CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

533 HLA GENOTYPE IS ASSOCIATED WITH PRETHERAPY ACCESSORY InSTI RESISTANCE MUTATION L74I Guinevere Q. Lee 1 , Suzanne McCluskey 2 , Nicholas Musinguzi 3 , Zabrina Brumme 4 , Yap Boum 3 , Bosco M. Bwana 3 , Conrad Muzoora 3 , Simone Kigozi 3 , Peter W. Hunt 5 , Jeffrey Martin 5 , David R. Bangsberg 6 , Mark J. Siedner 2 , P. Richard Harrigan 7 , Jessica E. Haberer 2 1 Weill Cornell Medicine, New York, NY, USA, 2 Massachusetts General Hospital, Boston, MA, USA, 3 Mbarara University of Science and Technology, Mbarara, Uganda, 4 Simon Fraser University, Burnaby, BC, Canada, 5 University of California San Francisco, San Francisco, CA, USA, 6 Oregon Health and Sciences University, Portland, OR, USA, 7 University of British Columbia, Vancouver, BC, Canada Background: Naturally-occurring HLA-driven polymorphisms in HIV-1 may confer decreased susceptibility to antiretroviral therapies, but our knowledge of such polymorphisms in non-B subtypes remains incomplete. Here, we examine whether HLA-genotype is associated with pre-therapy integrase strand transfer inhibitors (INSTI) resistance mutations in a treatment-naïve Ugandan cohort. Methods: HIV-1 integrase bulk sequencing and HLA-genotyping were performed on pre-therapy-initiation plasma and PBMC collected between 2005-2010 from n=511 INSTI-naïve participants in the Uganda AIDS Rural Treatment Outcomes (UARTO) cohort. RIP 3.0 was used for HIV-1 subtyping. Major INSTI-associated resistance mutations were defined by the Stanford HIV database version 8.8 as T66AIK, E92Q, G118R, E138KAT, G140SAC, Y143RCH, S147G, Q148HRK, N155H, R263K. Only one minor mutation, L74I, was included due to recent reports linking it to cases of INSTI-treatment failures in subtype A1. R was used for all statistical and phylogenetic analyses. Multivariable logistic regression models were described for six resistance-mutation-HLA-genotype pairs that had n=8 or more individuals fromwhich resistance mutations were observed. Results: We identified major INSTI mutations in 1.2% (6/511) of participants: T66I (n=1; subtype D), E138K (n=3; all subtype D), and E138T (n=2; all subtype A1). L74I was found in 6% (n=16/247 subtype A1) and 4% (n=8/200 subtype D) of individuals. None of these polymorphisms, when occur alone, were associated with reduced INSTI susceptibility according to Stanford HIVdb. Multivariate logistic regression analyses revealed associations between A*02, B*4415 and Cw*0407 with L74I (p=0.03, 0.01, 0.007, Fig 1) after adjusting for gender, age, subtype, and interactions between subtype and HLA-genotypes. Cohort prevalence of A*02, B*4415 and Cw*0407 were 37%, 10% and 9%, respectively. Sequences containing L74I did not cluster into a monophyletic group in phylogenetic analyses. Conclusion: Our data suggest that certain polymorphisms associated with INSTI resistance in specific viral subtypes may be HLA-driven. L74I have not been previously associated with HLA-escape in any viral subtype, suggesting the epitope responsible is not immuno-dominant. Lack of phylogenetic clustering suggests results are not attributable to viral founder effects. Effects of L74I on INSTI-based therapy, its link to HLA-genotypes, and whether it lowers genetic barrier to INSTI require additional large-scale population-level validation.

Background: In past studies, we performed in vitro washout experiments to show a more durable suppression of wild-type (WT) and integrase-resistant HIV-1 by dolutegravir (DTG) as compared to Raltegravir (RAL) or Elvitegravir (EVG) following release of drug pressure. In this study, we were interested in reproducing these observations on two newly developed integrase strand transfer inhibitors (INSTIs), Bictegravir (BIC) and Cabotegravir (CAB). Methods: Site directed mutagenesis generated pNL4-3 plasmid constructs harbouring Wild Type (WT), R263K, G118R, and G140S/Q148H integrase. MT-2 cells were infected with WT or resistant clones to establish IC 50 and IC 90 concentrations. MT-2 cells were then subjected to maximal drug pressure, using 20 times the IC 90 for each drug. Three days post-exposure, drugs were washed out from the cells. Viral rebound was assessed at days 3, 7 and 11 post-infection. Results: BIC showed a higher genetic barrier to resistance than DTG and CAB, based on IC 50 values. The R263K G118R, G140S/Q148H clones showed 1-, 1.4-, and 3.5-fold resistance to BIC relative to WT, respectively. This compares to 3.5-, 1.7- and 6.6-fold resistance to DTG and 0.8-, 6.4-, and 6.8-fold resistance to CAB against R263K, G118R, and Q140S/Q148H clones, respectively. In our washout experiments, WT and R263K were viral suppressed by all three drugs during selective pressure (20 x IC 90 ) and following drug washout (day 11). With G118R infected cells, viral rebound occurred following DTG washout with minimal increase in replication following CAB washout and no rebound following BIC washout. The G140S/Q148H clones were not susceptible to CAB prior to and following drug washout. While DTG could suppress replication t of G140S/Q148H infected cells, viral rebound occurred following washout (day 7). In contrast, BIC successfully suppressed replication through the 11 days of infection, showing minimal rebound after drug removal. Conclusion: Overall, we observed an extended duration of viral suppression of HIV-1 replication following release of drug pressure with BIC than either DTG or CAB. This included WT virus and viruses harboring mutations conferring low- level, moderate and high-fold drug resistance. These findings show that BIC may be pharmacologically more forgiving than DTG and CAB.

Poster Abstracts

535LB MUTATIONS IN THE HIV-1 ENVELOPE GLYCOPROTEIN CONFER BROAD DRUG RESISTANCE Yuta Hikichi 1 , Rachel Van Duyne 1 , Phuong Pham 1 , Jennifer L. Groebner 1 , Ann Wiegand 1 , John W. Mellors 2 , Mary F. Kearney 1 , Eric O. Freed 1 1 National Cancer Institute, Frederick, MD, USA, 2 University of Pittsburgh, Pittsburgh, PA, USA Background: We recently reported that, in vitro, HIV-1 can acquire resistance to the potent IN inhibitor, dolutegravir (DTG), by acquiring mutations in the envelope (Env) glycoprotein that enhance viral spread via cell-to-cell transmission. The aim of this study is to clarify the mechanism of Env-mediated HIV-1 drug resistance. Methods: Virus replication in the presence of ARVs was measured by propagating viruses in a spreading infection in SupT1 cells. Sensitivity to neutralizing antibodies (NAbs) was measured using TZM-bl indicator cells. gp120 shedding from virus particles was measured by western blotting. To examine possible in vivo relevance of Env-mediated drug resistance, we performed single-genome sequencing using plasma from 5 patients failing a raltegravir (RTG)-containing regimen (La Rosa et al., 2016).

534 HIV-1 VIRAL REBOUND AFTER BICTEGRAVIR, DOLUTEGRAVIR, AND CABOTEGRAVIR WASHOUT Ernesto Cuadra Foy 1 , Nathan Osman 1 , Ruxandra-Ilinca Ibanescu 2 , Maureen Oliveira 2 , Bluma G. Brenner 1 1 McGill University, Montreal, QC, Canada, 2 Lady Davis Institute for Medical Research, Montreal, QC, Canada

CROI 2020 191