CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

1 Gladstone Institutes, San Francisco, CA, USA, 2 University of California San Francisco, San Francisco, CA, USA Background: Reducing the size of the latent HIV reservoir and controlling subsequent viral rebound by immune engineering could lead to a sustained viral remission in HIV-infected individuals in the absence of ART. CTLs could reduce the size of the reservoir by recognizing and killing reactivated reservoir cells. However, cellular exhaustion and the presence of CTL-resistant viruses may undermine their effectiveness. We have tested a new approach to reservoir reduction where convertibleCAR-T cells (cCAR-Ts) programmed with multiple HIV-specific broadly neutralizing antibodies (bNAbs) are deployed. Methods: cCAR-Ts utilize a mutated, inert form of the NKG2D receptor. Orthogonal MIC ligands that bind to inert NKG2D but not wild-type NKG2D are fused to antibodies to generate bispecific MicAbodies for directing cCAR-T targeting and activation. cCAR-Ts can therefore be readily redirected by altering the antibody component of the MicAbody and furthermore, MicAbodies can be multiplexed. 4 bNAbs were engineered as MicAbodies and tested for their ability to kill tonsil, spleen, or blood cells infected with GFP-tagged R5 or X4-tropic or transmitted/founder viruses. Specificity of infected cell killing was monitored by loss of GFP+ vs GFP- cells. Reactivated CD4 T cells from HIV- infected individuals on ART were assayed for loss of cell-associated viral RNA in the presence cCAR-Ts either armed or not armed with bNAbs. The platformwas checked in vivo, in NSG mice model of cancer, by measuring size reduction of cancer tumors. Results: In the presence of bNAb-MicAbodies, CD8 cCAR-Ts effectively killed HIV-infected, but not uninfected, cells from tonsil, spleen and blood. Killing was strictly dependent on the presence of bNAb-MicAbodies targeting HIV Env. Multiplexing of four MicAbodies increased the breadth of killing. cCAR-T cells also reduced by more than half the inducible reservoir present in blood of HIV-infected individuals on ART. Administration of cCAR-Ts cells in a mice cancer model, demonstrated highly effective in vivo killing. Conclusion: An attractive feature of cCAR-Ts is that it is a modular platform that not only allows for multiplexing of MicAbodies, but also targeted delivery of kill switches if needed or cytokines for cCAR-T rejuvenation. This platform could be an important tool for reducing and controlling the size of the latent HIV reservoir. 337 CAR-T CELLS AT 15 YEARS: PERSISTENCE OF CD4-ZETA TRANSGENE AND EFFECT ON RESERVOIR Naomi E. Aronson 1 , James L. Riley 2 , Linda Jagodzinski 3 , Sodsai Tovanabutra 3 , Denise C. Hsu 4 , Lydie Trautmann 3 , Anwar E. Ahmed 1 , David Wallace5, Simon Lacey 2 , Bruce L. Levine 2 , Carl H. June 2 , Julie Ake 3 , Wendy B. Bernstein 5 1 Uniformed Services University of the Health Sciences, Bethesda, MD, USA, 2 University of Pennsylvania, Philadelphia, PA, USA, 3 Walter Reed Army Institute of Research, Silver Spring, MD, USA, 4 Armed Forces Research Institute of Medical Sciences in Bangkok, Bangkok, Thailand, 5 Walter Reed National Military Medical Center, Bethesda, MD, USA Background: Despite effective antiretroviral therapy, cellular reservoirs of HIV persist. CD4ζ is a chimeric T cell receptor with the intracellular and transmembrane domain of CD4 linked to the zeta signaling chain of the CD3 T cell receptor. The long term persistence of this CAR-T cell therapy was previously estimated. Methods: Fifteen individuals were randomized to 3 groups (cells, IL-2, cells + IL-2) to receive a single infusion of 5-9 x 109 autologous CD4ζ gene modified T cells ± subcutaneous IL-2 at 1.2 million IU/m 2 for 56 days. Inclusion criteria included CD4≥200, viral load<50, stable HAART for ≥8 weeks. Pheresis and rectal biopsy were performed at baseline and at 13-15 years follow up. Real- time PCR was used to detect and measure the CD4ζ transgene and the HIV-1 gag gene in PBMCs and rectal tissues. RNAscope using HIV-1 Clade B probe was performed on formalin fixed rectal tissue at long term follow-up. Total and integrated HIV DNA were measured in PBMCs using a highly sensitive nested PCR assay. Mixed models and ANCOVA were used to assess the effects of treatment arms on CD4, CD4%, CD4:CD8, total and integrated HIV DNA over time. Results: Fifteen persons enrolled (mean age 38.4 ± 7.9 years) and thirteen individuals,11 males and 2 females, completed the long term follow up (LTFU). Race/ethnicity of the participants included one Asian, four Blacks, two Hispanics and six Caucasians. The median CD4 count on enrollment was 821 (IL-2), 712 (cells) and 822 (cells + IL-2), p=0.468. At LTFU median CD4 counts were 779, 720 and 1047 respectively, p=0.376. HIV viral loads were suppressed except in one nonadherent subject at LTFU. No differences by race or sex were seen. There was

