CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

1 University of North Carolina at Chapel Hill, Chapel Hill, NC, USA, 2 Asociacion Civil Impacta Salud y Educacion, Lima, Peru, 3 Fred Hutchinson Cancer Research Center, Seattle, WA, USA, 4 University of Washington, Seattle, WA, USA, 5 Yale University, New Haven, CT, USA, 6 University of California San Francisco, San Francisco, CA, USA Background: The CNS is a sanctuary and reservoir for HIV, which is known to enter the CNS within days of primary infection. We sought to assess whether early ART would improve neurocognitive performance (NP) in the Sabes study, a cohort of men who have sex with men (MSM) and transgender women (TW) in Lima, Peru randomized to start ART at diagnosis (within 30-90 days of estimated date of detectable infection (EDDI)) vs. after a short delay. We hypothesized that, by limiting CNS infection, ART initiation within 30 days of HIV acquisition would improve NP compared to ART initiation later in early infection. Methods: A subset of Sabes participants had neurocognitive assessments and blood collection, and in some cases, lumbar puncture. NP was measured with a 15-test battery covering Gross motor, Attention, Executive, Learning, Memory, Speed of Processing, and Fine Motor domains at weeks 12, 24, 48, 72, 96 and 120 after randomization. Estimated date of infection (EDDI) was derived from an algorithm compiling test parameters of last negative and first positive HIV tests. Results: The 112 participants were all hispanic MSM or TW, had a mean age of 26.4 years (SD=7.4), mean education grade level of 12.5 (SD=2.3) and mean baseline CD4+ cell count of 443.4 (SD=210.4). Seventy-seven observations came from participants with an EDDI to ART initiation interval of <30 days, 190 from participants who started ART after 31-90 days, and 262 from those who started ART 91 to 249 days after EDDI (>90 days). NP did not differ significantly between the EDDI to ART categories over time (mean total z for <30 = .42, 31-90= .39, >90=.32; p=ns). However, NP significantly improved with ART out to 120 weeks of follow up (F=(5, 394)=35.9, p<.0001), across categories (mean total z score at week 0=0.09, week 12=0.28, week 24=0.41, week 48=0.47, week 72 = 0.50, week 96=0.52). As a check for practice effects, gait (resistant to practice) was significantly improved over time (F(5.391)=2.94, p<.05), indicating that ART had a substantial impact on NP. Conclusion: In this unique early infection cohort, time between primary infection and ART initiation was not associated with neurocognition. Initiation of ART improved neurocognitive functioning regardless of treatment category: this cohort of persons who started ART during or just following acute infection had improved cognitive performance as time on ART increased. These findings underscore the importance of initiating ART early to protect the CNS.

1 University of North Carolina at Chapel Hill, Chapel Hill, NC, USA, 2 University of Gothenburg, Gothenburg, Sweden, 3 Yale University, New Haven, CT, USA, 4 University of Torino, Torino, Italy, 5 San Raffaele Scientific Institute, Milan, Italy, 6 University of California San Francisco, San Francisco, CA, USA Background: Some HIV-infected individuals presenting with neurologic symptoms have increased levels of cerebrospinal fluid (CSF) HIV-1 RNA despite being on antiretroviral therapy (ART) and having undetectable, or low, plasma viral load (VL)-i.e. neurosymptomatic (NS) CSF escape. Genetic diversity and phenotypic characteristics of NS CSF escape viruses have not been previously examined, and the cells producing these populations are unknown. Methods: We examined archived blood plasma and CSF samples from 11 individuals with NS CSF escape. All individuals were ART-treated and had a CSF VL >40 copies/ml and greater than the VL in plasma. Single genome amplification (SGA) and/or Illumina MiSeq deep sequencing with Primer ID were used to assess diversity in env and drug resistance in pro-pol. Full-length env genes were cloned from the CSF of three individuals and assessed for macrophage tropism based on their ability to efficiently enter cells with a low density of CD4 on their surface. Results: For these 11 participants, median values were: CSF VL = 1,493 copies/ml, plasma VL = 163 copies/ml, blood CD4 count = 552 cells/μl and CSF WBC = 44 cells/ μl. 73% (8/11) of participants had a genetically diverse CSF escape HIV-1 population. 67% of individuals examined for drug resistance (4/6) had mutations in their CSF virus conferring at least partial resistance to their current ART regimen. 5 of 6 participants experienced an improvement in neurologic symptoms upon ART optimization. Four individuals were examined longitudinally and three had persistent CSF escape. The three participants examined for viral tropism had CSF HIV-1 variants that were adapted to entering CD4+ T cells rather than macrophages (i.e. R5 T cell-tropic). Conclusion: Most individuals with symptomatic CSF escape have characteristics that are consistent with ongoing viral replication such as genetically diverse CSF viral populations, CSF drug resistance and resolution of neurologic symptoms after ART optimization. The results here suggest that NS CSF escape virus can be adapted to entering T cells and is likely produced by CD4+ T cells in the CNS (Example shown in Fig. 1). It remains unknown whether these infected cells represent long-lived viral reservoirs in the CNS or transient populations producing virus during treatment failure due to drug resistance or nonadherence.

Poster Abstracts

446 NEUROSYMPTOMATIC HIV CSF ESCAPE CAN BE PRODUCED BY REPLICATION IN T CELLS IN THE CNS

447 RELAPSE OF SYMPTOMATIC CSF HIV ESCAPE UPON PREVIOUSLY OPTIMIZED cART REGIMEN CHANGES Francesca Ferretti, Valentina De Zan, Filippo Turrini , Enzo Boeri, Simonetta Gerevini, Nicola Gianotti, Hamid Hasson, Adriano Lazzarin, Cinque Paola San Raffaele Scientific Institute, Milan, Italy Background: Neuro-symptomatic cerebrospinal fluid (CSF) viral escape is a condition of persons receiving combination antiretroviral treatment (cART), who show a discordant HIV replication between CSF and plasma, associated with

Laura P. Kincer 1 , Magnus Gisslén 2 , Serena S. Spudich 3 , Mattia Trunfio 4 , Andrea Calcagno 4 , Cinque Paola 5 , Ronald Swanstrom 1 , Richard W. Price 6 , Sarah B. Joseph 1

CROI 2019 162