CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

an independently reduced risk of VF, adjusted hazard ratio (aHR)=0.23 (95% CI: 0.05–0.97; P=0.045), while those with CD4 <200 cells/µL had a three-fold raised risk, aHR=3.49 (95% CI: 2.00-6.14; P<0.0001). Conclusion: Despite initial concerns of reduced ART adherence amongst clinically well HIV-positive people initiating ART with high CD4 counts, participants in this study initiating ART with CD4 count ≥500 cells/µL had much better virological outcomes than those with baseline CD4 count <500 cells/µL.

findings was statistically significant. New strategies need to be developed to engage hard-to-retain patients, and to increase VS among those retained under care.

514 PROJECT RHAE: A PILOT STUDY OF RAPID ART START AND RESTART IN BALTIMORE CITY Joyce Jones , Yu-Hsiang Hsieh, Geetanjali Chander, Kathleen Page, Richard Rothman, Richard D. Moore Johns Hopkins University School of Medicine, Baltimore, MD, USA Background: Rapid HIV Treatment Initiation (RHTI) has shown good clinical outcomes in a variety of settings. Data are limited on the feasibility of RHTI in areas predominately affecting African Americans with high rates of poverty and in previously diagnosed patients. We conducted a pilot study of RHTI and treatment reinitiation (RHTRI) in an academic medical center and public health STD clinic in Baltimore. Methods: We recruited patients newly diagnosed (ND) or previously diagnosed (PD) with HIV not on ART from the Johns Hopkins (JH) John G. Bartlett Specialty Practice, the JH inpatient HIV service, the JH Emergency Department and the Baltimore City Health Department STD clinics. A baseline and 4 week survey of demographics; mental health and substance use; barriers and facilitators to care; and acceptability of RHTI/RHTRI was performed. A survey-only phase began 2/13/17 and 8/30/17 a RHTI and RHTRI phase was added in which clinic and inpatient providers prescribed ART at first clinic visit or during hospitalization. We evaluated survey, ART initiation (rapid vs. delayed) and VL data for patients recruited through 9/1/18 with >12 week follow up. VL ≥ 200 copies/mL or no post-ART VL was considered detectable. Results: From 2/13/17 to 9/1/18, 70 patients enrolled (38 ND, 32 PD). Most were African American (84%), male (70%) with HIV risk factor MSM (34%) or heterosexual sex (30%). Mean age was 35±12 years and 41% had an annual household income of <$5,000/yr. 25% reported recent panic symptoms (PHQ-A), 22%major or severe depressive symptoms (PHQ-8) and 41% at-risk alcohol use (AUDIT-C). 99% reported they would start same-day ART if available. 87% of ND and 66% of PD patients received rapid ART (Figure 1). 22/28 (79%) rapid ND patients achieved an undetectable VL (UDVL) vs. 2/4 (50%) delayed. 9/19 (47%) rapid PD patients had UDVL vs. 5/9 (56%) delayed. Median time to UDVL was 50 days for ND (50 days rapid vs. 42 days delayed) and 56 days for PD (31 days rapid vs. 83 days delayed). Conclusion: RHTI and RHTRI were highly acceptable and demonstrated promising rates of UDVL in this predominately African American population with high rates of poverty, mental health issues and hazardous alcohol use. Rapid PD patients had lower rates of UDVL than delayed PD but median time to UDVL was faster in the rapid group and sample size is small. Ongoing recruitment and follow up will help characterize the effectiveness of RHTI/RHTRI in achieving and maintaining durable VL suppression in these key patient groups.

Poster Abstracts

513 RCT OF INDIVIDUALIZED COMMUNICATION STRATEGY AMONG HIV- INFECTED PERSONS STARTING ART Marcelo H. Losso, Marisa Fuentes, Valentina Losso, Martha Michaan, Lea Pastore,

Lorena Puente, Amalia Vazquez, Manuela Bullo Hospital JM Ramos Mejía, Buenos Aires, Argentina

Background: Several strategies have been designed to improve retention in care and viral suppression (VS) in patients starting antiretroviral therapy (ART). In Latin America few strategies have been locally tested in a randomized way and evidence-based decisions are scarce. Methods: We conducted a multicentric RCT including naïve patients prescribed ART. The main goal was to compare retention in care and VS between arms. Subjects starting ART were randomly assigned to standard of care follow-up (SOC) or SOC plus an individualized communication strategy (ICS). At weeks 2,4 and every 4 weeks up to 1 year, trained personnel contacted patients using patient-selected communication method to screen for retention and adherence problems with a semi-structured interview. The primary outcome was successful linkage, defined as ambulatory care in the last 6 months with no ART interruption; secondary outcomes were VS defined as HIV-RNA <200 cps in the last measurement and successful treatment defined as the presence of successful linkage and/or VS. Descriptive statistics were used for baseline characteristics. Risk differences (RD) and 95% confidence intervals (CI) were estimated using linear regression for primary and secondary outcomes obtaining crude and adjusted estimates. Results: A total of 207 participants were randomized (107 to SOC, 100 to ICS). Median age was 31 yrs (IQR 26,40), 80.2%were male, 59%were MSM, 26.4% were immigrants. Median baseline HIV-RNA log was 4.43 (IQR 3.73,5.05) and CD4 count/mm3 was 398 (IQR 220,576). There was not significant RD in treatment success across arms in the crude (SOC=0.62, ICS=0.68, RD=0.06, 95%CI: -0.07,0.19) and adjusted by sex, age, transmission category, immigration status and baseline CD4 (SOC=0.29, ICS=0.38, RD=0.09, 95%CI: -0.04,0.22) estimates. VS was higher in the SOC arm, but not statistically significant (SOC=0.55, ICS=0.49, RD=-0.06, 95%CI: -0.15, 0.12). Successful treatment was lower among SOC (0.66) than among ICS (0.72) but not statistically significant (RD=0.06, 95%CI: -0.07, 0.18). Conclusion: In a mostly male MSM cohort in Argentina, linkage was successful one year after ART initiation in approximately two thirds of the population. However, over one quarter was not linked to care at one year. The strategy showed increased linkage and treatment success but lower VS. None of these

CROI 2019 192