CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

FSW (≤24 yrs) were less likely to know their HIV status (47% (212/447) vs. 61% (928/1510), p<0.01) and less likely to be on ART (18% (81/447) vs. 40% (600/1510), p<0.01). Conclusion: HIV prevalence among AGYW in Southern Africa was exceptionally high in this pooled analysis of women who sell sex. Limited knowledge of HIV status and low ART coverage in this age group suggests that even among key populations such as FSW, HIV risk may not be evenly distributed. HIV prevention interventions for AGYW that target those who engage in sex work and those who are vulnerable to early entry into sex work may be most effective in maximizing prevention impact.

mortality in a single-factor analysis and remained predictive after adjusting for other factors (Table) with AHRs ranging from 0.69 to 0.77 depending on the considered model. Conclusion: In a large cohort of naïve HIV-1 infected subjects, the achievement of undetectable viral load within 6 months from ART start was associated with a lower risk of all-cause mortality.

912 CONTRIBUTION OF HIV DISEASE AND CARE STAGES TO HIV TRANSMISSION AMONG BALTIMORE MSM

Romain Silhol 1 , Marie-Claude Boily 1 , Dobromir Dimitrov 2 , Danielle German 3 , Colin Flynn 4 , Jason Farley 3 , Marcy Gelman 5 , James P. Hughes 2 , Deborah J. Donnell 2 , Adeola Adeyeye 6 , Robert H. Remien 7 , Chris Beyrer 3 , Gabriela Paz- Bailey 8 , Cyprian Wejnert 8 , Kate M. Mitchell 1 1 Imperial College London, London, UK, 2 Fred Hutchinson Cancer Research Center, Seattle, WA, USA, 3 Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA, 4 Maryland Department of Health and Mental Hygiene, Baltimore, MD, USA, 5 Fenway Health, Boston, MA, USA, 6 NIH, Bethesda, MD, USA, 7 Columbia University, New York, NY, USA, 8 CDC, Atlanta, GA, USA Background: HIV incidence remains high among men who have sex with men (MSM) in the United States. We estimated the contributions of MSM at different stages of HIV infection and the HIV care continuum to HIV transmission among MSM in Baltimore, MD, over the past 30 years. Methods: A deterministic compartmental model of HIV transmission among MSM was parameterised using, and fitted to demographic, epidemiological and care continuum data for MSM in Baltimore, using multiple combinations of plausible parameter values. We estimated the fraction of new direct and secondary HIV infections transmitted by MSM in different disease or care continuum stages, using population attributable fractions (PAFs) estimated over 10-year periods between 1988 and 2017, obtained by comparing the estimated number of new HIV infections with the numbers that would have occurred if the population in that stage could not transmit HIV. Average per-capita HIV transmissions per 100 people living with HIV (PLHIV)/year were also estimated over these periods. Results: We estimated that treated and untreated MSM transmitted a median 15% (95% uncertainty interval: 7-31%) and 89% (79-95%) of all new infections among MSM over 2008-2017 respectively. Untreated PLHIV in the acute stage were estimated to transmit 36 new HIV infections per 100py over 2007-2018, vs 25 and 8/100py for those with AIDS (CD4 <200 cells/μl) and those in the chronic non-AIDS stages (CD4 >200 cells/μl), respectively. The model PAF for undiagnosed MSM has declined over time, from 91% (68-98%) over 1988- 1997 to 38% (30-49%) over 2008-2017, when undiagnosed MSM represented 87% and 22% of all HIV-positive MSM, respectively. PAFs for diagnosed MSM increased from 15% (5-44%) over 1988-1997 to 82% (67-87%) over 2008- 2017. Most new infections attributable to diagnosed PLHIV were transmitted by untreated MSM (PAF for diagnosed untreated MSM: 67% (48-78%) over 2008-2017), who represented 41% of PLHIV (22-48%). Diagnosed MSM (including those treated) transmitted HIV to two-thirds as many individuals as undiagnosed PLHIV per capita over the last ten years (7 vs 11/100py), but around three times more than PLHIV on treatment (2/100py). Conclusion: Increases in the relative contribution to transmission of diagnosed MSM reflect improvements in HIV testing, but the majority of these transmissions arise from those who remain untreated, showing gaps in treatment provision and retention. Future interventions will need to address the remaining diagnosis and treatment gaps.

