CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

341 VIRAL RESERVOIR DISRUPTION WITH PANOBINOSTAT AND IFN-Α: FIRST RESULTS Ciputra A. Hartana 1 , Theresa Flynn 2 , Amy Sbrolla 2 , Carina Bannach 1 , Jane Blackmer 1 , Joshua Chevalier 1 , Pilar Garcia Broncano 1 , Xu G. Yu 1 , Rajesh T. Gandhi 2 , Daniel R. Kuritzkes 3 , Mathias Lichterfeld 3 1 Ragon Institute of MGH, MIT and Harvard, Cambridge, MA, USA, 2 Massachusetts General Hospital, Boston, MA, USA, 3 Brigham and Women's Hospital, Boston, MA, USA Background: Reactivation of viral transcription can sensitize viral reservoir cells to immune-mediated killing which may reduce long-term persistence of virally-infected CD4+ T cells in ART-treated individuals. The ACTIVATE study is an ongoing, prospective, randomized, dose-escalation clinical trial in which the histone deacetylase inhibitor (HDACi) panobinostat is administered as a latency-reversing agent in combination with pegylated IFN-α2a as an innate immune modulator. Methods: ART-treated participants were randomized to receive three consecutive doses of 5mg (phase I) or 10mg (phase II) of panobinostat alone (Arm A, n=2 participants in stages I and II each), or in combination with one dose of pegylated IFN-α2a (Arm B, n=6 participants in stages I and II each). Before and at multiple timepoints after study drug administration, cell- associated HIV-1 RNA from the CD4+ T cells were quantified using ddPCR; moreover, innate and adaptive immune responses and acetylated H3 expression were analyzed by flow cytometry. HIV-1 DNA was evaluated using the IPDA. Results: Relative to baseline, the expression of acetylated histone H3 increased 1.5 times (p=0.025) on day 4 after 3 doses of panobinostat, an effect that was most visible in naïve, stem cell memory and central-memory CD4+ T cells. In parallel, a significant increase of HIV-1 gene expression relative to baseline levels was seen for TAR transcripts (p=0.0234) and long-LTR transcripts (p=0.0156) in stage II, but not in stage I. The frequency of activated CD38+ NK cells and NKp30+ NK cells increased significantly at day 4 and day 10 from participants receiving IFN-α2a in stages I and II, which was mostly seen in the cytokine producing (CD16- CD56+), cytotoxic (CD16+ CD56+) and immature (CD16+ CD56-) NK cell subsets. Moreover, the proportion of IL-2-producing HIV-1-specific CD4+ T cells increased during treatment with IFN-α2a, while IFN-g secreting CD4+ T cells were reduced. There were no changes in HIV-1 DNA levels among timepoints and between medication arms in both phases. No unexpected or severe clinical adverse events occurred so far. Conclusion: First results indicate that the medication induces HIV-1 transcription and augments innate and adaptive immune cells. Phase III with 15mg panobinostat administered is ongoing. 342 IAP ANTAGONISM PROMOTES PD-1 BLOCKADE-MEDIATED ELIMINATION OF HIV IN HUMANIZED MICE Michael Bobardt 1 , Joseph Kuo 1 , Udayan Chatterji 1 , Norbert Wiedemann 2 , Gregoire Vuagniaux 2 , Philippe Gallay 1 1 The Scripps Research Institute, La Jolla, CA, USA, 2 Debiopharm, Lausanne, Switzerland Background: The immune checkpoint programmed cell death protein 1 (PD-1) plays a major role in T cell exhaustion in cancer and chronic HIV infection. Inhibitor of apoptosis protein antagonists (IAPa) reverse HIV latency and costimulate T-cells through modulation of NF-kB signaling in vitro. Methods: We asked in this study whether a new IAPa would stimulate the potency of an anti-human PD-1 monoclonal antibody (mAb) to reduce HIV loads in humanized mice. Results: Four weeks of Anti-PD-1 mAb treatment decreased the PD-1+ CD8+ cell population among CD45+ in blood by 22% compared to vehicle, while IAPa co-treatment reduced it by 50%. Anti-PD-1 mAb administration reduced HIV load in blood by 94%with detectable levels in 8 of 8 mice, and addition of the IAPa further enhanced this reduction from 94 to 97%with undetectable levels in 5 of 8 mice. 2 weeks after drug treatment interruption, Anti-PD-1 mAb administration had reduced HIV loads in CD4+ cells also in all tissues analyzed compared to vehicle, including spleen (5.6 to 2 log in viral RNA copies), lymph nodes (5.6 to 1.1 log in viral RNA copies), liver (5.4 to 1.6 log in viral RNA copies), lung (5.6 to 2 log in viral RNA copies) and thymic organoid (5.5 to 1.2 log in viral RNA copies). IAPa further enhanced the anti-PD-1-mediated reduction of HIV tissue loads achieving a >5 log reduction in all tissues analyzed, notably with undetectable levels in some individual organs; spleen (5.6 to 0.2 log in viral RNA copies), lymph nodes (5.6 to 0.2 log in viral RNA copies), liver (5.4 to 0.3 log in viral RNA copies), lung (5.6 to 0.2 log in viral RNA copies) and thymic organoid

