CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

488 EFFICACY AND DURABILITY OF 2-DRUG VS 3-DRUG InSTI-BASED REGIMENS: DATA FROM REAL LIFE Massimiliano Fabbiani 1 , Barbara Rossetti 1 , Arturo Ciccullo 2 , Maria Letizia Oreni 3 , Filippo Lagi 4 , Luigi Celani 5 , Manuela Colafigli 6 , Andrea De Vito 7 , Maria Mazzitelli 8 , Alex Dusina 2 , Stefano Rusconi 3 , Amedeo Capetti 3 , Gaetana Sterrantino 4 , Gabriella d'Ettorre 5 , Simona Di Giambenedetto 2 1 Siena University Hospital, Siena, Italy, 2 Catholic University of the Sacred Heart, Rome, Italy, 3 Luigi Sacco University Hospital, Milan, Italy, 4 University of Florence, Florence, Italy, 5 Sapienza University of Rome, Rome, Italy, 6 San Gallicano Dermatology Institute, Rome, Italy, 7 University of Sassari, Sassari, Italy, 8 Magna Græcia University of Catanzaro, Catanzaro, Italy Background: Due to high efficacy and tolerability, the use of integrase strand transfer inhibitors (InSTI) is increasing not only in standard 3-drug regimens (3DR) but also in 2-drug regimens (2DR). However, few data are available about comparison of these strategies in a real-life setting. Methods: Retrospective multicentre (8 clinical centers in Italy) observational study including HIV+ treatment-experienced patients with HIV-RNA (VL)<50copies/mL switching to InSTI-based 2DR or 3DR with at least one follow-up visit. Major outcomes were virological failure (VF, defined as 1 VL>1000copies/mL or 2 consecutive VL>50copies/mL) and regimen discontinuation due to any reason. Survival analyses were performed to estimate the probability of VF and discontinuation, and to evaluate their predictors. Results: Overall, 1666 patients [73%males, median age 51 years, 26% previously exposed to InSTI, median time from last VL>50 copies/mL 55 months, current and nadir CD4+ 676 and 184 cells/mmc, respectively] were included, of which 1334(80%) treated with 3DR [n=265 elvitegravir(EVG), n=334 raltegravir(RAL), n=735 dolutegravir(DTG)] and 332(20%) with 2DR [n=263 lamivudine+DTG; n=69 rilpivirine+DTG]. Over a median follow-up of 100 weeks(IQR 52-150), 52(3.1%) patients experienced VF with an incidence of 1.5 per 100 PYFU; the estimated 48-week probability of VF was not different between 2DR and 3DR(1.4% vs 1.8%;p=0.53), but it was higher for EVG(3.5%) and RAL(3%) when compared to DTG(1%)(p=0.04). By multivariate analysis, previous VF (aHR 2.7;p<0.001) and shorter time from last VL>50copies/ mL(aHR 0.9;p=0.04) predicted VF. Four-hundred(24%) patients discontinued InSTI-based regimen with an incidence of 11.3 per 100 PYFU. Main reasons for discontinuation were toxicity[n=159 (40%) of which 51(13%) CNS toxicity] and simplification(n=119, 30%). The estimated 48-week probability of discontinuation for any reason was 20% for RAL, 10% for DTG and 16% for EVG(p<0.001), without differences comparing 2DR and 3DR DTG-based(9% vs 10%;p=0.21). By multivariate analysis, there was higher risk of discontinuation in 3DR(vs 2DR aHR=2.1;p<0.001) and lower risk in MSM(vs heterosexuals aHR=0.75;p=0.02) and regimens started for simplification(aHR 0.5;p<0.001). Conclusion: In our real-life setting, both 2DR and 3DR InSTI-based regimens showed high efficacy and durability. Regimens including DTG were associated with a lower risk of VF and discontinuation 489 ASSESSING THE VIROLOGIC IMPACT OF ARCHIVED RESISTANCE IN AN HIV-1 SWITCH STUDY, TANGO Ruolan Wang 1 , Jonathan Wright 2 , Mounir Ait-Khaled 3 , Allan Raymond Tenorio 1 , Maria Claudia Nascimento 3 , Thomas Lutz 4 , Daniel Podzamczer 5 , Richard Moore 6 , Miguel Górgolas Hernández-Mora 7 , Clifford Kinder 8 , Jean van Wyk 3 , Mark Underwood 1 1 ViiV Healthcare, Research Triangle Park, NC, USA, 2 GlaxoSmithKline, Uxbridge, UK, 3 ViiV Healthcare, Brentford, UK, 4 Infektiologikum, Frankfurt, Germany, 5 Hospital Universitario de Bellvitge, Barcelona, Spain, 6 Northside Clinic, Fitzroy, Australia, 7 Hospital Universitario Fundación Jiménez Díaz, Madrid, Spain, 8 AHF Healthcare Center, Miami, USA Background: TANGO study demonstrated that switching to DTG/3TC fixed dose combination (FDC) 2-Drug Regimen (2DR) was non-inferior to continuing a TAF- based 3-drug regimen (TBR, 3DR) in maintaining virologic suppression in HIV-1 infected, ART-experienced adults through Week 48. The impact of pre-existing, HIV-1 drug resistance on virologic outcomes through Week 48 was assessed. Methods: Participants with historical IAS major NRTI or INSTI resistance associated mutations (RAMs) were excluded from the study. Pro-viral DNA genotyping was conducted retrospectively on baseline samples from randomized participants by Monogram Bioscience using GenoSure Archive assay. Virologic outcomes based on IAS major NRTI, NNRTI, PI and INSTI RAMs were determined by last available on-treatment HIV-1 RNA through Week 48

