CROI 2016 Abstract eBook

Abstract Listing

Oral Abstracts

93LB PD-1 Blockade as an Adjunct Therapy to ART and Potential to Destabilize SIV Reservoir Geetha Mylvaganam 1 ; Sakeenah Hicks 1 ; Benton Lawson 1 ; Melon Nega 1 ;VijayakumarVelu 1 ; Rafi Ahmed 1 ; Gordon J. Freeman 2 ; Rama Amara 3 1 Emory Univ, Atlanta, GA, USA; 2 Dana-Farber Cancer Inst, Boston, MA, USA; 3 Yerkes Natl Primate Rsr Cntr, Emory Univ, Atlanta, GA, USA

Background: The expression of the inhibitory receptor programmed death-1 (PD-1) on anti-viral CD8 T cells and virally infected CD4 T cells provides an immunological signature for both T cell dysfunction and viral latency during chronic SIV/HIV infection. We hypothesized that PD-1 blockade administered during the initiation of anti-retroviral therapy (ART) and under fully suppressive ART would have direct effects on both dysfunctional CD8 T cells and latently infected CD4 T cells. To test our hypothesis we developed a primatized anti-human PD-1 Ab to allow for repeated infusions in rhesus macaques (RMs) and administered PD-1 blockade to chronically SIV infected RMs in combination with ART. Methods: SIVmac251 infected RMs were administered 5 infusions (over 14 days) of a 3mg/kg dose of primatized anti-PD-1 Ab 10 days prior to the initiation of ART or saline. 26-30 weeks post ART, RMs received 3 monthly infusions of 10mg/kg anti-PD-1 or saline. ART was interrupted 2 weeks after the final PD-1 Ab infusion. Results: PD-1 blockade administered during the initiation of ART enhanced proliferation of anti-viral CD8 T cells (p=0.02), increased their cytotoxic potential (p=0.04) and polyfunctionality (p=0.01). Importantly, the PD-1 Ab treated animals showed more rapid viral suppression (42 days in the PD-1 group versus 140 days in saline group; p = 0.01) and greater reconstitution of Th17 cells in the rectal mucosa (p = 0.01) following initiation of ART. Moreover, PD-1 blockade administered under suppressive ART resulted in significant and detectable plasma viral blips, suggesting possible effects on destabilizing the latent viral reservoir. Following ART interruption, PD-1 Ab treated animals showed up to an 80-fold reduction in set point viremia compared set point levels prior to initiation of ART. Conclusions: These results reveal for the first time the potential of PD-1 blockade both on restoring anti-viral CD8 T cell function and possibly destabilizing the viral reservoir under ART. They highlight the potential of PD-1 blockade to work synergistically with other therapeutic agents such as vaccines and latency reversing agents to effectively diminish HIV reservoir under ART as a means to establish a functional cure. 94 Follicular CTL Accumulate in SIV-Infected Lymph Nodes Due to Immune Activation Costantinos Petrovas 1 ; Sara Ferrando-Martinez 1 ; Amarendra Pegu 1 ; Sarah F. Andrews 1 ; David Ambrozak 1 ; Adrian B. McDermott 1 ; Jason M. Brenchley 2 ; John R. Mascola 3 ; Richard A. Koup 3 1 NIH, Bethesda, MD, USA; 2 Frederick Natl Lab for Cancer Rsr, Frederick, MD, USA; 3 VRC, NIAID, NIH, Bethesda, MD, USA Background: Strategies aimed to eradicate HIV/SIV infection require the characterization of cytolytic CD8 T cells that can migrate and recognize infected cells at the sites of active replication. Follicular CD4 T cells (T FH ), located within the germinal centers (GC) of the lymph node (LN) B cell follicles support active HIV/SIV replication even under suppressive combined antiretroviral therapy (cART). Thus, a comprehensive analysis of CD8 T cell dynamics in the LN, particularly in the GC area, will provide valuable information towards HIV/ SIV eradication. Methods: Multi-parametric flow cytometry, multiplexed confocal imaging, histocytometry and fluidigm-based analysis were used to characterized PBMC, LN-derived cells and LN tissues from (1) SIV-negative rhesus macaques (RM), (2) acute and chronic SIV-infected RM and (3) chronic SIV-infected African green monkeys (AGM). CM9 (Gag) and TL8 (Tat) tetramers and intracellular cytokine staining after peptide pool stimulation identified virus-specific CD8 T cells while in vitro killing assays using sorted LN CD8 T cell populations and an anti-CD3/VRC07 antibody proved the cytolytic activity of LN-resident CD8 T cells. Results: Chronic SIV infection in RM was characterized by the accumulation of memory (CD28 hi CD95 hi ) and effector (CD28 lo CD95 hi ) CD8 T cells with a follicular (CCR7 lo CXCR5 hi ) phenotype (fCD8). Imaging and histocytometry confirmed the accumulation of fCD8 within the GC. fCD8, when compared to CCR7 hi CXCR5 lo non-fCD8, express higher amounts of Granzyme B and exert a higher capacity to mediate in vitro bispecific-mediated redirected killing of infected cells. Chronically infected AGM, on the