CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

346

Relationship Among Viral Load Outcomes in HIV Treatment Interruption Trials Graham C. Treasure 1 ; Evgenia Aga 2 ; Ronald Bosch 2 ; JohnW. Mellors 3 ; Daniel Kuritzkes 4 ; RajeshT. Gandhi 5 ; Michael Para 6 ; Jonathan Z. Li 1 1 Brigham and Women’s Hosp, Harvard Med Sch, Boston, MA, USA; 2 Harvard Sch of PH, Boston, MA, USA; 3 Univ of Pittsburgh, Pittsburgh, PA, USA; 4 Harvard Med Sch, Boston, MA, USA; 5 Massachusetts General Hosp, Boston, MA, USA; 6 Ohio State Univ, Columbus, OH, USA Background: Analytic treatment interruption (ATI) trials are crucial for evaluating potential HIV curative strategies, but interpretation of these trials is complicated by the use of varying virologic outcome measures and an incomplete understanding of relationships between potential virologic endpoints. While viral load (VL) set point has historically been the most common virologic outcome of ATI studies, there is increasing use of time to viral rebound as an alternative endpoint. In addition, other studies have presented lower ATI VL set point compared to pre-ART VL as evidence of therapeutic efficacy in clinical trials. Using a pooled dataset of AIDS Clinical Trials Group (ACTG) ATI studies, we completed an in-depth characterization of ATI VL endpoints. Methods: We performed a pooled analysis of 6 ACTG ATI studies with 235 total participants who were virologically suppressed prior to ATI. We evaluated the relationship between two common VL outcomes—timing of confirmed VL rebound ≥200 HIV RNA copies/mL and VL set point (mean log 10 VL during ATI weeks 12-16). Associations between set point and timing of viral rebound were analyzed using Spearman correlation, Kruskal-Wallis, Mann-Whitney, and Chi-square tests. Results: For participants who had HIV rebound at or before week 12, there was no significant association between timing of viral rebound and the VL set point. However, participants with viral rebound after 12 weeks also had significantly lower set points (rebound ≤12 [N=176] vs. >12 weeks [N=14]: median 4.1 vs. 1.9 log 10 HIV RNA copies/mL, P<0.001, Figure 1). Participants treated during chronic infection had higher ATI VL set points than those treated during early infection (chronic [N=141] vs. early [N=50]: median 4.2 vs. 3.4 log 10 HIV RNA copies/mL, P<0.001). Pre-ART VL was correlated with ATI VL set point (Spearman r =0.38, p<0.001), but not rebound timing. In 68% of participants, ATI VL set point was lower than pre-ART VL with a median 2-fold decrease, but there was significant variation in the fold-change between ATI VL set point and pre-ART VL (Q1, Q3: 5-fold decrease, 1.7-fold increase). Conclusions: Our results reveal a complex relationship between several key virologic factors in HIV ATI trials, including pre-ART VL, viral rebound timing, and ATI VL set point. The association of delayed viral rebound with lower ATI VL set point suggests the presence of virologic and/or immunologic factors mediating both outcomes.

