CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

523 BENEFITS OF InSTI-BASED REGIMEN AT THE TIME OF PRIMARY HIV INFECTION Raphaël Veil 1 , Isabelle Poizot-Martin 2 , Jacques Reynes 3 , Cécile Goujard 4 , Remonie Seng 4 , Pierre Delobel 5 , Laurent Cotte 6 , Claudine Duvivier 7 , David Rey 8 , Laurent Tran 9 , Laure Surgers 10 , Clotilde Allavena 11 , Caroline Lascoux-Combe 12 , Antoine Chéret 4 , Laurence Meyer 9 1 Assistance Publique – Hôpitaux de Paris, Paris, France, 2 Assistance Publique– Hopitaux Marseille, Marseille, France, 3 CHU de Montpellier, Montpellier, France, 4 Hôpital Bicêtre, Le Kremlin-Bicetre, France, 5 CHU de Toulouse, Toulouse, France, 6 CHU de Lyon, Lyon, France, 7 Necker Hospital, Paris, France, 8 Hôpitaux Universitaires de Strasbourg, Strasbourg, France, 9 INSERM, Le Kremlin-Bicetre, France, 10 Saint-Antoine Hospital, Paris, France, 11 CHU Hôtel-Dieu, Nantes, France, 12 Hôpital Saint-Louis, Paris, France Background: Combined antiretroviral therapy (cART) containing integrase strand transfer inhibitor (INSTI) has been shown to be superior to boosted protease inhibitor (PI) in chronic HIV-infected patients. We thus compared, on a large dataset of 712 patients, the efficacy of INSTI versus PI-containing cART initiated at the time of primary HIV infection (PHI), a key period in the HIV natural history. Methods: This multicentre observational cohort study was conducted among patients initiating cART between 2013 and 2017. Data were pulled from 2 sources: the ongoing ANRS PRIMO cohort which enrolls patients during PHI, and Dataids’, a French hospital database which uses computerised medical record collected during clinical visits. The primary outcome was the time from cART initiation to reach plasma HIV-1 RNA below 50 copies/ml. Turnbull interval- censored estimator was used to draw cumulative-event curves, and groups were compared with a generalized logrank test. The CD4 T cells restoration was estimated with a segmented mixed linear model. Results were adjusted for the data source. Results: 712 patients initiating cART during PHI were included in the study, 299 initiating with an INSTI and 413 with a PI. Patients characteristics (age, sex, sub-Saharan origin, transmission group, and baseline HIV-1 RNA, CD4 count and CD4/CD8 ratio) were similar in the two groups, except for the year of treatment initiation: the proportion of patients initiating an INSTI-based regimen increased in more recent years. Time to virological response was faster for INSTI-treated patients vs PI-treated ones (logrank test: p<0.001). Proportions of patients showing a virological response was 37% vs 6% at 1 month (±7 days) of treatment, 77% vs 38% at 3 months (±15 days), 92% vs 79% at 6 months (±30 days) and 93% in both groups at 12 months (±60 days). During the first month, INSTI-treated patients gained on average 41 more CD4 cells/μl (p = 0.046) than PI-treated ones; mean CD4 counts were similar in the 2 groups at 1 year. CD4/ CD8 ratio followed the same pattern. Results were similar on a per-protocol analysis, or when comparing only Dolutegravir vs Darunavir-containing cART. Conclusion: We show here, using ‘real-life’ data, both an earlier virological response and a faster immune restoration in PHI patients treated with an INSTI- based regimen.

Long-term follow-up of this cohort is ongoing to understand whether benefits of starting ART during acute infection are maintained over years.

522 VIRAL BLIPS AFTER TREATMENT INITIATION DURING ACUTE HIV INFECTION Trevor A. Crowell 1 , Suteeraporn Pinyakorn 1 , Carlo Sacdalan 2 , Eugène

Kroon 2 , Donn J. Colby 2 , Suwanna Pattamasavin 2 , Sasiwimol Ubolyam 3 , Rapee Trichavaroj 4 , Oratai Butterworth 1 , Ellen Turk 1 , Corinne McCullough 1 , Mark de Souza 2 , Nittaya Phanuphak 2 , Jintanat Ananworanich 1 , for the RV254/SEARCH010 Study Group 1 US Military HIV Research Program, Silver Spring, MD, USA, 2 SEARCH, Bangkok, Thailand, 3 HIV–NAT, Thai Red Cross AIDS Research Centre, Bangkok, Thailand, 4 Armed Forces Research Institute of Medical Sciences in Bangkok, Bangkok, Thailand Background: Transient episodes of low-level HIV viremia, or blips, are observed in up to 50% of individuals on suppressive antiretroviral therapy (ART) initiated during chronic HIV infection and may be associated with clinical failure, viral evolution, and blunted reservoir decay. We described the incidence and predictors of blips after ART initiation during acute HIV infection (AHI). Methods: Participants were offered ART during AHI fromMay 2009 to August 2018 in Bangkok, Thailand. Those who continued ART for ≥1 year after viral suppression (confirmed HIV RNA <50 copies/mL) were included in these analyses. A blip was defined as any HIV RNA 50-999 copies/mL immediately preceded and followed by HIV RNA <50 copies/mL without a change in ART. Negative binomial regression was used to calculate rate ratios (RRs) and 95% confidence intervals (CIs) for associations of participant characteristics at ART initiation with blips. Fiebig stage and factors that were significant (p<0.05) in unadjusted models were included in the final multivariable model. Results: Of 299,004 samples screened, 564 participants were enrolled during AHI and 416 satisfying inclusion criteria were monitored for blips for a median of 2.7 (interquartile range [IQR] 1.9-3.9) years after achieving viral suppression. Participants had median age 26 (IQR 23-31) and were predominantly men who have sex with men (92.6%) with HIV subtype CRF01_AE (77.2%). Thirty (7.2%) participants demonstrated blips with incidence 2.7 (95% CI 1.8-3.7) per 100 person-years. Among 35 blips observed, 18 (51.4%) were 50-75 copies/mL, 14 (40.0%) were 76-199 copies/mL, and 3 (8.6%) were 200-999 copies/mL. Characteristics at ART initiation that were independently associated with blips included HIV RNA >6 log10copies/mL (RR 2.51 [95% CI 1.04-6.04], compared to ≤6 log10copies/mL) and CD4 ≤350 cells/mm3 (RR 2.46 [95% CI 1.00-6.03], compared to >350 cells/mm3). There was a non-significant trend towards increased blips after ART initiation in later Fiebig stages (Fiebig III/IV RR 1.45 [95% CI 0.63-3.31], Fiebig V [RR 2.74 [95% CI 0.73-10.35], compared to Fiebig I/ II), controlling for HIV RNA and CD4. Conclusion: Viral blips were uncommon and of generally lowmagnitude after ART initiation during AHI, suggesting a potential benefit of early ART initiation. As with ART initiation during chronic infection, higher HIV RNA and lower CD4 were predictive of blips. Further follow-up is needed to evaluate associations with viral reservoirs and clinical outcomes.

Poster Abstracts

CROI 2019 196