CROI 2025 Abstract eBook

Abstract eBook

Poster Abstracts

724

In Vitro Resistance Profile for GS-1720, a Potent Once-Weekly Oral InSTI in Clinical Development Derek Hansen, Matthew R. Hendricks, Joseph Campbell, Brie Falkard, Gregg Schwarzwalder, Ana Z. Gonzalez, Tomas Cihlar, Stephen R. Yant Gilead Sciences, Inc, Foster City, CA, USA Background: Integrase strand transfer inhibitors (INSTIs) are the cornerstone of current HIV therapies, but their oral dosing remains limited to once-daily administration. Herein we describe the in vitro resistance profile for GS-1720, a potent and selective investigational INSTI with Phase 1 clinical pharmacokinetic, safety and antiviral efficacy data supportive of once-weekly oral dosing. Methods: GS-1720 activity against HIV-1 reporter viruses containing integrase (IN) substitutions associated with INSTI class resistance was assessed at Monogram Biosciences. In vitro resistance barrier was studied by serial passage of HIV-1 IIIb- and HXB2-infected MT-2 cells in the presence of increasing concentrations of GS-1720. GS-1720 was also evaluated in a 35-day fixed-dose viral breakthrough assay in HIV-1 BaL-infected human PBMCs. Emergent viral variants were characterized genotypically and phenotypically. Results: GS-1720 retained potent antiviral activity, defined as an EC50 fold change (FC) < 10, against 91% of INSTI-resistant isolates tested (n=55), as compared to 96% and 75% for bictegravir (BIC) and cabotegravir, respectively. Reduced virus susceptibility to GS-1720 (FC>10) required 3 IN mutations (G140A/S+Q148H/K+L74I, T97A or E138A/K) previously defined as primary or secondary drivers of INSTI class resistance. In vitro dose escalation resistance selections with GS-1720 against IIIb virus progressed at a rate comparable to BIC and significantly slower than elvitegravir (EVG), with the emergence of IN variants S153F (GS-1720), S153S/Y (BIC) and P145S (EVG) after >300, 140 and 119 days in culture, respectively. The S153Y variant also emerged in two dose escalation selections with HXB2 virus at GS-1720 concentrations equal to 16-fold its EC50 after 112 days in culture. Site-directed mutants of S153 variants showed high susceptibility to GS-1720 and all 5 approved INSTIs (FC<2.7). In a fixed-dose viral breakthrough assay, clinically relevant concentrations of GS-1720 or BIC prevented viral breakthrough while EVG and raltegravir, when tested at their equivalent clinical trough plasma concentrations, showed breakthrough with resistance in 45% and 18% of PBMC cultures (n=44-48/condition), respectively. Conclusions: GS-1720 exhibits an in vitro resistance profile comparable to the best-in-class INSTI bictegravir, including high barrier to resistance and minimal cross-resistance within the INSTI class. These observations support the ongoing clinical development of GS-1720 as a novel once-weekly oral INSTI. Effect of SNG001, Inhaled IFN-β1a, on SARS-CoV-2 Diversity and Evolution Gregory E. Edelstein 1 , Tatum N. Sass 1 , Rinki Deo 1 , Prasanna Jagannathan 2 , Kara Chew 3 , Mark J. Giganti 4 , Michael Hughes 4 , Carlee Moser 4 , Joseph J. Eron 5 , Judith Currier 3 , Upinder Singh 6 , Davey Smith 7 , William A. Fischer II 5 , Manish C. Choudhary 1 , Jonathan Li 1 , for the ACTIV-2/A5401 Study Team 1 Brigham and Women's Hospital, Boston, MA, USA, 2 Stanford University, Stanford, CA, USA, 3 University of California Los Angeles, Los Angeles, CA, USA, 4 Harvard TH Chan School of Public Health, Boston, MA, USA, 5 University of North Carolina at Chapel Hill, Chapel Hill, NC, USA, 6 University of Iowa, Iowa City, IA, USA, 7 University of California San Diego, La Jolla, CA, USA Background: Interferon resistance has been implicated in SARS-CoV-2 escape from innate immune responses, but interferon’s specific impact on viral evolution is unknown. SNG001 is an inhaled interferon-β1a treatment for COVID-19 which was evaluated in the ACTIV-2/A5401 randomized placebo controlled trial. Although SNG001 did not accelerate viral decay, there was a trend towards fewer hospitalizations in the treated group. Sequence analysis of samples collected in ACTIV-2 provide a rare opportunity to investigate SARS CoV-2 evolution and diversity in response to exogenous interferon treatment. Methods: The study population included 220 adult outpatients with COVID-19 enrolled in ACTIV-2/A5401 and randomized to receive either SNG001 (n=110) or placebo (n=110). Participants collected nasal swabs on days 0, 3, 7, 14, and 28 for SARS-CoV-2 viral load (VL) testing. Whole genome sequencing was performed on specimens with VLs ≥2 log 10 copies/mL. Viral sequences were analyzed for diversity using average pairwise distance (APD) and for the emergence of amino acid polymorphisms by gene and treatment arm. Results were analyzed using Wilcoxon Rank Sum Tests. Results: Overall, 187 participants (SNG001 n=99; placebo n=88) had a quantifiable baseline VL which could be used to generate a variant call. The Alpha, Delta, and other variants comprised 28%, 40%, and 32% of infections, respectively. The viral diversity analyses included 136 participants with