other hand, did not accumulate fCD8. Fluidigm analysis showed that both memory and effector fCD8 clustered together and were mostly affected by the SIV-infection status. Besides, the frequency of LN CD8 T cells was significantly correlated with LN CD14 hi CD16 hi monocytes. Upon stimulation, monocytes from SIV-infected RM secreted increased amounts of CXCL10, a CXCR3 ligand. Confocal imaging confirmed the proximity of CXCR3-expressing cells and monocytes within the LN. Conclusions: Cytolytic fCD8 T cells are able of killing infected cells, further justifying the use of the SIV model to develop new eradication strategies. Our results suggest that chronic immune activation is a major force driving the accumulation of fCD8, pointing to specific targets as mediators of the intra-follicular CD8 trafficking. 95LB Repeated TLR7 Agonist Treatment of SIV+Monkeys on ART Can Lead to Viral Remission James B. Whitney 1 ; So-Yon Lim 1 ; Christa E. Osuna 1 ; Srisowmya Sanisetty 1 ;Tiffany L. Barnes 2 ;Tomas Cihlar 2 ; Michael Miller 3 ; Romas Geleziunas 2 ; Joseph Hesselgesser 2 1 Beth Israel Deaconess Med Cntr/Harvard, Boston, MA, USA; 2 Gilead Sciences, Inc, Foster City, CA, USA; 3 Gilead Scis, Inc, Foster City, CA, USA Background: The identification of pharmaceutical agents capable of safely reversing HIV-1 latentcy in ART-treated patients is urgently needed. We have previously reported that an oral TLR7 agonist GS-986, at doses that produce peripheral IFN-alpha, induces transient plasma viremia in SIV-infected rhesus macaques (RMs) on antiretroviral therapy (ART). In this follow up study, we assessed if lower clinically relevant doses of the TLR7 agonist GS-986 and the clinical compound GS-9620, which produce low or no detectable peripheral IFN-alpha, could induce transient plasma viremia or perturb viral reservoirs. Methods: 11 RMs were infected with SIVmac251 and started on daily suppressive ART at 65 days post-infection. Virologic suppression (<50 SIV RNA copies/mL) was achieved and maintained through 67 weeks. The cohort was divided into 4 groups receiving: (1) 19 doses of vehicle (n=2), (2) 19 doses of 0.1mg/kg GS-986 (n=3), (3) 19 doses of 0.05mg/ kg GS-9620 (n=3), or (4) 10 doses of 0.15mg/kg GS-9620 (n=3) once every two weeks while maintaining ART. We longitudinally assessed plasma SIV RNA, total SIV DNA in PBMCs, lymph node (LN) and colon biopsies, and ex vivo SIV production from LN mononuclear cells and PBMC cultures stimulated with ConA. Results: The first two TLR7 doses did not induce detectable plasma viremia. However, doses 3-10 led to transient but inconsistent production of plasma SIV RNA “blips” in all TLR7- treated RMs, while additional doses (11 to19) did not induce SIV. After completing TLR7 dosing but prior stopping ART, SIV DNA levels were reduced in PBMC, colon and LN biopsies, and levels of ex vivo stimulated virus production were also diminished, with no such changes in the control group. To assess the impact of TLR7 treatment on viral reservoirs, ART was discontinued 2 weeks after the last TLR7 dose. Plasma virus rebound in 7 of the 9 TLR7-treated RMs was similar to that of the control group. However, two RMs having received either 19 doses of GS-986 or 10 doses of GS-9620 maintained undetectable plasma viral load for >60 days after stopping ART. Prior to ART cessation, these same two RMs were negative for ex vivo virus induction. Conclusions: Repeated low doses of TLR7 producing minimal to undetectable peripheral IFN-alpha in SIV-infected ART-suppressed RMs induced transient plasma viremia, decreased viral DNA levels, and delayed plasma virus rebound after stopping ART in some RMs. These data support the ongoing clinical testing of GS-9620 in HIV-infected persons on ART. 96 Pharmacologic Measures of Adherence and Relationship to Drug Response Peter L. Anderson , Univ of Colorado, Denver, CO, USA Non-adherence occurs in about 40% of patients on chronic therapies. In studies, it complicates interpretation of study outcomes and may lead to inaccurate conclusions regarding biological drug efficacy. One approach to deal with this is to incorporate pharmacologic monitoring for adherence assessment. This has become commonplace for PrEP studies and has been evaluated for other HIV-related disease states. This presentation will discuss lessons learned from pharmacologic measures of adherence in PrEP studies and their utility in interpreting study outcomes, including guidance about pharmacokinetic forgiveness, predicted efficacy for intermittent dosing strategies, and estimation of concentration- response relationships. Parallels will be drawn for other HIV-related disease states such as virologic breakthrough or development of drug resistance.

Oral Abstracts

36

CROI 2016