CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

452 PHARMACOKINETICS OF RUXOLITINIB WITH ART IN HIV-SUPPRESSED INDIVIDUALS (ACTG # A5336) Selwyn Hurwitz 1 , Sijia Tao 1 , Yong Jiang 1 , Christina Gavegnano 1 , Charles W. Flexner 2 , Randall Tressler 3 , Athe Tsibris 4 , Steven G. Deeks 5 , Carlos del Rio 1 , Edgar T. Overton 6 , Jeffrey J. Lennox 1 , Vincent C. Marconi 1 , Raymond F. Schinazi 1 1 Emory University, Atlanta, GA, USA, 2 Johns Hopkins University, Baltimore, MD, USA, 3 NIH, Bethesda, MD, USA, 4 Harvard University, Cambridge, MA, USA, 5 University of California San Francisco, San Francisco, CA, USA, 6 University of Alabama at Birmingham, Birmingham, AL, USA Background: Ruxolitinib is an FDA-approved Janus kinase (JAK 1/2) inhibitor (myelofibrosis, polycythemia vera) that blocks key cytokines involved in HIV persistence including IL1, 6, 7 and 15. In A5336, low dose ruxolitinib (10 mg bid) was administered to healthy people living with HIV (PLWH) on antiretroviral therapy (ART) for 5 wk to investigate safety and to reduce ongoing inflammation that persists even with virologic suppression. Because ruxolitinib is metabolized via the cytochrome P450 system. Analysis sought to model variability of ruxolitinib pharmacokinetics (PK) between participants (inter individual variability, IIV) and assess PK interactions between ruxolitinib and ART. Methods: Steady-state plasma concentrations of ruxolitinib and coadministered ART were drawn on wk 1 and 4/5 and assayed by liquid chromatography and tandemmass spectrometry (LC-MS/MS). Population PK models were fitted using NONMEMÔ 7.4. Parameter distributions were assumed log-normal, and residuals having an additive and residual component. IIV of Parameter and the variations in fraction of oral dose absorbed (between occasion variability of F) were estimated. Models converged to >3 decimals using the FOCE (first-order conditional estimation) with interaction method and were evaluated using statistical and graphical methods. Results: No clinically relevant adverse events were observed across participants (33 male, 7 female), and HIV suppression was maintained. Ruxolitinib plasma concentrations versus time profiles from 39 and 38 participants on wk 1 and wk 4/5, respectively, were modeled. The PK profiles were adequately described using an open 2-compartment model with first-order absorption and elimination, and parameters were similar to reports in healthy volunteers and other indications: Distribution volumes V1/F = 61.83 L, 30.9% and V2/F = 2.36 L, 70.1% (normalized by body weight, mean 91.5 kg, IQ range 76.75-91.5 kg); Compartment clearance values were CL10/F = 14.47, 33.8% and CL12/F =4.84 L/hr; Absorption rate constant Ka = 4.96, 70.1%, and there was a 23% BOV in F. Area under the curve (AUC, dose/CL12) distributions were similar on wk 1 and wk 4/5. Overall, concentrations of ART were consistent with those reported in population PK studies without ruxolitinib. Conclusion: These data suggest that ruxolitinib can be safely administered to ART suppressed PLWH without adverse consequences regarding ruxolitinib or ART plasma levels, and variability of ruxolitinib plasma concentrations is similar to other populations. 453 INFILTRATION OF bNAb VRC01 INTO THE CEREBROSPINAL FLUID IN HUMANS IN THE RV397 STUDY Madhu Prabhakaran 1 , Sandeep Narpala 1 , Lucio Gama 1 , Donn J.Colby 2 , Phillip Chan 3 , Carlo Sacdalan 3 , Khunthalee Benjapornpong 3 , Jintanat Ananworanich 2 , Nittaya Phanuphak 3 , Suteeraporn Pinyakorn 2 , Trevor A. Crowell 2 , Serena S. Spudich 4 , Adrian B. McDermott 1 , for the RV397 Study Team 1 Vaccine Research Center, NIAID, Bethesda, MD, USA, 2 US Military HIV Research Program, Silver Spring, MD, USA, 3 Thai Red Cross AIDS Research Center, Bangkok, Thailand, 4 Yale University, New Haven, CT, USA Background: HIV may persist in the central nervous system (CNS) despite antiretroviral therapy (ART), creating a barrier to HIV eradication. Novel strategies to reduce the latent HIV reservoir may need to cross the blood brain barrier (BBB) into the cerebrospinal fluid (CSF). Targeting the CD4 binding site, VRC01 is a broadly neutralizing antibody (bNab) capable of potently neutralizing over 90% of HIV-1 strains. Methods: The RV397 study conducted in Bangkok, Thailand was a randomized, double-blind, placebo-controlled trial that randomized participants who initiated suppressive ART during acute HIV infection to receive VRC01 40mg/kg or placebo intravenously every 3 weeks during analytic interruption of ART (ATI). CSF samples were collected at two time points from 3 participants who received VRC01: pre-infusion and 2-4 days after first detectable plasma viral load. VRC01 levels were quantified using a standardized sensitive Singulex single molecule counting technology with a lower limit of quantitation (LLOQ) of 50pg/ml for