911 TIME TO UNDETECTABLE VIRAL LOAD ACHIEVEMENT AFTER ART START AND RISK OF MORTALITY Vincenzo Spagnuolo 1 , Laura Galli 2 , Alessandro Cozzi-Lepri 3 , Giuseppe Lapadula 4 , Andrea Antinori 5 , Giancarlo Orofino 6 , Nicoletta Bobbio 7 , Maria Cristina Moioli 8 , Andrea Calcagno 9 , Andrea De Luca 10 , Antonella D’Arminio Monforte 11 , Antonella Castagna 1 , for the ICONA Foundation Study Group 1 San Raffaele Vita-Salute University, Milan, Italy, 2 San Raffaele Scientific Institute, Milan, Italy, 3 University College London, London, UK, 4 San Gerardo Hospital, Monza, Italy, 5 IRCCS Lazzaro Spallanzani, Rome, Italy, 6 Amedeo di Savoia Hospital, Torino, Italy, 7 Galliera Hospital, Genoa, Italy, 8 ASST Grande Ospedale Metropolitano Niguarda, Milan, Italy, 9 University of Turin, Turin, Italy, 10 University of Siena, Siena, Italy, 11 University of Milan, Milan, Italy Background: No data on the association between the time to the first undetectable viral load (FUVL) after antiretroviral therapy (ART) initiation and mortality are available. In this study we evaluated whether time to FUVL after ART start is predictive of all-cause mortality in a large population of HIV-1- infected patients (pts). Methods: We included HIV-1-infected treatment-naïve pts, from the ICONA Cohort, who started ART (≥3 drugs) >1998, with ≥1 viral load (VL) and CD4+ values before and after ART start, who achieved undetectable VL (defined by a single HIV-RNA <50 copies/mL) after ART start. Results described as median (IQR) or frequency (%). Cumulative risk of all-cause mortality was summarized using Kaplan-Meier method, with follow-up for these analyses from the date of FUVL achievement until patient’s death, loss to-follow-up or last visit. Factors associated with all-cause mortality were identified using multivariate Cox proportional hazards regression models. Results: Overall, 10,000 subjects achieved undetectable VL after ART start and were included in the analyses. At ART start, age was 38 (32-46) years, 7805 (78%) males, 1701 (17%) HCV-coinfected, 1028 (10.3%) had a previous AIDS diagnosis, CD4+ 319 (172-464) cells/µL, CD4+/CD8+ ratio was 0.35 (0.20-0.53), HIV-RNA 4.77 (4.20-5.26) log10cps/mL; calendar-year of ART start was 2012 (2007-2015), 153 (1.5%) started a NRTI-, 3540 (35.4%) a NNRTI- , 4074 (40.7%) a PI- and 1956 (19.6%) an INSTI-based ART, 277 (2.8%) started more complex regimens. After ART start, 3161 (31.6%), 3399 (34%) and 3440 (34.4%) achieved the FUVL ≤3 months (M), 3-6 M and >6 M, respectively. During a median follow-up of 4.0 years (2.0-7.0), 300 deaths for any-cause occurred. Kaplan- Meier cumulative mortality estimates at 1, 3 and 5 years were higher (log-rank test: p=0.001) in subjects who achieved FUVL >6M [0.8% (95% CI 0.5-1.2), 2.4% (1.9-3.0) and 4.0% (3.2-4.9)] as compared to those who achieved FUVL ≤6M [0.6% (95% CI 0.4-0.8), 1.6% (1.2-1.9) and 2.3% (1.9-2.8)]. The achievement of FUVL ≤6M as compared to >6Mwas associated with a lower risk of all-cause

Poster Abstracts

CROI 2019 356