which affect the efficacy of all known INSTIs. We found that viruses harboring the complex IN mutant E138K/G140A/Q148R led to the largest decrease in EC 50 compared to WT viruses. Both the available INSTIs, and new compounds currently in pre-clinical trials, are less effective against the complex IN mutant E138K/G140A/Q148R, indicating that, with the increasing therapeutic use of INSTIs, the prevalence of this variant is likely to increase. Notably, Cabotegravir (CAB), which was approved by the US Food and Drug Administration in 2021 as a long-acting injectable, potently inhibits WT IN, but is particularly susceptible to this complex IN mutant. Using cryo-electron microscopy (cryo-EM), we solved new high-resolution structures of WT and drug-resistant HIV-1 IN assemblies with CAB bound. We also performed molecular dynamics-based free energy calculations to delineate the mechanism of resistance. Our study shows that the E138K/G140A/Q148R mutations disrupt the drug's efficacy by altering the DDE-Mg ( D 64+ D 116+ E 152- Mg ) network at the active site of integrase. The δ+ charge on Arg 148 forms a salt bridge with D64, E152, and D116. The mutations cause rearrangements in the active site that reduce the binding of CAB and, by extension, other INSTIs. Conclusions: Our combined virology, structural biology, and molecular dynamics analyses provides a detailed mechanistic understanding of CAB/INSTI resistance, yielding a foundation for the development of next-generation INSTIs capable of overcoming problematic complex resistant IN variants. The figure, table, or graphic for this abstract has been removed. The Barrier to Escape From Broadly Neutralizing Antibodies Varies Between Different HIV-1 Isolates Alex C. Stabell 1 , Debby Park 2 , Christy L. Lavine 3 , Sebastian Mejia Espinosa 4 , Viren Baharani 5 , Songhee Lee 5 , Michel Nussenzweig 5 , Marina Caskey 5 , Michael S. Seaman 3 , Paul Bieniasz 5 , Theodora Hatziioannou 5 1 Weill Cornell Medicine, New York, NY, USA, 2 University of Pennsylvania, Philadelphia, PA, USA, 3 Beth Israel Deaconess Medical Center, Boston, MA, USA, 4 Harvard University, Cambridge, MA, USA, 5 The Rockefeller University, New York, NY, USA Background: Broadly neutralizing antibodies (bNAbs) against HIV-1 have re-invigorated research into antibody therapies and vaccination strategies for prevention, treatment, and cure of HIV-1 because of their unique ability to neutralize diverse HIV-1 strains. While many HIV-1 isolates are initially susceptible to bNAbs, their ability to acquire resistance is unknown. Understanding the evolutionary barriers to bNAb resistance will inform the feasibility of using bNAbs as HIV-1 prevention, treatment, and cure. Methods: We developed a cell-culture based system to understand the evolutionary barriers to escape from neutralization by the bNAb 3BNC117. We performed numerous parallel selection experiments to capture the array of escape pathways available to a range of HIV-1 strains. We isolated 375 viruses that replicated in the presence of 3BNC117 from 16 starting viral strains. Each viral variant isolated by our assay was confirmed to be resistant to 3BNC117 and sequenced using Illumina-based short read sequencing. We developed a statistical model to distinguish mutations conferring bNAb resistance from passanger mutations and confirmed these predictions using a pseudotype neutralization assay. We performed this procedure on numerous group M HIV-1 isolates of various subtypes that capture the worldwide diversity of HIV-1. Results: 3BNC117-resistant variants were isolated from all HIV-1 strains tested. Some viral strains demonstrated many different mutational escape pathways, while others demonstrated only a single dominant escape mutation that was selected in several different resistant isolates. The vast majority of HIV-1 strains required only a single nucleotide substitution to escape 3BNC117 neutralization. These resistance mutations existed in the viral quasispecies prior to selection at variable frequencies, ranging from 1 in 10 2 viruses (common) to less than 1 in 10 6 viruses (rare). Each resistance mutation was tested in a pseudotype assay against a panel of bNAbs, demonstrating that some, but not all 3BNC117 escape mutations were able to confer resistance to additional CD4 binding site bNAbs. Conclusions: This study demonstrates that various pathways exist for escape from a CD4 binding site bNAb 3BNC117 among divergent HIV-1 isolates. Understanding the pathways enabling escape from bNAb neutralization as well as the frequency at which they are maintained in the population will help to tailor these antibody-based therapies and inform the design of vaccines that aim to elicit bNAb-like antibodies.

Poster Abstracts

723

725

CROI 2025 209