CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

86% of WVS switching to TAF and 85% continuing TDF (p=0.99). Overall TAF and TDF were well-tolerated. Discontinuation due to adverse event/death was 0% (TAF) vs. 1.6% (TDF) in WTN and 1.3% (TAF) vs. 2.2% (TDF) in WVS. At W96 there was less impact on renal biomarkers in WTN initiating TAF- vs TDF-based regimens (p<0.001; Table), and decreases in BMD were smaller (p<0.001; Table). Women switching from TDF to TAF experienced decreases in tubular proteinuria (p<0.001; Table) and increases in BMD (p<0.001; Table) at W96. Conclusion: Similar to the overall results in pivotal naïve and switch trials of FTC/TAF-based regimens, cis-women who initiated or switched to TAF had significantly improved bone and renal safety parameters compared to TDF, with similar rates of virologic suppression through W96. These pooled data from 7 studies demonstrate a safety advantage for initiating therapy with or switching to TAF compared to TDF in women.

significantly for women with CST2 and CST4 post-ART. Immune reconstitution was not associated with significant vaginal microbiome changes and pre-ART vaginal CST was not associated with immune reconstitution. Conclusion: ART initiation was associated with decreases in abundance in indicator bacteria from two vaginal CSTs, which are associated with bacterial vaginosis; however, other CSTs, including one characterized by Prevotella and one characterized by Lactobacillus iners remained stable. Overall, vaginal microbiome did not change significantly with immune reconstitution.

Poster Abstracts

521 IMMEDIATE ART INITIATION IN ACUTE INFECTION IMPROVES CLINICAL OUTCOMES, SABES STUDY Javier R. Lama 1 , Rachel A. Bender Ignacio 2 , Jorge A. Gallardo-Cartagena 1 , Jessica Rios 1 , Ricardo Alfaro 1 , Karin S. Sosa-Barbaran 1 , Carolyn Bain 3 , Christopher D. Pilcher 4 , Ann Duerr 5 , for the Sabes Study 1 Asociacion Civil Impacta Salud y Educacion, Lima, Peru, 2 University of Washington, Seattle, WA, USA, 3 PATH, Seattle, WA, USA, 4 University of California San Francisco, San Francisco, CA, USA, 5 Fred Hutchinson Cancer Research Center, Seattle, WA, USA Background: Current standard of care recommends immediate ART initiation after HIV diagnosis, but recommendations are grounded in data on established infections. Aside from the high public health benefit of reduced transmission, individual benefits of starting ART in early (acute and recent) HIV infection have not been well characterized. Methods: Two-hundred sixteen participants diagnosed with early HIV infection via monthly screening in the Sabes study (a treatment-as-prevention intervention among MSM and transwomen in Lima, Peru) were randomized to start ART immediately or after a short delay, and were categorized as having started ART within 30, 90, or >90 days after estimated date of detectible HIV infection (EDDI). Survival analyses with log-rank tests evaluated rates of virologic suppression and adverse events in the first year after HIV diagnosis. We tested differences in CD4+ counts with Kruskal-Wallis tests. Analyses were adjusted when appropriate, for time under observation or time on ART. Results: All 105 participants who were offered same day ART started ART; five of 111 offered delayed ART did not start during the study period (p=0.03). Total adverse events and non-ART-related adverse events were less frequent in persons starting ART within 30 days of EDDI, with a trend toward fewer ART- related events than in those who started ART after 30 or 90 days (Table). While a higher proportion of the >90-day group reached virologic suppression by 24 weeks on ART, enrollment HIV viral load was highest in the <30-day group. CD4+ counts and CD4+/CD8+ ratio increased in all groups but normalized more completely in the <30-day group; adjusted for time on ART, median CD4+ at 48 weeks remained significantly higher than in those who started ART after 30 days (p=0.009). Increase in CD4+ on study was not different when adjusted by time on ART, but the greatest improvements in CD4+/CD8+ ratio were in the 31-90 day group, which began with the lowest ratios (+0.55, p=0.005). Conclusion: In early HIV infection, those who began ART within 30 days of estimated date of HIV infection had better clinical outcomes, including fewer adverse events during the first year. While several observational studies have suggested similar findings, the Sabes study is likely the only demonstration of this effect in a randomized study, where risk of confounding is minimized.

520 EFFECT OF ANTIRETROVIRAL THERAPY AND IMMUNE RECONSTITUTION ON VAGINAL MICROBIOME Zoe Packman 1 , Cindy M. Liu 2 , David Serwadda 3 , Fred Nalugoda 3 , Maliha Aziz 2 , Alison Abraham 1 , Jessica Prodger 4 , Rupert Kaul 5 , Ronald H. Gray 6 , Aaron Tobian 1 , Thomas C. Quinn 7 , Steven J. Reynolds 1 1 Johns Hopkins University School of Medicine, Baltimore, MD, USA, 2 George Washington University, Washington, DC, USA, 3 Rakai Health Sciences Program, Kalisizo, Uganda, 4 Western University, London, ON, Canada, 5 University of Toronto, Toronto, ON, Canada, 6 Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA, 7 National Institute of Allergy and Infectious Diseases, Bethesda, MD, USA Background: Host factors, including menarche, menstruation, and pregnancy are known to impact vaginal microbiome composition. Both HIV infection and the immune reconstitution associated with antiretroviral therapy (ART) can cause broad immunological changes in the vaginal microbiome. We assessed vaginal microbiome after starting ART and its association with immune reconstitution (>50 increase of CD4+ T-cells post ART initiation). Methods: We characterized the vaginal microbiota of HIV-1 and HSV-2 coinfected women (n=94) who initiated ART in a trial of HSV-2 suppression with acyclovir in Rakai, Uganda. Vaginal swabs were collected 1-month pre-ART and at 4- and 6-months after ART initiation. Proportional and absolute abundance of vaginal bacteria was estimated by sequencing of the 16S rRNA V3V6 region. Vaginal community state types (CSTs) were identified using proportional abundance data with Bray-Curtis distance and hierarchal clustering by Ward’s method. Microbiome composition was compared using permutational MANOVA. Changes in absolute and proportional abundance of indicator bacteria were assessed using Wilcoxon signed-rank test. Characterizing anaerobes selected by indicator analysis. Results: We identified five vaginal CSTs among HIV+ women prior to ART initiation: one characterized by Gram-positive anaerobes (CST1), one characterized by Gram-negative anaerobes (CST2), one characterized by Gardnerella (CST3), and one characterized by Lactobacillus iners (CST4). Prior to ART, the likelihood of having a particular vaginal CST did not vary by HIV viral load or CD4+ T-cell count. ART did not have a significant impact on overall vaginal microbiome composition (p=0.74). However, among two CSTs-CST1 and CST3-abundance of Gram-positive (Anaerococcus, Finegoldia) and Gram- variable (Gardnerella) indicator bacteria decreased significantly six-months post-ART (Table 1). In contrast, indicator bacteria abundance did not change

CROI 2019 195