in order to assess pure virologic responses by censoring discontinuations due to non-efficacy reasons. This was repeated on the FDA Snapshot Algorithm at Week 48 as a sensitivity analysis. Results: 322 (87%) of participants in the DTG/3TC arm and 321 (86%) in the TBR arm had both pro-viral genotype data and at least one on-treatment HIV-1 RNA result. Archived major NRTI, NNRTI, PI and INSTI RAMs were observed in 42 (7%), 90 (14%), 43 (7%) and 6 (1%) participants, respectively across both arms (Table 1), and 474 (74%) participants were without any major RAMs at the baseline. The frequencies of NRTI RAMs M184V/I, K65E/N/R and thymidine analog mutations (TAMs) were low. Through Week 48, 322 (100%) of participants on DTG/3TC and 319 (>99%) on TBR were virologically suppressed (last on- treatment HIV-1 RNA <50 c/mL). For participants with any major NRTI, INSTI, NNRTI or PI RAMs, all were virologically suppressed. The results of a sensitivity analysis using the FDA Snapshot algorithmwere consistent with those using last available on-treatment HIV-1 RNA. One participant in TBR armwithout any archived RAMs met the protocol-defined, confirmed Virologic withdrawal criterion (CVW) with no emergent resistance. None in the DTG/3TC armmet CVW criteria through Week 48. Conclusion: In TANGO, archived major NRTI (e.g., M184V/I, K65E/N/R and TAMs) and INSTI (e.g., Q148R, Y143C/H, R263K) RAMs were infrequent. High rates of virologic suppression were maintained in participants on both treatment arms through Week 48. The presence of pre-existing, archived RAMS did not appear to impact virologic outcomes through Week 48.

Poster Abstracts

490 VIROLOGIC FAILURE AND RESISTANCE IN DOLUTEGRAVIR-BASED MAINTENANCE DUAL REGIMENS Colin Deschanvres 1 , François Raffi 1 , Jacques Reynes 2 , Bruno Hoen 3 , David Rey4, Romain Palich 5 , Olivier Robineau 6 , Firouze Bani-Sadr 7 , Claudine Duvivier 8 , Laurent Hocqueloux 9 , Lise Cuzin 10 , Véronique Joly 11 , André Cabié 10 , Clotilde Allavena 1 , for the the Dat'AIDS Study Group 1 CHU de Nantes, Nantes, France, 2 CHU de Montpellier, Montpellier, France, 3 CHU de Nancy, Vandœuvre-lès-Nancy, France, 4 Hôpitaux Universitaires de Strasbourg, Strasbourg, France, 5 AP–HP, Hôpitaux Universitaires Pitié Salpêtrière, Paris, France, 6 Centre Hospitalier de Tourcoing, Tourcoing, France, 7 CHU de Reims, Reims, France, 8 Necker Hospital, Paris, France, 9 Centre Hospitalier Régional d'Orléans, Orléans, France, 10 CHU Fort de France, Fort de France, Martinique, 11 AP–HP, Hôpital Bichat- Claude Bernard, Paris, France Background: Maintenance ART with dolutegravir (DTG)-based dual regimens (2DR) have proved their efficacy among HIV-1 infected subjects in large and randomized trials. Real-life data are scarce with limited population and follow-up. In a large cohort (Dat'AIDS), we evaluated virologic failure (VF) and resistance-associated mutations (RAMs) on DTG maintenance regimens in combination with rilpivirine (RPV) or 3TC/FTC (xTC), and we analyzed the factors associated with VF. Methods: Between 2014 and 2018, all HIV-1 adults starting DTG/RPV or DTG/xTC as a maintenance 2DR (i.e. with HIV RNA <50 c/mL) were enrolled in a retrospective analysis within the Dat’AIDS cohort (NCT02898987). VF was defined as 2 consecutive HIV RNA >50 c/mL or a single value >400 c/mL. We compared cumulative genotypes prior to 2DR and at VF (ANRS algorithm V29; 2018). Cox-models were used to analyze factors associated with VF. Results: 1374 subjects were included (DTG/RPV: 799, DTG/xTC: 575) with a median follow-up of 587 days [IQR 334-934] and 562 days [IQR 326-938],

CROI 2020 173