CROI 2016 Abstract eBook

Abstract Listing

Oral Abstracts

Results: We found, one hour after SPIONs were instilled at the cisterna magna, they were present throughout the perivascular space mainly as extracellular single beads with some being phagocytized by perivascular cells (PVCs). By 24 hours post inoculation all the SPIONs were intracellular and over the next 28 days the frequency of cells containing SPIONs fell throughout the CNS with the exception of the cranial nerves. There were larger numbers of SPIONS in the CNS of animals with SIV infection than non infected animals. Beginning at 7 days after intracisternal instillation of SPIONs we found SPION containing cells outside of the CNS with the greatest number in the cervical lymph node 7 days after SPION administration. This timing was consistent in both SIV-infected and control animals. We determined that SPIONs are carried out by cells (PVCs) and not in the fluid phase traced with dextran red. In a normal animal the cervical lymph node, with a volume of 0.5 ml, the number of SPIONs carrying cells is a million showing that there is a relative large CNS efflux of cells. We identified several points for emigration out of the CNS as those at the cribriform plate, the cranial nerves, along the neural cord at the ganglia at the brachial plexus and the sacral cord. Conclusions: The SPION+ cells were consistently CD163+ indicating that they were monocyte/macrophage lineage cells. Interestingly, rare SPION+CD163+ cells were also SIV+. This experiment shows for the first time that brain perivascular cells (macrophage lineage) leave the CNS into the periphery and may carry virus with the possibility of reseeding the body with potentially new viral clones. 142 CSF Lymphocyte and Monocyte Activation and Trafficking in Primary HIV Infection Xiang Li 1 ; Fangyong Li 2 ; AnjiYi 2 ; Julia Peterson 3 ; Brinda Emu 1 ; RichardW. Price 3 ; Elizabeth Sinclair 3 ; Serena S. Spudich 1 1 Yale Univ Sch of Med, New Haven, CT, USA; 2 Yale Univ Sch of PH, New Haven, CT, USA; 3 Univ of California San Francisco, San Francisco, CA, USA Background: Trafficking of immune cells to the central nervous system is hypothesized to facilitate HIV entry and immune-induced neuronal injury, and is mediated by surface proteins such as chemokine receptors and α4 integrin. We longitudinally assessed immune cell activation and surface marker expression in cerebrospinal fluid (CSF) and blood and their relationship with CSF HIV RNA during primary HIV infection (PHI) before and after combination antiretroviral therapy (cART). Methods: Longitudinal paired blood and CSF were obtained in initially cART-naïve PHI (<12 mo since infection) participants; some subjects independently initiated cART during follow up. Multiparameter flow cytometry on fresh samples was used to determine activation (% CD38+HLADR+) and chemokine receptor expression (% CCR5+CXCR3+) on CD4+ and CD8+ T cells, and activation (% CD14+CD16+) and α4 integrin expression (% and mean fluorescence intensity (MFI) of CD49d+) on monocytes. CSF chemokines (IP-10 and MCP-1) were quantified by ELISA. Analyses employed Spearman correlation, within-subject correlation, and linear mixed models. Results: 51 participants enrolled at a median 3.3 mo post infection had 168 total visits (113 untreated, 55 on cART) with median 7 mo follow up (range 0-40). Baseline CSF IP-10 but not MCP-1 correlated with CD4+ and CD8+ T cell activation and HIV RNA in CSF (r=0.48, 0.59, 0.77; all p<0.05). Pre-cART, rates of increase in T cell activation were 4 times higher in CSF than blood. In unadjusted longitudinal analysis, CSF CD4+ and CD8+ T cell activation correlated with CSF HIV RNA (all p≤0.01); in multivariate analysis CSF CD4+ but not CD8+ T cell activation was an independent predictor of CSF HIV RNA. CSF monocyte activation and α4 expression did not correlate with CSF HIV RNA. In blood but not CSF of untreated participants, monocyte α4 MFI correlated with CD4+ and CD8+ T cell activation (all p<0.05). During follow up on cART, blood but not CSF T cell activation declined with days on treatment (slope=-0.06, p=0.001). During cART, blood monocyte α4 MFI correlated with blood T cell activation, and CSF monocyte α4 MFI correlated with CSF CD4+ T cell activation (all p<0.05). Conclusions: In untreated PHI, T cell activation increases faster in CSF than blood, and CSF CD4+ T cell activation but not monocyte activation correlates with CSF HIV RNA. Intrathecal T cell activation does not decline during early follow up on cART. The role of α4 integrin in trafficking leukocytes to the CSF warrants further exploration. 