CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

population (14.9%). Lower nadir CD4 cell count, male sex, lower pre-ART HIV-1 RNA, older age at ART initiation and black ethnicity were identified as significant risk factors for lower ART-CD4-plateau in the multivariable model. Although the patients of black ethnicity were, when compared to patients of white ethnicity, younger at ART start (34 [28-39] vs. 40 [33-48]) and predominantly women (64.1% vs. 16.8%) the black population was found to have 61 cells/μL lower ART-CD4-plateau (95%-confidence interval [32-91]) even after adjusting for all cofactors (Fig. 1). Moreover, this finding was consistent among all the considered growth models. Lastly, the ART-CD4%-plateau was also found to be significantly lower for black ethnicity, indicating a different pattern of CD4 cell recovery. Conclusion: Our approach established a CD4 cell recovery phenotype based on longitudinal data and revealed black ethnicity as an important subpopulation with a distinct CD4 recovery profile. Enabling in-depth analysis of determinants of immune system recovery in treated HIV infection highlights the utility of our method as component for precision medicine.

1 Research Institute of McGill University Health Centre, Montreal, QC, Canada, 2 Clinique Médicale l’Actuel, Montreal, QC, Canada, 3 Clinique Médicale du Quartier Latin, Montreal, QC, Canada, 4 Centre Hospitalier de l’Université de Montréal, Montreal, QC, Canada Background: (1 → 3)-β-D-Glucan (βDG) is one of the most abundant components of fungal cell walls. People living with HIV (PLWH) without invasive fungal infection have been reported to have elevated plasma levels of circulating βDG. Such elevation is correlated with markers of gut damage, immune activation, and the occurrence of non-AIDS events. However, the mechanisms by which βDG induces immune activation and contributes to disease progression remain undefined. We aim to 1) Correlate βDG levels with CD4 and CD8 T cell activation markers as well as integrated HIV DNA, 2) Correlate plasma levels of βDG with expression of its receptors, and 3) Demonstrate a direct effect of βDG on the induction of immune activation in vitro. Methods: We analyzed plasma and peripheral blood mononuclear cells (PBMC) from participants receiving or not ART. We assessed the frequency CD4 and CD8 T cell activation (HLADR+CD38+) in PLWH PBMC and CD4 T cell bearing integrated HIV DNA by nested qPCR. We used flow cytometry to measure the expression of the βDG receptors Dectin-1 and NKp30 on monocytes and NK cells respectively. We assessed the dynamics of βDG receptor expression up to two days after Saccharomyces-derived βDG stimulation compared to bacterial lipopolysaccharide (LPS) stimulation in vitro. We analyzed indoleamine 2,3-dioxygenase-1 (IDO-1) expression by flow cytometry and cytokine secretion in the supernatant by ELISA following stimulation with βDG and/or LPS. Results: Higher plasma βDG levels correlated with higher frequencies of HLADR+CD38+ CD4 (r=0.69, p<0.001) and CD8 T cells (r=0.65, p<0.001, n=26), as well as HIV reservoir size in PLWH (r=0.41, p=0.04, n=24). Plasma βDG negatively correlated with the expression of its receptors Dectin-1 on monocytes (r=-0.58, p<0.01) and NKp30 on NK cells (r=-0.61, p<0.01) in 33 participants. In vitro, βDG stimulation prompted a reduction of Dectin-1 and NKp30 expression after 24 and 48 hours of stimulation. βDG stimulation predominantly induced IL-8, TNF-α and IDO-1 production over IL-1β and IL-6 in vitro. Conclusion: βDG elevation correlated with the frequency of activated CD4 and CD8 T cells and HIV reservoir size. βDG induced immune activation independently of LPS by triggering Dectin-1 and NKp30 on monocytes and NK cells respectively, and inducing cytokine secretion mostly IL-6, IL-8 and IDO expression. Our results pave the way to new treatment strategies to reduce inflammation and prevent the development of non-AIDS events. 246 MINING FOR CD4 CELL RECOVERY PHENOTYPE REVEALS DISTINCT PATTERN FOR BLACK ETHNICITY Teja Turk 1 , Christian Thorball 2 , Jacques Fellay 2 , Alexandra Trkola 3 , Peter Rusert 3 , Karin Metzner 1 , Dominique L. Braun 1 , Jürg Böni 3 , Sabine Yerly 4 , Vincent Aubert 5 , Thomas Klimkait 6 , Huldrych F. Günthard 1 , Roger Kouyos 1 , for the Swiss HIV Cohort Study 1 University Hospital Zurich, Zurich, Switzerland, 2 École Polytechnique Fédérale de Lausanne, Lausanne, Switzerland, 3 University of Zurich, Zurich, Switzerland, 4 University Hospitals of Geneva, Geneva, Switzerland, 5 Lausanne University Hospital, Lausanne, Switzerland, 6 University of Basel, Basel, Switzerland Background: With ever increasing majority of HIV-infected individuals receiving suppressive antiretroviral treatment (ART), one of the key questions is understanding the factors that govern the CD4 cell recovery in this population. Methods: We explored a range of asymptotic growth curve models to describe CD4 cell recovery after ART initiation and to capture its saturation. A well-characterized ART-naïve patient population of 2,583 individuals from the Swiss HIV Cohort Study (SHCS) receiving suppressive combination ART for at least 3 years (with >9 CD4 cell counts available) was used to establish the recovery phenotype and its population distribution, in particular the CD4 cell count plateau (ART-CD4-plateau) under suppressive ART (supp-ART). Individual ART-CD4-plateaus for a broader population of 4,089 ART-naïve SHCS patients under supp-ART (for at least 3 years with >5 CD4 observations) were inferred from the individual CD4 cell counts given the population distribution. The same approach was applied to CD4 cell percentage to obtain CD4 cell percentage plateaus (ART-CD4%-plateau). Results: Median ART-CD4-plateau in the supp-ART population based on a Janoschek growth model was 769 cells/μL (IQR [606-945]). Among patients of white ethnicity (76.1%) the median plateau was 785 cells/μL [620-968] compared to a lower median plateau of 720 cells/μL [560-872] observed in black