347 Viral and Immune Characteristics of HIV Post-Treatment Controllers in ACTG Studies

Behzad Etemad 1 ; Xiaoming Sun 2 ; Michael M. Lederman 3 ; Rachel Z. Gottlieb 1 ; Evgenia Aga 4 ; Ronald Bosch 4 ; Jeffrey Jacobson 5 ; RajeshT. Gandhi 6 ; Xu G.Yu 2 ; Jonathan Z. Li 1 1 Brigham and Women’s Hosp, Harvard Med Sch, Boston, MA, USA; 2 Ragon Inst of MGH, MIT, and Harvard, Cambridge, MA, USA; 3 Case Western Reserve Univ, Cleveland, OH, USA; 4 Harvard Sch of PH, Boston, MA, USA; 5 Drexel Univ Coll of Med, Philadelphia, PA, USA; 6 Massachusetts General Hosp, Boston, MA, USA Background: HIV post-treatment controllers (PTCs) are individuals who can maintain low levels of viremia after ART discontinuation. They have primarily been identified from patients treated during acute infection and little is known about PTCs who were treated during chronic infection. Understanding the mechanisms of HIV control in PTCs has implications for the design of novel strategies for HIV remission. Methods: We evaluated 497 participants from 8 ACTG analytic treatment interruption (ATI) studies to identify PTCs who generally maintained viral loads ≤400 HIV RNA copies/mL for ≥24 weeks. Non-PTC control participants were matched by study arm and selected randomly. Total HIV DNA and CA-usRNA were measured in PBMCs at 3 time points: pre-ATI, early and late post-ATI. T cell intracellular cytokine staining (ICS) was performed on PBMCs stimulated with HIV Gag peptide pool and NK cell ICS was performed with PBMCs stimulated with K562 cells. Soluble markers of inflammation were measured with ELISA-based assays. Results: 16 PTCs were identified (3.2% of participants), including 6 treated during early infection (6.6% of this group) and 10 treated during chronic infection (2.5% of this group). Median duration of documented virologic control was 96 weeks. There were no significant differences in pre-ATI CD4+ count, duration of ART, or frequency of protective HLA alleles between PTCs and non-PTCs. Pre-ATI HIV DNA and CA-RNA were detectable in 14% and 29% of PTCs, respectively. HIV DNA and CA-RNA levels did not significantly increase in PTCs after ATI (median DNA and CA-RNA <50 copies/10 6 PBMCs for all time points). For individuals treated during chronic infection, non-PTCs had higher HIV DNA levels pre-ATI and higher DNA and CA-RNA levels early post-ATI. No significant differences were detected in HIV-specific T or NK cell activity between PTCs and non-PTCs. While there were no significant pre-ATI differences in soluble and T cell markers of inflammation, non-PTCs had increased CD8+ cell activation as well as IP10 and sCD163 levels post-ATI. Conclusions: While rare, PTCs can be identified from individuals who were not treated during acute HIV infection. PTCs maintained a small HIV reservoir even after ART discontinuation without a significant increase in immune activation or HIV-specific T or NK cell activity. The detection of pre-ATI HIV DNA and CA-RNA did not preclude the possibility of post-treatment control, suggesting inefficient viral replication or antiviral immune activity may mediate this control. 348 HIV-1 in the Blood and Intestine Contributes to Viremia During Treatment Interruption Anni Winckelmann 1 ; Kirston Barton 2 ; Bonnie Hiener 2 ;Thomas A. Rasmussen 3 ; MartinTolstrup 1 ;Wei Shao 4 ; PaulW. Denton 1 ; Sharon R. Lewin 3 ; Ole S. Søgaard 1 ; Sarah Palmer 2 1 Aarhus Univ Hosp, Aarhus Nord, Denmark; 2 Cntr for Virus Rsr, Westmead Millennium Inst, Westmead, Australia; 3 Doherty Inst for Infection and Immunity, Univ of Melbourne, Melbourne, Australia; 4 Leidos Biomed Rsr, Inc, Frederick, MD, USA Background: Transcriptionally silent HIV-1 proviral genomes contribute to HIV-1 persistence in long-lived CD4+ T cells. Administration of panobinostat to HIV-1 infected individuals on long-term antiretroviral therapy (ART) activates HIV-1 RNA transcription in these latently infected cells. The extent to which the induced HIV-1 RNA transcripts and related HIV-1 DNA sequences represent virus able to contribute to viremia upon treatment discontinuation is unknown. In this study, we compared HIV-1 sequences in CD4+ T-cells from the peripheral blood and intestinal lamina propria mononuclear cells (LPMCs) during experimental therapy with panobinostat to sequences that were collected from the plasma during a post-panobinostat analytical treatment interruption (ATI).

Poster Abstracts

132

CROI 2016