335 ABX464 DECREASES THE TOTAL HIV RESERVOIR AND HIV TRANSCRIPTION INITIATION IN VIVO

Sara Moron-Lopez 1 , Silvia Bernal 2 , Jean-Marc Steens 3 , Joseph K. Wong 4 , Javier Martinez-Picado 2 , Steven A. Yukl 4 1 University of California San Francisco, San Francisco, CA, USA, 2 IrsiCaixa Institute for AIDS Research, Badalona, Spain, 3 ABIVAX, Paris, France, 4 San Francisco VA Medical Center, San Francisco, CA, USA Background: Antiretroviral treatment (ART) intensification and disruption of latent HIV infection (reversal or silencing) have been suggested as strategies to eradicate HIV. ABX464 (AbiVax) is a novel antiviral that binds to the cap binding complex, interfering with splicing and Rev-mediated export of newly transcribed HIV RNA. ABX464 has been shown to inhibit HIV RNA biogenesis in vitro and delayed viral rebound in a humanized mouse model. We investigated the effect of ABX464 on the HIV transcription profile and total and intact HIV DNA in circulating CD4+ T cells from ART-suppressed participants enrolled in the ABIVAX-005 clinical trial (NCT02990325). Methods: Eleven participants on suppressive ART were treated daily with 150mg of ABX464 for 4 weeks. Peripheral CD4+ T cells from nine study participants were available for HIV transcription profile and reservoir size analysis. Total HIV DNA, intact HIV DNA (IPDA), and Read-through, total/ initiated, 5’elongated, unspliced, polyadenylated and multiply-spliced HIV transcripts were quantified at weeks 0, 4 and 8 using ddPCR. Results: We observed a significant decrease in the total HIV DNA (p=0.008, median fold-change=0.8) and a lower median level of intact HIV DNA (p=n.s., median fold-change=0.8) after ABX464 treatment (wk0 vs. wk4). However, intact HIV DNA increased significantly (p=0.008, fold-change=1.6) after ABX464 discontinuation (wk4 vs. wk8). After 4 weeks of ABX464 treatment, we observed a decrease in total initiated HIV RNA per million CD4+ T cells and per provirus (HIV RNA/HIV DNA) (p=0.05, median fold-change=0.7; p=0.004, median fold-change=0.5, respectively), a trend towards a decrease in the 5’elongated HIV RNA per provirus (p=0.07, median fold-change=0.5), and a lower median level of unspliced HIV RNA (p=n.s., median fold-change=0.6), but no decrease in polyadenylated or multiply-spliced HIV RNA. However, 5’elongated HIV RNA per million CD4+ T cells increased significantly (p=0.04, fold-change=1.4) after ABX464 discontinuation (wk4 vs. wk8). Conclusion: In this substudy, ABX464 had a dual effect of decreasing total HIV DNA (and possibly intact proviruses) and decreasing the amount of HIV transcription per provirus, although these changes were reversed after drug discontinuation. Our data suggest that ABX464 acts as an ART intensifier in vivo. To further characterize its specific mechanism of inhibiting HIV transcription, long-term administration of ABX464 in a larger cohort should be studied. 336 ATTACKING LATENT HIV WITH CONVERTIBLE CAR-T CELLS, A MODULAR KILLING PLATFORM Eytan Herzig 1 , Kaman Chan Kim 2 , Thomas A. Packard 1 , Noam Vardi 1 , Roland Schwarzer 1 , Andrea Gramatica 1 , Steven G. Deeks 2 , Steven R. Williams 2 , Kyle Landgraf 2 , Nigel Killeen 2 , David W. Martin 2 , Leor Weinberger 1 , Warner C. Greene 1

Poster Abstracts

CROI 2020 116