CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

viral control with antiretroviral therapy (ART), but the mechanisms driving this phenomenon are unknown. Here we examined MAIT cell loss in ART-treated HIV-infected donors with or without CD4 T cell recovery. Methods: PBMCs from fresh whole blood from HIV-negative (n=25) or ART-treated HIV-positive donors with full (Immune Success [IS]; n=18) or impaired (Immune Failure [IF]; n=13) CD4 T cell recovery were analyzed by flow cytometry for T cell markers, TCR Vα7.2, and CD161. PBMCs from healthy controls were cultured with or without TCR-mediated stimulation, and CD161 expression was assessed on Vα7.2+ T cells. Interferon-γ (IFNγ) production was assessed by intracellular cytokine staining. Results: We find decreases in the percentage of CD3+ T cells that express CD161 (0.9% vs. 4.7%; P<0.0001) and the percentage of Vα7.2+ T cells that express CD161 (23% vs. 71%; P<0.0001) in HIV-infected individuals. We also find a significant increase in the percentage of T cells that are CD161-Vα7.2+ in IF compared to controls (3.3% vs. 1.7%; P=0.0003), accompanied by an increase in the percentage of CD161-Vα7.2+ T cells that express CD8 in IF (76% vs. 48%; P<0.0001), but not IS (P=0.2847), donors. After TCR stimulation in vitro, Vα7.2+ T cells reduced expression of CD161 by 49% (P=0.002), and CD161- Vα7.2+ cells produced less IFNγ than CD161+ Vα7.2+ cells (3.7% vs. 29.5%; P=0.016). Conclusion: Our findings suggest that in ART-treated HIV-infected subjects who do not recover CD4 T cell numbers, the reduction in peripheral MAIT cells is due at least in part to a loss in CD161 expression, and not merely trafficking into mucosal tissues or cell death. The residual CD161-Vα7.2+ cells may be impaired in their ability to synthesize IFNγ. 277 RESIDUAL LOW-LEVEL VIREMIA ON ART NOT ASSOCIATED WITH LOWER SYSTEMIC EXPOSURE TO ARVs Monica Gandhi 1 , Andrei Stefanescu 2 , Ronald Bosch 2 , Rajesh T. Gandhi 3 , Joshua C. Cyktor 4 , Howard Horng 1 , Joseph J. Eron 5 , John W. Mellors 4 , Deborah McMahon 4 1 Univ of California San Francisco, San Francisco, CA, USA, 2 Harvard Univ, Boston, MA, USA, 3 Massachusettes General Hosp, Boston, MA, USA, 4 Univ of Pittsburgh, Pittsburgh, PA, USA, 5 Univ of North Carolina at Chapel Hill, Chapel Hill, NC, USA Background: The persistence of HIV has been postulated by some to be due to ongoing viral replication in reservoir sites because of inadequate antiretroviral (ARV) exposure. Few data exist on the relationship between cumulative systemic exposure to ARVs and levels of residual viremia on ART. If residual viremia were a consequence of marginal ARV exposure, then lower ARV exposure would correlate with higher levels of residual viremia. Methods: Scalp hair and blood samples were collected as part of a longitudinal cohort study (ACTG 5321) of 112 participants with chronic HIV infection with plasma HIV RNA <50 copies/ml on ART over a prolonged duration. Small samples of hair were analyzed for ARV concentrations (as cumulative measures of exposure) using validated liquid chromatography/tandemmass spectrometry assays. Plasma samples obtained close to time of hair sampling were assayed for residual viremia (HIV RNA) by a single copy RT-PCR assay targeting integrase (LOD 0.4 cps/ml). Spearman correlations evaluated the relationship between hair ARV levels and residual viremia. Results: Median duration of ART at study entry was 7 yrs. All participants were on TDF/FTC and a PI, INSTI or NNRTI. There was substantial (5-fold) person-to-person variability in tenofovir (TFV) levels in hair (10th-90th percentile: 0.02-0.09 ng/mg). This variability was similar to that reported for ARV-taking patterns of 4-7 doses /week in other cohorts with less well-suppressed viremia. No significant correlation was found between ARV hair concentrations and residual viremia: FTC/TFV r= 0.03 (p= 0.78); RAL r= 0.16 (p= 0.56); EFV r=-0.02 (p= 0.94); DRV r=0.11 (p= 0.73). Of note, TFV concentrations when co-administered with boosted ARVs (ATV/r, DRV/r, EVG/cobi) were 1.3-fold higher than when co- administered with unboosted ARVs (RAL, EFV, RPV) (Wilcoxon ranksum; p=0.018). Conclusion: Hair concentrations of TFV as a cumulative measure of TFV exposure were distributed over a wide range and pharmacoenhancers (like ritonavir) increased TFV levels by 30%. Despite the variability in ARV concentrations, there was no apparent relationship between ARV exposure and the level of residual viremia. This result argues against inadequate ARV exposure allowing ongoing viral replication and resulting in residual viremia. Whether ARV exposure is associated with other markers of HIV persistence or with persistent inflammation or immune activation should be explored. 