CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

532 FREQUENT DISCORDANCE BETWEEN ETRAVIRINE PHENOTYPE & GENOTYPE IN SUBTYPE cART FAILURE Kevin D. McCormick 1 , Kerri J. Penrose 1 , Chanson J. Brumme 2 , P. Richard Harrigan 2 , Raquel V. Viana 3 , John W. Mellors 1 , Urvi Parikh 1 , Carole Wallis 4

533 IN-DEPTH CHARACTERIZATION OF HIV RESISTANCE TO INTEGRASE INHIBITORS IN BRAZIL Nazle Veras , Rosana Elisa G. Pinho, Ana R. Pati Pascom, José Boullosa, Fernanda Rick, Adele Benzaken Ministry of Health, Brasilia (DF), Brazil Background: Due to increasing HIV drug resistance, Brazil was one of the first countries to adopt Dolutegravir (DTG) in first-line antiretroviral therapy (ART). The Ministry of Health of Brazil offers genotyping tests to all individuals under an integrase inhibitor (INI) based regimen experiencing virological failure. Using real life data, we aimed to characterize HIV genotypic resistance to Raltegravir (RAL) and DTG in Brazil in order to better understand factors related to the development of INI resistance-associated mutations (RAM), and to depict INI RAM transmission chains. Methods: HIV integrase sequences from 2012-2018 were selected from the National System for Genotyping Control. The presence of INI RAM (Stanford HIVdb Program) and HIV subtype (Rega HIV Subtyping tool) were characterized. Socio-demographic, clinical (CD4 count and viral load/VL), and ART history data were assessed. A Pearson Chi-square test was carried out. INI RAM transmission clades were characterized by Bayesian phylogenetics. Results: We analyzed 1,467 HIV integrase sequences from RAL- and/or DTG- experienced individuals. HIV resistant strains were identified in 21.7% for RAL and 0.7% for DTG. In 2017, following the use of DTG in first-line ART, individuals on RAL-based regimen switched to DTG. As a reflection of DTG’s higher genetic barrier, resistance to INI has have been slightly decreasing to 13.7% and 0.3% in 2018 for RAL and DTG, respectively. Indeed, we did not identify any DTG resistant lineages in samples from individuals under DTG first-line ART. The prevalence of RAL and DTG resistant strains was similar, regardless of demographic and clinical data, including regional sustained VL levels. INI RAMs at positions G140 (7.0%) and E138 (1.0%) were most prevalent. Overall, subtype B (69.9%) was the most prevalent, followed by C (13.7%), F (8.9%) and recombinant forms (6.7%). Sequences presenting INI RAMwere dispersed in phylogenetic trees for subtypes B and C, showing no specific INI RAM transmission clade, considering both the national level and the five Brazilian geographic regions, separately. Conclusion: INI RAMmonitoring revealed a short-term decrease in resistance to INI, even after DTG large-scale use. In addition, phylogenetics revealed that INI RAM does not occur in a particular population group or geographic region. Hence, the successful pioneering implementation of DTG goes beyond costs savings but healthcare efficacy, corroborating to sustaintability of DTG as first- line ART in a public health program. 534 HIV DRUG RESISTANCE AMONG PWID IN EASTERN EUROPE AND ASIA, HPTN 074 Yinfeng Zhang 1 , Philip J. Palumbo 1 , Jessica M. Fogel 1 , Xu Guo 2 , William Clarke 1 , Paul Richardson 1 , Erica Hamilton 3 , Ngo T. Hoa 4 , Kostyantyn Dumchev 5 , Zubairi Djoerban 6 , Irving Hoffman 7 , Brett Hanscom 2 , William C. Miller 8 , Susan H. Eshleman 1 , for the HPTN 074 Study Team

1 University of Pittsburgh, Pittsburgh, PA, USA, 2 British Columbia Centre for Excellence in HIV/AIDS, Vancouver, BC, Canada, 3 Lancet Laboratories, Johannesburg, South Africa, 4 Lancet Labs and BARC SA, Johannesburg, South Africa Background: Etravirine (ETR) is a second-generation NNRTI that is used as a component of combination ART for treatment-experienced persons. The extent of cross-resistance between nevirapine (NVP) and efavirenz (EFV) and ETR is not well defined especially in low and middle-income countries (LMIC) where switches from first-line ART may be delayed. To address this gap, we investigated the susceptibility to ETR of subtype C HIV-1 among individuals on failing first-line NNRTI-containing regimens in South Africa (SA) and compared containing bulk-cloned full-length RT amplified from plasma of 100 HIV-1 subtype C-infected individuals failing first-line ART (>10000 cp/ml and >1 NNRTI RAM) were phenotyped for ETR susceptibility in TZM-bl cells. Fold-change was calculated using a composite IC 50 of 12 treatment-naïve individuals from SA. Genotypic scores (Stanford HIVdb v8.4) were categorized as partial or complete discordance if deviated from phenotype clinical cut-offs (DUET trials) by one or two tiers respectively. Correlations were determined using Pearson›s coefficient (r). WT reversions of K65 were made in clonally isolated plasmids with the QuikChange II Site- Directed Mutagenesis Kit. Results: Of 100 first-line ART failures, 54 had reduced ETR susceptibility above the clinical cut-off of 2.9-fold higher than the control IC 50 . The fold-change (FC) did not strongly correlate with genotypic score (r=0.47) with 44% of samples partially and 4% completely discordant. Of the 33 samples with FC>10, 26 samples were categorized as ‘low’ or ‘intermediate’ resistant by the HIVdb (Figure). The ETR-associated mutations L100I, Y181C and/or M230L were present in 79% (26/33) of samples with FC>10 but only in 4% (2/46) of samples with a FC<2.9. By contrast, the HIVdb NNRTI mutations A98G, K101H, E138A/K, V179D, Y188L, G190A, H221Y and P225H did not correlate with ETR resistance. The NRTI mutation 65R was associated with ETR resistance but reversion to 65K had no effect on ETR susceptibility. Rather, 65R was a marker of more prolonged ART failure and the accumulation of NNRTI mutations that conferred ETR resistance. Conclusion: Phenotypic cross-resistance to ETR is common after first-line NNRTI-containing ART failure in SA. Genotype-based algorithms differentially classify ETR susceptibility in Subtype C. More appropriate weighting of combinations of ETR associated mutations are needed to improve genotype predictions of ETR phenotype. ETR phenotype to genotype. Methods: Recombinant HIV-1 LAI

Poster Abstracts

CROI 2019 200