CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

signaling. If so, activation of Wnt/b-catenin in astrocytes may emerge as a prominent pathway to reverse astrocyte senescence associated with aging and/or neurodegenerative diseases beyond HAND. 508 Wnts-Mediated Astrocyte/CD8+ T-Cell Interactions Impacting HIV Neuropathogenesis Maureen H. Richards 1 ; Melanie S. Seaton 1 ; Stephanie Kim 2 ; Srinivasa Narasipura 1 ; Lena Al-Harthi 1 1 Rush University Medical Center, Chicago, IL, US; 2 Brown University, Boston, MA, US Background: Depletion of CD8+ T cells accelerates neuroAIDS, although there is no clear evidence that CD8+ T cells control HIV/SIV in the CNS. A subset of CD8+ T cells, CD4 dim CD8 bright T cells (DP T cells), is enriched in anti-HIV responses. This phenotype is generated through induction of Wnt/b-catenin signaling in CD8+ T cells. Wnts, small secreted glycoproteins, initiate Wnt/b-catenin signaling leading to gene regulation. Given that astrocytes are a rich source of Wnts, we evaluated whether astrocytes drive DP expression in vitro and in vivo. Methods: Wnts mRNA and protein from human progenitor-derived astrocytes (PDAs) were determined by qRT-PCR and western blot, respectively. Day three-Astrocyte Conditioned Media (ACM) was added to α CD3/CD28 activated PBMCs and percentage of DP T cells was assessed at day six by flow. Wnts were depleted from ACMs by protein A/G magnetic beads coated with anti-Wnt 1, Wnt2b, Wnt 3, Wnt 5b, Wnt 10b, or IgG1 isotype control. NOD/SCID/IL-2rc γ -/- (NSG) mice were reconstituted with 2x10 7 PBMCs (NSG- HuPBMCs) and infected by intraperitoneal injection with 10 4 TCID 50 HIV BAL at week 1. Three weeks post-infection, brain lymphocytes were isolated and expression of CD8, CD4, and HIVp24 determined by flow. Lastly, violet-labeled CD8+ T cells were injected into NSG brain, re-isolated at 1 week, and evaluated for CD4, CD45RA, CD27, CD28, and CCR7 expression by flow. Results: PDAs secrete Wnts 1, 2b, 3, 5b, and 10b. ACM induced DP expression by 2-fold in a Wnt-dependent manner. CD8 Single Positive (SP) and DP T cells are found in the brain of HIV+ NSG-HuPBMCs mice. DP cells harbor HIV, but interestingly, at three weeks post-infection, greater than 90% of CD4 SP T cells were depleted, while 25% of DP cells survived. Brain injection of Violet-labeled CD8+ T cells induced DP expression by 10-fold, where >80% of DP and CD8 SP T cells expressed surface markers indicative of a Terminal Effector Memory (TEMRA) phenotype (CCR7 - CD27 - CD28 - CD45RA + ). Conclusions: Astrocytes drive DP expression through Wnts. DP T cells are present in the CNS, are TEMRA, and HIV infected. Unlike CD4 SP T cells, DP T cells survive longer and may constitute a novel HIV CNS reservoir. These studies highlight a dynamic interaction between astrocytes and CD8 T cells, which can skew their differentiation and lead to a phenotype that, on one hand, can control HIV in the CNS but, on the other, heightens inflammatory responses in the CNS, contributing to HAND pathology. 2:30 pm– 4:00 pm Pharmacokinetics, Pharmacodynamics, and Adherence 509 HIV-1 Attachment Inhibitor Prodrug BMS-663068: Model-Based Dose Selection Ishani Savant Landry 1 ; Li Zhu 1 ; Malaz Abutarif 1 ; Matthew Hruska 1 ; Brian M. Sadler 2 ; Maria Pitsiu 3 ; George J. Hanna 1 ; DavidW. Boulton 1 ; Richard Bertz 1 1 Bristol-Myers Squibb, Princeton, NJ, US; 2 ICON plc, Cary, NC, US; 3 ICON plc, Manchester, United Kingdom Background: BMS-663068 is a prodrug of BMS-626529, an attachment inhibitor that binds to HIV-1 gp120, preventing initial viral attachment and entry into host CD4+ T-cells. Pharmacokinetic (PK)/pharmacodynamic (PD) modeling of Phase 2 data was performed to guide dose selection for subsequent studies. Methods: PK/PD data from two Phase 2 studies (AI438006, AI438011) in HIV-1-infected subjects (N=244) were used to develop a population PK model for BMS-626529 using non-linear mixed effects modelling (NONMEM v.7.2.0). Subsequently, relationships between IC 50 -normalized BMS-626529 systemic exposure (C max , C ssavg , C tau ) and efficacy/safety variables during monotherapy or after 24/48 weeks of combination therapy, were investigated using linear models. Monte-Carlo methods were used to assess the probability of achieving a decline of >0.5 (per FDA draft guidance) or >1.0 log 10 c/mL in HIV-1 RNA from baseline (BL) as a function of BMS-626529 systemic exposure after 7 days of BMS-663068 monotherapy for 5 proposed BMS-663068 doses (400mg/600mg [not studied clinically]/800mg BID and 600mg/1200mg QD). Results: A 2-compartment model with 1st-order elimination from the central compartment, zero-order release into a hypothetical absorption compartment, and 1st-order absorption into the central compartment described the PK of BMS-626529. The categorical covariates (ritonavir co-administration, BMS-663068 formulation type) and the continuous covariates (lean body mass, BL CD8%) were significant in the model. After 7 days of BMS-663068 monotherapy, BL BMS-626529 protein-binding adjusted IC 50 (PBAIC 50 ) influenced HIV-1 RNA decline and a clear relationship was observed between log e -transformed PBAIC 50 -adjusted C tau and change in HIV-1 RNA from BL (log 10 c/mL). No trends were observed between BMS-626529 systemic exposure and other efficacy/safety variables during BMS-663068 monotherapy or after 24/48 weeks of combination therapy. After 7 days of simulated BMS-663068 monotherapy, the probability of achieving a decline in HIV-1 RNA of >0.5 log 10 c/mL or >1 log 10 c/mL as a function of log e PBAIC 50 -adjusted C tau was predicted to be 99–100% and 57–73%, respectively across the doses (Table). TUESDAY, FEBRUARY 24, 2015 Session P-H1 Poster Session Poster Hall

Poster Abstracts

Predicted probability of achieving a decline in HIV-1 RNA of >0.5 or >1.0 log 10

copies/mL after 7 days of BMS-663068 monotherapy, as a function of log e PBAIC 50

-adjusted C

tau

Conclusions: Simulations showed an advantage of BID over QD dosing and a similar probability of achieving a decline of >1 log 10 c/mL across the 400, 600 and 800mg BID doses. Combined with clinical and safety observations, a 600mg BID dose of BMS-663068 was predicted to have the best benefit–risk profile and was selected for further study.

335

CROI 2015