143 Compartmentalized HIV DNA Populations Persist in CSF Despite Suppressive ART Michelli Faria de Oliveira ; Antoine Chaillon; Scott R. Letendre; Matt F. Strain; Ron Ellis; Sheldon R. Morris; Susan J. Little; Davey M. Smith; Sara Gianella Univ of California San Diego, San Diego, CA, USA Background: Persistence of HIV DNA in the central nervous system likely contributes to inflammation, brain damage and neurocognitive (NC) impairment during antiretroviral therapy (ART). The effects of early ART initiation on these parameters are still unknown. Methods: Paired blood and 40 mL cerebrospinal fluid (CSF) samples were collected from 16 HIV+ individuals on suppressive ART (<50 copies/ml): 9 subjects started ART ≤4 months from estimated date of infection (EDI) and 7 started ART >14 months from EDI. NC functioning was measured by Global Deficit Score (GDS). HIV DNA levels were measured in peripheral blood mononuclear cells (PBMC) and CSF cells by droplet digital PCR; soluble inflammatory markers (sCD163, IL-6, MCP-1, TNF-α) and marker of neuronal damage (neurofilament chain [NFL]) were measured in blood plasma and CSF supernatant by immunoassays. Next generation sequencing (NGS) data by Roche 454 were successfully generated for HIV env from 8 paired blood and CSF cell pellets (3 with early and 5 with late ART). Viral compartmentalization analysis via Fst statistics was performed using NGS data and repeated using representative haplotypes to guard against possible skewing of allelic frequencies due to PCR amplification and other biases. Cross-sectional comparisons between groups (early versus later ART) were performed using non-parametric statistical analysis. Results: In these suppressed HIV+ individuals (median duration on ART: 2.6 years), HIV DNA was detected in 62.5% (10/16) of CSF cell pellets and 93.8% (15/16) of PBMCs. Early initiation of ART was associated with lower CSF levels of IL-6 (p=0.03) and TNF-α (p=0.02), but no difference in GDS, NFL, or HIV DNA as compared to later ART group. Significant compartmentalization of HIV DNA populations between blood and CSF were detected in 7 out of 8 subjects, and these findings were congruent between the two approaches mentioned above. Phylogenetic analysis confirmed presence of monophyletic HIV DNA populations within the CSF for each participant (aLRT > 0.9), and their persistence over time in the two participants with longitudinal sampling (2 and 5 months between time points). Conclusions: A compartmentalized HIV DNA population in CSF was detectable in the majority of HIV+ individuals despite long-term suppressive ART and even when ART was started during early HIV-infection. Also, early ART start was associated with lower inflammation in CSF (IL-6 and TNF-α), but no difference in GDS compared to later start of ART. 144 Mitochondrial DNA Copy Number and Neurocognitive Impairment in HIV-Infected Persons David Samuels 1 ; Asha R. Kallianpur 2 ;Yan Guo 3 ;ToddT. Brown 4 ; Sanjay R. Mehta 5 ; Ron Ellis 5 ; Scott R. Letendre 5 ;Todd Hulgan 3 ; for the CHARTER Study Group 1 Vanderbilt Univ Sch of Med, Nashville, TN, USA; 2 Cleveland Clinic/Lerner Rsr Inst, Cleveland, OH, USA; 3 Vanderbilt Univ, Nashville, TN, USA; 4 Johns Hopkins Univ, Baltimore, MD, USA; 5 Univ of California San Diego, San Diego, CA, USA Background: Mitochondrial DNA (mtDNA) content in peripheral blood mononuclear cells (PBMC) declines with age and has been associated with neurocognitive function in non- HIV-infected persons. LowmtDNA copy number may indicate disordered mtDNA replication and high copy number may indicate a cellular response to mitochondrial dysfunction. Extracellular (or “free”) mtDNA is a TLR-9 ligand affecting the innate immune response. We tested relationships between both PBMC cellular mtDNA content and cerebrospinal fluid (CSF) cell-free mtDNA levels and neurocognitive impairment in the CNS HIV Antiretroviral Therapy Effects Research (CHARTER) study. Methods: PBMC mtDNA content was measured in 1011 CHARTER participants. The effect of PBMC mtDNA content on global deficit score (GDS), GDS impairment (GDS≥0.5), and HIV-Associated Neurocognitive Disorder (HAND) were tested by logistic regression, adjusting for platelet count, age, gender, genetic ancestry, comorbidity (incidental or contributing to neurocognitive impairment), nadir CD4 and HIV RNA in plasma. CSF free mtDNA was assessed by droplet digital PCR in a subset of 335 participants. Cell-free mtDNA associations with CSF inflammation and iron-related biomarkers CXCL10, IL-6, IL-8, TNF-a, transferrin (TF), ceruloplasmin (CP), and vascular endothelial growth factor (VEGF), HIV viral load (VL), and GDS were evaluated. Results: Lower PBMC mtDNA copy number per cell was associated with lower platelet count (p=3e-7), older age (p=0.002) and longer ART duration (p=0.0008). PBMC mtDNA content was associated with GDS (p=0.02), GDS impairment (p=0.0003) and HAND (p=0.009). CSF free mtDNA levels were positively associated with CSF CXCL10 (p<0.001) and TNF-a (p<0.05). After adjusting for CSF WBC and VL, free mtDNA levels were associated with CSF inflammation- and iron-related biomarkers TF (p<0.05) and CP (p<0.05). With correction for ART, free mtDNA was associated with CSF VEGF (<0.05) and IL-6 (p=0.05). Unlike PBMC mtDNA, CSF free mtDNA was not associated with age or NCI.

Oral Abstracts

53

CROI 2016