VRC01. CSF VRC01 concentration was compared to concurrent plasma level for each participant. Results: Three males, aged 18-47 years, initiated ART during acute HIV (Fiebig stages 1 or 2) and were on ART for at least 28 months before ATI. Pre-infusion, pre-ATI, CSF HIV RNA was <80 copies/ml, CSF WBC <2 cells/µl, CSF protein <38 mg/dL, blood HIV RNA <20 copies/ml and CD4 >400 cells/µL. Post ATI, post-VRC01 infusion CSF HIV RNA was <80 copies/ml, CSF WBC <2 cells/µl, CSF protein <43 mg/dL, blood HIV RNA 418-1789 copies/ml and CD4 T >400 cells/ µl. VRC01 levels in CSF were below LLOQ preinfusion and ranged between 0.35 – 0.75ug/ml post-infusion (see Table). Concurrent VRC01 levels in plasma were 200-600ug/ml, indicating a 100-1000 fold lower penetration into the CNS compartment. Conclusion: We report here the successful quantification of the bNab VRC01 in the CSF from 3 persons living with HIV. The bioavailability of this potent and broad HIV monoclonal antibody in the CNS is critical considering that the CNS can facilitate the generation of resistant HIV quasi-species that are distinct from virus in systemic circulation. These results thus serve to inform the design of immunotherapies to target HIV infection in the CNS.

Poster Abstracts

454 BICTEGRAVIR/FTC/TAF CSF DIFFUSION IN HIV-INFECTED PATIENTS WITH CNS IMPAIRMENT Antoine Chéret 1 , Thibaut Gelé 1 , Alicia Castro Gordon 1 , Valérie Furlan 1 , Coralie Pallier 1 , Pierre-Hadrien Becker 1 , Pilartxo Catalan 1 , Cécile Goujard 1 , Anne-Marie Taburet 1 , Jacques Gasnault 1 , Hélène Gouget 2 , Aurélie Barrail-Tran 1 1 Assistance Publique – Hôpitaux de Paris, Paris, France, 2 INSERM, Le Kremlin- Bicetre, France Background: The penetration of antiretroviral drugs in deep compartments, like the central nervous system (CNS), is a crucial part of strategies towards HIV cure. This study aimed to estimate cerebrospinal fluid (CSF) diffusion of bictegravir (BIC), that has high protein binding which could limit diffusion, emtricitabine (FTC) and tenofovir alafenamide (TAF) in patients with HIV-related CNS impairment (HCI) enrolled in a real-life observational study. Methods: Patients (pts) with HCI on treatment by an optimized antiretroviral therapy including BIC since at least 1 month were enrolled between 2019 January and February (NeuroHIV Rehabilitation Care Unit, AP-HP, Bicêtre Hospital, France). Blood and CSF samples were collected simultaneously in the setting of routine care. Plasma and CSF HIV RNA were quantified by PCR (Abbott Realtime®, threshold=40 copies/mL). Total plasma (Tot) and CSF BIC/ FTC/tenofovir (TNF) concentrations and unbound plasma (U) BIC concentrations, separated by ultrafiltration (Centrifree devices, cutoff, 30 kDa; Millipore), were measured by quality controls validated assays (LC-MS/MS). The albumin quotient (QA), calculated as the ratio of CSF to plasma albumin, was used to evaluate the blood-brain barrier (BBB) function. All numerical variables were expressed as median (IQR). Results: Twelve pts (6 females) were enrolled. Age was 44 (12) years. HCI were: progressive multifocal leukoencephalopathy (PML, n=7), cerebral toxoplasmosis (CT) (n=3), CT combined with HIV encephalitis (n=1) and VZV meningoencephalitis (n=1). Backbone therapy co-administered to BIC was: TAF+FTC (n=10) or TAF+FTC+Maraviroc (n=2). Plasma HIV RNA was undetectable in 10 (83%) pts and <3 log 10 copies/mL in others. Two (17%) pts had a detectable CSF viral load (1.7 and 1.9 log 10 copies/mL). All concentrations and ratios are shown in table below. There are correlations between CSF and Tot concentrations for BIC and FTC (p=.008 for BIC and p=0.002 for FTC) and between CSF and U concentrations for BIC (p=.049). The median QA was 5.5 (1.8); 1 (8%) patient had a damaged BBB, but not related with a higher CSF BIC/ FTC/TNF diffusion. Conclusion: Total plasma concentrations remained as previously reported. Almost all CSF concentrations were above the in vitro 50% inhibitory concentration (IC 50 ). BIC with FTC/TAF backbone should be effective to target

CROI 2020 159