CROI 2024 Abstract eBook

Abstract eBook

Poster Abstracts

533

Drug-Controlled Anti-PD-1 CAR T-Cells to Target the Replication Competent Reservoir in Tfh Cells Karsten Eichholz 1 , Yoshinori Fukazawa 2 , Christopher W. Peterson 3 , Francoise Haeseleer 3 , Benjamin Varco-Merth 2 , Sandra Dross 4 , Haesun Park 2 , Caralyn S. Labriola 2 , Michael Axthelm 2 , Jeremy Smedley 2 , Hans-Peter Kiem 3 , Louis Picker 2 , Afam A. Okoye 2 , Corey Larry 3 1 Fred Hutchinson Cancer Research Center, Seattle, WA, USA, 2 Oregon Health and Sciences University, Portland, OR, USA, 3 Fred Hutchinson Cancer Center, Seattle, WA, USA, 4 University of Washington, Seattle, WA, USA Background: Programmed cell death protein 1 (PD-1) is an immune checkpoint marker expressed on memory T cells and enriched in latently-infected CD4+ T cells and T follicular helper cells (TFH) containing replication-competent human immune deficiency virus 1 (HIV) provirus in people with HIV on antiretroviral therapy (ART). We recently tested a novel anti-PD-1 chimeric antigen receptor (CAR) T cell approach in vivo in simian immunodeficiency virus (SIV) mac239 infected and SIV-naïve rhesus macaques (RM) to assess the impact of PD-1 depletion on viral reservoirs and rebound dynamics. We found that anti-PD-1 CAR T cells expanded efficiently and rapidly eradicated all TFH cells from the germinal centers with concomitant depletion of detectable SIV RNA from this sanctuary site. This occurred even before release from ART. The anti PD-1 Car T cells persisted for up to 100 days concomitant with the depletion of PD-1+ memory T cells in blood and tissues, resulting in off target immune depletion and a marked increase in SIV replication in extrafollicular portions of lymph nodes, a 2-log higher plasma viremia relative to controls and accelerated disease progression, associated with the acute depletion of CD8+ memory T cells after CAR T expansion. The rapid depletion of TFH sanctuary site in GC while on ART offered the potential to an important advance in the quest for HIV cure if we could develop a second-generation product that had less off target long tern effects. Methods: Towards these ends, we integrated a Hepatitis C virus-derived non-structural protein 3 (NS3)-based ON switch into the intracellular domain of our anti-PD-1 CAR (NS3) T cell platform that can be controlled exogenously by administration of a NS3-specific protease inhibitor Grazoprevir (GZV). Results: Mechanistically, the NS3 domain undergoes autocleavage in absence of GZV, thus abrogating any CAR signaling and CAR T cell function. In vitro, primary T cells expressing the anti-PD-1 CAR NS3 kill PD-1+ cells and secret IFNg in presence of GZV in a dose-dependent manner indicating that the NS3 domain functions as an ON switch. Conclusion: Combined, these data indicate that a drug-controlled anti-PD 1 CAR NS3 is highly functional and can be controlled exogenously through administration or withdrawal of GZV. These data warrant further investigation of drug controlled-anti-PD-1 CAR T cells in ART-treated SIVmac239-infected rhesus macaques to transiently deplete TFH cells to determine their contribution to the latent reservoir. Refractoriness to SIV Reinfection Is Induced by Anti-IL-10/PD-1 Therapy in Rhesus Macaques Susan P Ribeiro 1 , Felipe Ten Caten 1 , Khader Ghneim 1 , Zachary Strongin 2 , Kevin Nguyen 1 , Justin L. Harper 1 , Robert Balderas 3 , Luca Micci 2 , Jeffrey Lifson 4 , Daria Hazuda 5 , Daniel Gorman 6 , Bonnie Howell 2 , Mirko Paiardini 1 , Rafick P. Sekaly 1 1 Emory University, Atlanta, GA, USA, 2 Merck & Co, Inc, Rahway, NJ, USA, 3 BD Biosciences, Franklin Lakes, NJ, USA, 4 Frederick National Laboratory for Cancer Research, Frederick, MD, USA, 5 Generate:Biomedicines, Somerville, MA, USA, 6 Merck & Co, Inc, Palo Alto, CA, USA Background: IL10 and PD1 contribute to HIV reservoir persistence impeding HIV eradication. We hypothesized that the dual blockade of IL10 and PD1 could synergize and lead to SIV-functional cure by triggering intrinsic antiviral immunity in monocytes and CD4 T cells resulting in lack of viral re-seeding post-ATI (Analytical Treatment Interruption) and by boosting innate and adaptive cellular effector function. Methods: 18 rhesus macaques (RMs) infected with SIVmac239 under antiretroviral therapy (ART) for 14 months started at 42 days post-infection received Rhesusized aIL10/aPD1 (Combo n=10) or vehicle (control n=8), administered every 3-weeks between 12 weeks pre-ATI and 14-weeks post-ATI. Animals were followed longitudinally for biospecimen collections. Results: Durable viral load control post-ATI (<1000 cps/mL in 9/10 animals) was observed in the combo treated RMs up to 6 months post-ATI (Fig 1). Post-ATI stable IPDA and 2LTR readouts was also observed in combo-treated animals as compared to the control group. Gene expression analysis revealed significantly elevated restriction factor machinery in the lymph nodes (LN) cells from the

