CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

540 IDENTIFICATION OF ARV-RESISTANCE MUTATIONS OUTSIDE OF THE DRUG-TARGET GENE Phuong Pham 1 , Rachel Van Duyne 1 , Eric O. Freed 2 1 NIH, Frederick, MD, USA, 2 National Cancer Institute, Frederick, MD, USA Background: Resistance to antiretrovirals (ARVs) continues to impair the overall quality of life for some HIV-infected individuals, despite the effectiveness of combination antiretroviral therapy (cART). The goal of this study was to explore the ability of HIV-1 to escape inhibition by ARVs in vitro by acquiring resistance mutations outside of the drug-target gene. Methods: We propagated HIV-1 in T-cell lines and measured virus replication kinetics in the presence or absence of low (sub-IC50) concentrations of ARVs, testing at least one representative of each class of inhibitor. We selected for viral escape mutants exhibiting at least partial resistance to ARVs as indicated by efficient replication in the presence of the inhibitors. A number of analyses were then performed to validate the ability of the selected mutations to confer ARV resistance. Results: Long-term passage of wild-type (WT) virus in the presence of ARVs led to the selection of ARV-escape mutants lacking changes in the target gene, but instead containing substitutions in the envelope (Env) glycoprotein and occasionally in Vpu. We have now identified a panel of partially ARV-resistant NL4-3 Env mutants that arose in the presence of protease, reverse transcriptase (RT), and integrase inhibitors. Mutations were selected in the context of two different T-cell lines, Jurkat and CEM12D7, that favor cell-cell and cell-free transmission, respectively. Remarkably, the same ARV-resistant Env mutant was selected in both cell lines. We extended our analyses to a transmitted-founder, subtype C virus, CH185_TF, which acquired a mutation in Env when propagated in the presence of Dolutegravir (DTG). These data demonstrate that ARV- resistant Env mutants arise in the context of three different T-cell lines and two viral subtypes with different coreceptor tropism. Finally, we found that several of the Env mutation positions are highly conserved within and across HIV-1 clades but that these mutations do appear in patient isolates. Conclusion: These results demonstrate that mutations in Env can contribute to HIV drug resistance in vitro. A combination of in vitro selections and in vivo analyses is ongoing and may establish a role for Env mutations in ARV resistance in patients and help guide the development of more effective therapies. 541 EMERGENCE OF GAG MUTATION, A364V, IDENTIFIED AS THE KEY IN VITRO RESISTANCE MUTATION Charlene McDanal 1 , Pauline Schipper 2 , Monique Nijhuis 2 , Jerry Jeffrey 1 , Brian A. Johns 1 1 ViiV Healthcare, Research Triangle Park, NC, USA, 2 University Medical Center Utrecht, Utrecht, Netherlands Background: The in vitro virology profile has been previously presented (Jeffrey et al, CROI 2015). These data demonstrated that GSK2838232, a second-generation HIV maturation inhibitor, has a broad spectrum antiviral profile against viruses from various clades and viruses resistant to marketed antiretrovirals. The current study was aimed to identify drug resistance mutations via in vitro resistance passage. Methods: Recombinant viruses containing the gag/protease fragments of two representative protease-treated HIV-infected individuals susceptible to GSK2838232 (R6877: IC50 1.8nM and R7104: 0.9nM) and laboratory strain NL4-3 (IC50 1.5nM) were serially passaged in SupT1 cells. Experiments were started with drug concentrations around the IC50 value and gradually increased in each passage (all experiments were done in 5-fold increments). Results: After 5 passages, at GSK2838232 concentrations 10-20 fold over the initial IC50 (~24nM) for inhibiting the parent virus, gag and protease was fully sequenced. Remarkably, in all experiments the gag A364V amino acid change at the p1’ site in the CA/P2 cleavage site was observed. A site-direct mutant containing the A364V was generated in the NL4-3 parental virus and demonstrated a high level of resistance to GSK2838232 (>400nM). Lastly, the frequency of A364V among HIV gag sequences in the Los Alamos National Labs HIV database was investigated and found to be less than 0.1% of these sequences. Conclusion: The resistance profile of GSK2838232 is consistent with previous maturation inhibitors. Based on the infrequent presence of the A364V mutation, pre-existing resistance in an HIV-positive human patient population is expected to be low. As GSK2838232 progresses through clinical development, these in vitro resistance data will help decipher the genotypic and phenotypic observations from those clinical studies.

Poster Abstracts

539 A COMPARATIVE EVALUATION OF HIV-1 CAPSID INHIBITOR SUSCEPTIBILITY

Alicia Newton , Charity Ting, Lan Trinh, Christos Petropoulos, Wei Huang Monogram BioSciences, San Francisco, CA, USA Background: Inhibitors that target HIV capsid assembly and virion maturation represent a promising new class of antiretroviral compounds. In this study, we used an enhanced cell-based infectivity assay, based on the assembly of resistance test vectors (RTV), to evaluate the susceptibility of patient isolates and gag gene site-directed mutants to several maturation and assembly inhibitors. Methods: Gag-protease coding regions from 111 HIV patient isolates, previously submitted for routine drug resistance testing, were amplified from plasma specimens and used to generate gag-pro RTVs that express firefly luciferase. In addition, gag-pro substitutions associated with reduced susceptibility to assembly or maturation inhibitors were introduced into an RTV containing a “wild-type” gag-pro sequence; site-directed mutants (SDM). Susceptibility to two maturation (CA-SP1 cleavage site) inhibitors and one capsid assembly/disassembly inhibitor (CAI) were determined. Results: Susceptibility to both CA-SP1 cleavage site inhibitors varied more than 100-fold across the 111 patient isolates, while susceptibility to the CAI varied less than 4-fold. Consistent with previous studies, viruses containing naturally occurring polymorphisms (68/111, 61%), or site directed mutations, within the “QVT” motif (aa positions 369-371) exhibited large reductions in CA-SP1 cleavage site inhibitor. In addition, six of 43 isolates lacking QVT polymorphisms also exhibited notable reductions in CA-SP1 inhibitor susceptibility. In contrast, only one patient isolate contained a polymorphism (N74D) that has been associated with reduced susceptibility to CAI (L56I, M66I, Q67H, N74D, A105E). SDMs containing single L56I, M66I and A105E substitutions exhibited large reductions in CAI susceptibility (FC>200), whereas the impact of Q67H and N74D was small (FC=1.8 and 2.6, respectively). Notably, L56I, M66I and Q67H substitutions also conferred modest cross-resistance (3 to 10-fold) to the CA-SP1 cleavage site inhibitors. Conclusion: Susceptibility to HIV-1 capsid inhibitors that vary in their mechanism of action were assessed using a cell-based pseudovirus reporter assay. Variation in susceptibility across more than 100 patient isolates was much more pronounced for CA-SP1 cleavage site inhibitors compared to a CAI. A small number of mutations conferred large reductions in CAI susceptibility and cross resistance to CA-SP1 cleavage site inhibitors.

CROI 2019 203