Poster Abstracts

247 SPD-L1: A POTENTIAL NOVEL IMMUNE MARKER FOR HIV-1 INFECTION AND VIROLOGIC FAILURE Jose Avendaño-Ortiz 1 , Marina Rubio Garrido 2 , Emilio Llanos-González 1 , Jorge Romero 3 , Santiago Moreno 2 , Carmen Rodriguez 3 , Eduardo López-Collazo 1 , Africa Holguin 2 1 Hospital La Paz Institute for Health Research, Madrid, Spain, 2 Instituto Ramón y Cajal de Investigación Sanitaria, Madrid, Spain, 3 Centro Sandoval, Madrid, Spain Background: Despite viremic control, basal chronic inflammation and its related comorbidities remains an unsolved problem among people living with HIV (PLWH). We explored the impact of HIV-1 infection, antiretroviral therapy (ART) exposure, viral load (VL) and sexual transmitted coinfections (STI) in soluble Programmed death-ligand 1 (sPD-L1) levels, a well-described inductor of T-cell exhaustion in other clinical contexts. Methods: Plasma samples were collected from 69 adults under clinical follow up in Madrid, Spain. Forty-nine were HIV-1-infected (20 drug naïve and 29 under ART) and 20 HIV-1 free. Among ART treated, 13 were under virological failure and 16 had suppressed VL (<1.6log or <40 HIV-1 RNA copies/ml), with sexual transmitted coinfections (STI) in 5 of 16 cases. Plasmatic sPD-L1 levels were measured using ELISA Kit for Programmed Cell Death Protein 1 Ligand 1, Cloud Clone Corp. Results: All 49 HIV-infected patients exhibited significant higher sPD-L1 levels than 20 uninfected adults (1.05ng/ml vs. 0.52ng/ml; p<0.001). Levels remained elevated in HIV-infection despite VL control, after comparing 16 infected with undetectable VL with 20 uninfected (0.75ng/ml vs. 0.52ng/ml; p=0.02). ART exposure seemed not to decrease sPD-L1 levels when comparing 16 treated infected with undetectable viraemia vs. 20 naïve (0.75ng/ml vs. 0.87ng/ml; p=0.199). We also found a significant impact of VL on sPDL1 values. Thirteen ART treated subjects under virological failure exhibited the highest sPDL1 levels, being significantly higher than in naive (1.68ng/ml vs 0.87ng/ml; p=0.002) or than in 16 ART treated subjects with supressed viraemia (1.68 ng/ ml vs. 0.79ng/ml, p=0.002). The last could be explained by differences in mean VL (5.1log vs. 3.7log vs. <1.6log, respectively). Along these line, there was a positive correlation between VL and sPD-L1 levels in plasma in the whole cohort

88

CROI 2019