confirm specificity. Tissues and plasma from BLT humanized mice were used to measure IFNα and IFNβ expression as well as ISGs expression after HIV infection through qPCR, western, ELISA, IHC and RNAscope. BLT hu-mice were infected with transmitted/ founder (T/F) virus HIVsuma and following 4 to 5 weeks of viremia were suppressed on cART (FTC+TDF+RAL). Infusions of anti- IFNα/β specific or IgG control antibodies were administered to the HIV fully suppressed animals intraperitoneally over three independent studies. Viral measures (plasma viral load, proviral DNA load), leukocyte subset activation, and HIV antigen specific T cell function were measured by flow cytometry. Analysis was done with Prism software. Results: Neutralizing anti-IFNα or anti-IFNβ-specific antibodies are capable of specifically blocking IFNα or IFNβ signaling in vitro and in vivo. We confirmed IFNα and IFNβ expression in BLT humanized mice, and the increase in ISGs expression after HIV infection. We found the infusions of anti-IFN-beta but not IFNα or IgG control can repeatedly elicit repeated low level viremic blimps under ART. HIV proviral/total DNA size were increased right after the IFNβ-specific antibody treatment supporting viral expression. HIV gag-specific CD8+ T cell frequency (IFN-gamma and CD107a) was also increased on ART after anti-IFNβ specific antibody treatment consistent with HIV antigen expression. However, no change in viral rebound after stopping ART was detected between groups. Conclusion: Blockade of IFN-beta signaling can reactivate HIV reservoir and stimulate HIV-specific T cells under ART in HIV infected BLT humanized mice. The potential use of anti-IFNβ strategies together with direct anti-HIV "kill" or clearance strategies should be further investigated. The figure, table, or graphic for this abstract has been removed. Maura Statzu 1 , Cristina Micali 1 , Tomas Raul Wiche Salinas 1 , Catherine Gurley 1 , Brandon Healy 1 , Diane G. Carnathan 1 , Brandon Keele 2 , Jeffrey Lifson 2 , Gregory M. Laird 3 , David M. Margolis 4 , Mirko Paiardini 1 , Guido Silvestri 1 1 Emory University, Atlanta, GA, USA, 2 Frederick National Laboratory for Cancer Research, Frederick, MD, USA, 3 Accelevir Diagnostics, Baltimore, MD, USA, 4 University of North Carolina at Chapel Hill, Chapel Hill, NC, USA Background: HIV infection cannot be cured despite suppressive antiretroviral therapy (ART) due to a persistent reservoir of latently-infected memory CD4 T cells harboring replication competent virus. Here, we tested the SMACm/cIAP antagonist AZD5582 in combination with the heterodimeric interleukin-15 (hetIL-15), the native form of the cytokine that activates and expands cytotoxic T and NK cells, as a novel approach to prevent the establishment of the virus reservoir during the early stages of infection and/or ART. Methods: 31 rhesus macaques (RMs) were infected with genetically barcoded SIVmac239M and started ART at 2-weeks p.i.. 9 RMs received hetIL-15 via sq injections of escalating dose of 10-40 ug/kg (weekly for 10 weeks) starting 3 days before ART initiation; 9 RMs received iv AZD5582 at 0.1 mg/kg weekly for 10 weeks starting at ART initiation; 9 RMs received the combination of hetIL-15 and AZD5582 over the same experimental phase as outlined above. 4 animals served as treatment-naïve, ART-only controls. Plasma viral loads were measured longitudinally for 16 weeks and the reservoir size was estimated in PBMC and lymph nodes (LN) by IPDA. CD8 T cells, CD4 T cells, and NK cells phenotypes were characterized by flow cytometry. Results: Treatment with AZD5582, alone or in combination with hetIL-15, resulted in slower decline of plasma viremia after ART initiation, and SIV viral load was significantly higher at day14-15-17-21-25-49 and 70p.i. in AZD5582 treated RMs, indicating that treatment with the SMACm was associated with a longer lifespan of the productively infected cells. Levels of peripheral and LN CD4 T cell intact proviral SIV DNA declined in all the groups over the treatment course. The frequency of IPDA+ CD4 T cells tended to be lower in the animals receiving the SMACm, alone or in combination with hetIL-15, with a statistically significant difference at day25p.i in both PBMC and LN. Interestingly, levels of granzyme B at day25p.i. were higher in LN CD8 T cells of the treated animals compared to the controls. Conclusion: Altogether, these findings suggest that treatment with AZD5582, alone or in combination with hetIL-15, may reduce the size of virus reservoir when administered at the time of ART initiation during acute SIV infection, thus suggesting a disruptive effect on the reservoir establishment. These data are consistent with previous work on the latency reversing activity of AZD5582 and provides rationale for further exploring this compound as a curative agent for HIV infection. Treatment With AZD5582 + hetIL-15 Disrupts the Reservoir Establishment in SIV-Infected Macaques

532

Poster Abstracts

534

CROI 2024 142