CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

922 CD31 Expression on CD4 Cells Predicts Clinical Course of HIV in a Perinatally HIV-Infected Cohort Ramia Zakhour 1 ; Gilhen Rodriguez 2 ; Cynthia Bell 2 ; Guenet Degaffe 2 ; Laura Benjamins 2 ; Gabriela DelBianco 2 ; Elizabeth Donnachie 2 ;Tran Dat 2 ; Gloria P. Heresi 2 ; James R. Murphy 2 1 University of Texas, Houston, TX, US; 2 UTHealth Medical School, Houston, TX, US

Background: Thymic production of T-cells is important for maintenance of good immunological function in HIV infected individuals, however methods of measurement of recent thymus emigrant (RTE) T-cells are not easily adapted to use in HIV clinics. CD31 expression on CD4 cells is a marker of RTE CD4 cells and can be easily incorporated into common clinical monitoring procedures. We hypothesized that quantification of RTE measured as CD4+CD31+ predicts subsequent clinical course of HIV infection. Methods: We performed a 2 year longitudinal prospective study of 69 perinatally HIV infected individuals from a single HIV clinic. CD31 expression was measured using a modified FACS procedure incorporating a FITC-conjugated anti CD31 (MBC78.2) into a standard CD4+measurement routine. Other data are frommedical records. For analyses, patients’ HIV clinical status was categorized as good (CD4 ≥ 25% and HIV RNA ≤ 50 copies/ml), intermediate ( ≥ 25%, > 50) or poor (< 25%, >50). Spearman’s rho, Kruskal-Wallis, Kaplan-Meier, Log Rank tests and Cox proportional hazards models were used. Results: CD4+CD31+ cells correlated positively with CD4+ numbers but were independent of plasma HIV RNA level. Using an experiment determined discriminator of 50% CD4+CD31+, we used Kaplan-Meier analyses to test whether baseline CD4+CD31+ value predicts subsequent clinical status. Patients in good or intermediate clinical status who had CD4+CD31+ values < 50% at enrollment had significant (p 0.0006) deterioration to the clinically poor classification over the subsequent 2 years as compared to those with CD4+CD31+ ≥ 50% (Figure). Reciprocally, patients in poor or intermediate clinical status who had CD4+CD31+ ≥ 50% at enrollment had significant improvement to clinically good status (p 0.054) over 2 years as compared to those with CD4+CD31+< 50%. CD4+CD31+measurements provided a predictive capacity that did not overlap with T-cell activation measured as CD8+CD38+>17%.

CD31 on CD4+ cells is a practical addition to established routine methods for immunological assessment of HIV patients and is applicable even with rudimentary FACS hardware. Conclusions: For perinatally HIV infected patients in good immunological status, the finding of < 50% of CD4 cells expressing CD31 is a predictor of subsequent clinical deterioration. The population identified by this CD31-based criterion did not overlap with that classified as having poor prognosis using the CD8+CD38+marker of T-cell activation. 923 Premature Aging and Immune Senescence in HIV-1-Infected Children Background: Several data indicate that HIV-1-infected adults undergo premature aging and immune senescence; chronic immune activation may play a critical role in these dysfunctions. Limited data are available for HIV-1-infected children, in whom immune activation and senescence are likely to be more deleterious, since their immune system co-evolves from birth with HIV-1. Methods: 71 HIV-1-infected (HIV + ) children, aged from 0-5 years, 65 HIV-1-exposed-uninfected (HEU) and 49 HIV-1-unexposed-uninfected (HUU) age-matched children were studied. 41% of the HIV + children were not on antiretroviral therapy (ART). Telomere length (TL) and T-cell receptor rearrangement excision circle (TREC) levels were quantified in peripheral blood cells by Real-Time PCR. Subgroups of 18 HIV + , 21 HEU and 19 HUU children were studied for CD4 and CD8 cell differentiation (CD45RA,CD27), senescence (CD28,CD57) and activation/exhaustion (CD38,HLA-DR,PD1) markers by flow cytometry. Statistical analyses were performed with SPSSv21. Results: TL were significantly shorter in HIV + than in HEU and HUU children (overall p=0.009, adjusted for age); moreover, HIV + ART-naïve children had shorter TL compared with children on ART (median 2.1[interquartile range 1.7-2.4] vs 2.6[2.0-3.0]; p=0.002). CD8 naïve cells (CD45RA + CD27 + ) and TREC levels were significantly lower in HIV + than in HEU and HUU groups (overall, p<0.001 and p<0.001, respectively), while percentages of CD8 effector memory (CD45RA - CD27 - ) and terminally differentiated cells (CD45RA + CD27 - ) were higher in the former (overall, p=0.025, and p<0.001, respectively). CD8 senescent cells (CD57 + CD28 - ) were higher in HIV + than in HEU and HUU children (33.0[17.8-50.0] vs 10.6[3.4-36.2] vs 11.8[5.9-22.6]; p=0.002) as were CD8 activated cells (HLA-DR + CD38 + ) (7.9[4.1-17.2] vs 4.7[3.9-7.7] vs 3.4[2.8-7.6]; p=0.09). PD1 expression on CD8 cells was higher in HIV + children than in HEU and HUU groups (overall, p<0.001) and strongly related to CD8 activated cells (r=0.806, p<0.001). Within CD4 cell subset, percentages of activated and senescent cells did not differ between HIV + and controls, although PD1 expression tended to be up-regulated in HIV + children (overall, p=0.094). Conclusions: HIV-1-infected children exhibit a premature biological aging with accelerated immune senescence which affects the CD8 cell subset in particular. HIV-1 infection per se seems to influence the aging process, rather than exposure to ART for prophylaxis or treatment. 924 Background: HLA class I molecules are ligands for killer cell immunoglobin like receptors (KIR) that affect innate and adaptive immunity, and are of importance in controlling the antiviral response of NK cells. Despite their significance in the immune response, effects of KIR/HLA on HIV disease of children have not been previously studied. We hypothesized that the interactions of KIR and HLA alleles would affect HIV pathogenesis in children. Methods: 993 antiretroviral naïve children with symptomatic HIV infection from PACTG protocols P152 and P300 were genotyped for KIR and HLA alleles using Luminex platform. Linear regression with a robust variance estimator was used to test association between genotypes and baseline HIV RNA, CD4 + count, and cognitive score, adjusting for age, race/ ethnicity and study. Because the role of innate and adaptive immunity can differ by age, interaction between genetic markers and age group <2 years (n=460) and >2 years Ketty Gianesin 1 ; Antoni Noguera-Julian 2 ; Marisa Zanchetta 3 ; Osvalda Rampon 1 ; Clàudia Fortuny 2 ; Mireia Camós 2 ; Carlo Giaquinto 1 ; Anita De Rossi 1 1 University of Padova, Padova, Italy; 2 Hospital Sant Joan de Déu-Universitat de Barcelona, Barcelona, Spain; 3 Istituto Oncologico Veneto–IRCCS, Padova, Italy KIR/HLA Alleles Alter CD4 + Lymphocyte Count and Viral Load in HIV-Infected Children Kumud Singh 1 ; Min Qin 2 ; Sean Brummel 2 ; Konstantia Angelidou 2 ; RodneyTrout 1 ;Terrence Fenton 2 ; Stephen Spector 1 1 University of California San Diego, La Jolla, CA, US; 2 Harvard School of Public Health, Boston, MA, US

Poster Abstracts

554

CROI 2015