delivered. Logistic regression was used to determine factors associated with test result delivery. Results: Among 341 individuals screened (from November 2016 to July 2017), 200 met inclusion criteria with the following characteristics: mean [SD] age, 39 [10] years; 46%White, 25% Black, 21% Hispanic, 7% other; and 58% (117/200) injected drugs, 31% (63/200) shared needles, 87% (149/171) had unprotected sex in the 6 months prior to admission. Of the 200 randomized participants, 99 received RT and 101 LBT; 17 were discharged prior to testing (4 RT, 13 LBT) and 7 assigned to LBT had hemolyzed samples that were not tested. Among all participants, 0.5% (1/200) had a reactive HIV test and 48% (96/200) reactive HCV tests; 96% (95/99) received test results in the RT arm and 51% (52/101) in the LBT arm. Test modality (aOR, 19.25, 95% CI, 7.05-52.50) and Black race (aOR, 0.34, 95% CI, 0.12- 0.98) were independently associated with result delivery after multivariable adjustment controlling for age, gender, race/ethnicity, high-risk behavior, unstable housing, past HIV or HCV testing, and treatment for mental health disorder. Conclusion: To maximize HIV and HCV test result delivery to high-risk transient populations, use of RT should be considered over LBT in certain settings, such as residential detoxification centers. Additional studies are needed to identify reasons for lower test result delivery among Black individuals in this setting.

Poster Abstracts

992 HIV SELF-TESTING INCREASES TESTING IN YOUNG SOUTH AFRICAN WOMEN: RESULTS OF AN RCT Audrey Pettifor 1 , Kathleen Kahn 2 , Linda Kimaru 1 , Zola Mayakayaka 2 , Amanda Selin 1 , Noah A. Haber 1 , Ryan Wagner 3 , Xavier Gomez-Olive 2 , Hailey Gilmore 4 , Memory Mhembere 2 , Daniel Westreich 1 , Rhian Twine 2 , Sheri A. Lippman 4 1 University of North Carolina Chapel Hill, Chapel Hill, NC, USA, 2 University of the Witwatersrand, Acornhoek, South Africa, 3 University of the Witwatersrand, Johannesburg, South Africa, 4 University of California San Francisco, San Francisco, CA, USA Background: To meet the UNAIDS 90/90/90 goal it is imperative to increase HIV testing uptake. HIV Self-Testing (HIVST) may increase early HIV detection, particularly among young people. Secondary distribution of test kits to peers/partners may be a good way to reach young people who do not attend traditional health services. Methods: From December 2016 – July 2017 we enrolled 284 South African young women age 18-24 years in a randomized control trial to examine HIV testing uptake among those randomized to either: 1) an invitation to a local

991 RAPID VS. STANDARD TESTING FOR HIV AND HCV AT A DRUG DETOX: A RANDOMIZED TRIAL Sabrina A. Assoumou 1 , Samantha M. Paniagua 2 , Benjamin Linas 1 , Jianing Wang 1 , Jeffrey H. Samet 2 , Jonathan Hall 1 , Curt G. Beckwith 3 1 Boston Medical Center, Boston, MA, USA, 2 Boston University, Boston, MA, USA, 3 Brown University, Providence, RI, USA

CROI 2018 379