278 ANTIGP41 TITERS REFLECT RESIDUAL REPLICATION IN VIROLOGICALLY SUPPRESSED ART PATIENTS Heloise M. Delagreverie 1 , Maxime Grudé 2 , Sidonie Lambert 3 , François Raffi 4 , Catherine Leport 5 , Cécile Goujard 6 , Lorraine Plessis 7 , Christine Katlama 8 , Francis Barin 9 , Constance Delaugerre 10 1 Hôpital Saint Louis, PARIS, France, 2 INSERM, Paris, France, 3 Institut Pierre Louis d’Epidémiologie et de Santé, Paris, France, 4 CHU Hotel-Dieu Nantes, Nantes, France, 5 Hôpital Bichat, Paris, France, 6 AP-HP, Le Kremlin-Bicêtre, France, 7 INSERM, Villejuif, France, 8 Pierre and Marie Curie Univ, Paris, France, 9 INSERM Tours, France, 10 Hôpital Saint Louis, Paris, France Background: A major challenge to HIV cure strategies is the measurement of the total burden of persistent replication-competent HIV. Suppression of viral replication is likely to lessen antigen presentation to B cells, leading to a decrease in anti-HIV antibody production. Our aimwas to determine whether anti-gp41 titer was correlated with reservoir size and could be valuable as a surrogate marker of HIV-1 reservoir in suppressed patients on antiretroviral therapy (ART). Methods: Anti-gp41 antibody titers were assayed in an cross-sectional study of HIV-1 infected ART treated patients with plasma HIV-1 RNA viremia strictly below 50 copies/ml for 12 months, from 3 ANRS cohorts (COPANA, MONOI and APROCO). Binding to the immunodominant epitope of gp41 was measured by enzyme-linked immunoassay. Pearson correlation with age, gender, duration of HIV infection, time on ART, duration of viremia suppression, nadir and current CD4+ T-cell counts, current CD4/CD8 ratio, plasma HIV-1 RNA (ultrasensitive viral load, usVL, LOQ 1 RNA copy/ml) and blood HIV-1 DNA (LOQ 1.6 log10 DNA copies/10^6 PBMC) were tested Results: 812 patients were included. They were mostly male (77%) with a median age of 41 yrs. Nadir CD4+ cells was 237/µl, median time since HIV diagnosis was 12 yrs and median ART duration was 9.5yrs with 4.2yrs of viral suppression. Median CD4+ T-cell count was 566/µl (CD4/CD8 ratio =0.8). Amongst the 213 patients with usVL data, 46% (n=98) tested twice consecutively below 1 copy/ml, and nadir CD4+ cells was 237/µl. Median (IQR) total HIV-DNA (n=364 patients tested) was 2.6 (2.2-3.1) log copies/10^6 PBMC. In 809 patients, the median (IQR) titer of anti-gp41 antibody was 1.5 (0.7-2.1). There was a significant inverse correlation between anti-gp41 antibody titer and time on ART, duration of viral control, current CD4+ cell count and CD4+/CD8+ ratio. Consistent usVL below 1 copy/mL was associated with a median gp41 titer of 1.1 (0.5-1.6) vs 1.4 (0.7-1.9) in patients with at least one usVL above 1 copy/ml (p=0.009). A trend towards a lower titer of anti-gp41 antibodies was observed in patients with undetectable HIV DNA: 0.7 (0.5-1.6) vs 1.3 (0.7-1.8) with detected HIV DNA, p=0.06. Conclusion: Maximal viral suppression, leading to minimal antigen stimulation, was associated with a decrease in anti-gp41 antibodies titers. Importantly, low anti-gp41 titers were combined with a lower HIV DNA burden. Anti-gp41 titration may be an additional biomarker of effective HIV replication suppression in patients on ART. 279 ANTIBODY LEVELS CORRELATE WITH THE INFECTED CELL POPULATION IN HIV PATIENTS ON ART Sheila Keating 1 , Christina Lalama 2 , Ronald Bosch 2 , Deborah McMahon 3 , Joshua Cyktor 3 , John Mellors 3 , Joseph J. Eron 4 , Rajesh T. Gandhi 5 , Michael P. Busch 1 , for the ACTG 5321Team 1 Blood Systems Rsr Inst, San Francisco, CA, USA, 2 Harvard Univ, Boston, MA, USA, 3 Univ of Pittsburgh, Pittsburgh, PA, USA, 4 Univ of North Carolina at Chapel Hill, Chapel Hill, NC, USA, 5 Massachusetts General Hosp and Ragon Inst, Boston, MA, USA Background: Current methods to quantify the HIV reservoir in individuals on ART use large volumes of blood or tissue biopsies. In contrast, small volume, easy to perform assays to assess changes in HIV reservoir size and expression are needed. Antibody (Ab) assays can be performed with a fewmicroliters of serum or plasma, are reproducible, and results can be obtained within hours. We assessed whether plasma Ab levels and Ab avidity to HIV are associated with the frequency of HIV-infected cells and proviral transcriptional activity in blood. Methods: Longitudinal PBMC and plasma samples were obtained from an ART-treated cohort of individuals with virologic suppression (ACTG A5321) who had HIV RNA levels <50 cp/mL with no blips or ART interruption (≥21 consecutive days) documented from year 1 of follow-up. Virologic and immunologic testing was performed on stored PBMC and plasma samples from pre-ART (n=101), year(yr) 1 (n=101), yr4 (n=101), and yrs 6-15 (n=66) on ART. Less-sensitive VITROS (S/Co) and Avidity VITROS (AI) Ab measurements were performed on plasmas at all timepoints except pre-ART. Plasma HIV RNA, unspliced cell-associated HIV DNA and unspliced HIV RNA were measured by qPCR assays. Results: Both HIV Ab measurements showed a statistically significant decline between yr1 and yr4 on ART with median participant-specific decline of 19%/yr for S/Co and 5.6%/ yr for AI (p<0.001 for each) (Figure). No significant associations were found between yr1 HIV Ab levels and pre-ART HIV RNA, CD4 T-cell count or CD4:CD8 ratio. S/Co and AI were

Poster and Themed Discussion Abstracts

CROI 2017 109