(5.5 to 0.1 log in viral RNA copies). Following the 4 weeks of in vivo treatments, ex vivo anti-CD3/CD28 stimulation increased the ability to activate CD8+ T cells in infected mice having received in vivo anti-PD-1 treatment by 7.9-fold (5 to 39.6%), and an additional increase by 1.7-fold in mice having received IAPa co-treatment (39.6 to 67.3%). Conclusion: These findings demonstrate for the first time that an IAPa greatly enhances the effects of an immune checkpoint inhibitor on antiviral immunity resulting in undetectable HIV titers in blood and organs of humanized mice. This suggests that the combination of two distinct classes of immunomodulatory agents constitutes a promising immunotherapeutic approach to cure HIV. 343 IMPACT OF GS-986, PGT121 AND N6-LS ON CNS IMMUNE ACTIVATION IN SHIV-INFECTED MACAQUES Denise C. Hsu 1 , Decha Silsorn 2 , Rawiwan Imerbsin 2 , Amarendra Pegu 3 , Dutsadee Inthawong 2 , Jumpol Sopanaporn 2 , Alexandra Schuetz 1 , James Demarest 4 , Merlin L. Robb 5 , John R. Mascola 3 , Romas Geleziunas 6 , Richard A. Koup 3 , Dan Barouch 7 , Nelson L. Michael 5 , Sandhya Vasan 5 1 US Military HIV Research Program in Thailand, Bangkok, Thailand, 2 Armed Forces Research Institute of Medical Sciences in Bangkok, Bangkok, Thailand, 3 NIH, Bethesda, MD, USA, 4 ViiV Healthcare, Research Triangle Park, NC, USA, 5 US Military HIV Research Program, Silver Spring, MD, USA, 6 Gilead Sciences, Inc, Foster City, CA, USA, 7 Beth Israel Deaconess Medical Center, Boston, MA, USA Background: Kick and kill strategies using TLR-7 agonist and broadly neutralizing antibodies (bNab) have shown promise in non-human primates, but effects on the central nervous system (CNS) have not been evaluated. Methods: Rhesus macaques (n=16) were intrarectally inoculated with SHIV- 1157ipd3N4 at wk0 and initiated on ART (PMPA, FTC, DTG) on Day14. Active group (n=8) received GS-986 every 2 weeks fromwk14 and intravenous N6-LS and PGT121 every 2 weeks fromwk24. The development of anti-drug antibodies limited number of bNAb administrations. Active group animals received at least 7, 2 and 2 doses of GS-986, PGT121 and N6-LS, respectively. ART was ceased 2 weeks after plasma levels of bNAbs <0.25ug/mL. Control animals (n=8) received intravenous saline and ART was ceased at wk40. Plasma and cerebral spinal fluid (CSF) SHIV RNA levels were measured by PCR and soluble markers of immune activation by multiplex assay using Luminex. Results: Median wk2 (pre-ART) plasma and CSF SHIV RNA was 5.7 (range 4.1-6.8) and 3.1 (range 2.2-4.2) log 10 copies/mL respectively. After ART initiation on Day14, plasma SHIV RNA was undetectable in all animals by wk8 and remained undetectable until ART interruption. CSF SHIV RNA was also undetectable in all animals during GS-986 dosing (wk24). Median time to viral rebound was 6 weeks in active arm and 3 weeks in control arm (p=0.024). At 12 weeks post rebound, median plasma SHIV RNA was 1.2 (range 1.0-2.2) and 2.1 (range 1.0-2.8) log 10 copies/mL in the active and control arm respectively. CSF SHIV RNA was only detectable at low levels in 1 active and 1 control arm animal. Longitudinal CSF samples from the active group showed significant increases in IL-15 (p=0.008), MCP-1 (p=0.008), IL-8 (p=0.008), IL-1RA (p=0.016), IL-2 (p=0.031), and G-CSF (p=0.008) at wk2 when compared to pre-infection, that decreased after ART initiation to pre-infection levels. Importantly, levels remained similar post GS-986 administration at wk24 and post bNabs prior to ART interruption. At 12 weeks post rebound, CSF IL-2 (p=0.031), and G-CSF (p=0.008) were increased relative to pre-infection levels. Conclusion: Administration of GS-986, PGT121 and N6-LS did not increase SHIV RNA or markers of immune activation in CSF, suggesting that this strategy may be pursued in humans without impacting CNS activation.

Poster Abstracts

CROI 2020 118