CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

401 HIV Compartmentalization in the CNS Is AssociatedWith Neurocognitive Impairment Natalie M. Bowman 1 ; Sarah B. Joseph 1 ; Laura P. Kincer 1 ; SarahYosief 1 ; Serena S. Spudich 2 ; Joseph J. Eron 1 ; RichardW. Price 3 ; Ronald I. Swanstrom 1 ; Kevin R. Robertson 1 ; for theTHINC Group 1 Univ of North Carolina at Chapel Hill, Chapel Hill, NC, USA; 2 Yale Univ Sch of Med, New Haven, CT, USA; 3 Univ of California San Francisco, San Francisco, CA, USA Background: The effect of compartmentalized HIV replication in the central nervous system (CNS) on neurocognitive performance (NP) is still not well characterized. Low level HIV replication in the CNS may sustain inflammation causing lasting neuronal damage. We examined the effect on NP of compartmentalization of HIV between the blood and CNS in a cohort of treatment-naïve patients. Methods: We recruited HIV+ treatment-naïve subjects starting ART with CD4<400. Blood and cerebrospinal fluid (CSF) were collected at entry and 2-4 weeks and 54 weeks after initiating antiretroviral therapy (ART) and analyzed for HIV viral load (VL) and CD4 count. Comprehensive NP assessments were performed at enrollment and 24 weeks and 54 weeks after initiating ART. We genetically characterized virus in CSF and blood at entry to evaluate for compartmentalized lineages in the CNS (see figure). cDNA was generated from viral RNA, and partial env genes were amplified and sequenced by Illumina with Primer ID to quantify the number of templates examined and/or full length env

genes were amplified and sequenced by single genome amplification. Primary outcome was NP. Compartmentalized and equilibrated groups were compared using Wilcoxon rank-sum test and multivariable generalized models. Results: Of 28 subjects whose viral populations were genetically characterized, 10 (36%) had compartmentalized HIV in the CNS pre-ART. Subjects with and without compartmentalization did not differ in demographic variables or mean baseline VL, CSF VL, and CD4 or viral decay rates in the blood and CSF after starting ART. Subjects with CSF compartmentalization had numerically greater NP impairment than subjects lacking compartmentalization pre-ART (global deficit score 1.31 vs. 0.87, p=0.2), and this difference was significant 24 weeks (1.31 vs. 0.63, p=0.04) and 54 weeks (1.25 vs. 0.59, p=0.05) after ART initiation. Conclusions: Compartmentalized HIV lineages in the CNS were common in treatment-naïve subjects. Despite the fact that compartmentalization was not associated with baseline VL, CD4 count, or demographic characteristics, subjects with compartmentalized HIV trended towards impaired NP at baseline compared to non-compartmentalized subjects, and this difference was significant at 24 and 54 weeks after ART initiation. Compartmentalization was associated with NP impairment, and improvement on ART may be blunted in those with compartmentalized virus suggesting that local viral replication may have an impact on the CNS that is less reversible.

402 Persistent HIV-1 in the CNS During Therapy: Evidence of a Viral Reservoir in the CNS Sarah B. Joseph 1 ; Laura P. Kincer 1 ; Natalie M. Bowman 1 ; SarahYosief 1 ; Angela Kashuba 1 ; Kevin R. Robertson 1 ; Serena S. Spudich 2 ; Joseph J. Eron 1 ; RichardW. Price 3 ; Ronald I. Swanstrom 1 1 Univ of North Carolina at Chapel Hill, Chapel Hill, NC, USA; 2 Yale Univ Sch of Med, New Haven, CT, USA; 3 Univ of California San Francisco, San Francisco, CA, USA Background: In some untreated subjects, HIV-1 forms diverse, CNS-specific lineages with an enhanced ability to infect myeloid-lineage cells. It is unknown whether these viral lineages persist during antiretroviral therapy (ART). Methods: We examined Cerebrospinal fluid (CSF) viral escape in a cohort of 96 HIV-infected subjects on ART for at least one year. All subjects had neurocognitive assessments. CSF and blood plasma samples were collected from each subject and examined for HIV-1 RNA and drug concentrations. CSF escape was defined as having HIV-1 RNA concentrations ≤ 40 cp/ml in blood plasma and ≥ 40 cp/ml in CSF. Subjects whose CSF viral load exceeded 100 cp/ml were examined at a follow-up visit. Viral RNA was isolated from the CSF of these subjects and converted to cDNA. Single genome amplification (SGA) was used to amplify full-length env genes for sequencing and cloning. Pseudotyped viruses were generated from the cloned env genes and used in entry assays (3 replicates per clone) to determine whether the encoded Env proteins were well-adapted to entering cells expressing low levels of CD4 (i.e. were macrophage-tropic). Results: Four of 96 subjects had CSF escape (4%; 47, 89, 265 and 1295 cp of RNA/ml of CSF). Subjects with and without CSF escape did not differ in their neurocognitive performance (overall global deficit score; Wilcoxon rank-sum, p=0.64). Two subjects had follow-up samples; after 9 months one subject was suppressed (transient escape) but after 8 months the second subject still had 356 cp of RNA/ml of CSF (persistent CSF escape). Analyses of env genes amplified from the CSF revealed that the subject with transient CSF escape had a homogeneous viral population that was poorly adapted to entering macrophages (Fig. 1A). In contrast, the subject with persistent CSF escape had a genetically diverse viral population with an elevated ability to infect macrophages (Fig. 1B) at a level that is virtually never detected in blood-derived viruses. There was no association between CSF escape in these subjects and low CSF drug concentration. Conclusions: HIV-1 can persist in the CNS during suppressive antiretroviral therapy. Trafficking of an infected T cell into the CNS or activation of a small number of infected resident cells could generate a transient, homogeneous viral population in the CNS. In contrast, our observation of genetically diverse CSF viral populations eight months apart is consistent with the production of HIV-1 by a persistently infected resident cell population.

Poster Abstracts

Figure 1: Phylogenetic and analysis of CD4 entry phenotype of transient (A) and persistent (B) viral populations in the CSF of two subjects with CSF viral escape.

1A. Transient CSF escape Homogeneous, R5 T cell-tropic viral population

1B. Persistent CSF escape Diverse viral population with elevated macrophage tropism

Phenotypic controls

R5 T cell-tropic controls (negative controls)

Macrophage- tropic controls (positive controls)

3025A4 3025A9 3025A8 3025 C10 3025A1 3025 E4 3025 E2 3025 E10 3025 B5 3025 C12 3025 E12

3026 T2 F2 3026 T2D12 3026 T2E9 3026T2A11Final

81

10 20 30 40

40

87

82

3026 T2 F10 3026 T2H8

100

3026 T1H3

30

3025 F1 3025 E1 3025 B8 3025A7 HXB2gp160

3026 T1C8

Axis

HXB2gp160

A B C D

0.01

.01

0.01

20

0.01

0 Percent infectivity at low CD4 10

Percent Infectivity at Low CD4

E4 E10 E12 D12 E9 F10 A11 A7 E4 E10 E12 12 E9 F10 A11

4013C 4051C 4059C 5002C 7115C Ba−L 4013P 4051P 4059P 5002P 7115P JRCSF D7

152

CROI 2016