CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

for demographic factors, HCV co-infection and HIV acquisition risk, higher BIL levels at the time of reporting CNS AE were independently associated with 6.5% increased risk of developing CNS AE (OR=1.065 (95% CI (1.001, 1.134) p=0.04)). Conclusion: The prevalence of CNS AE with DTG was higher than clinical trials had suggested but in line with those reported with real world use. BIL levels were independently associated with increased risk of developing CNS AE. How BIL levels relate to DTG levels or UGT1A1 activity requires further investigations. 437 ART INITIATED AT HIGH CD4 NADIR DOES NOT NORMALIZE CSF MARKERS OF IMMUNE ACTIVATION Frida Rydberg 1 , Aylin Yilmaz 1 , Lars Hagberg 1 , Dietmar Fuchs 2 , Magnus Gisslén 3 , for the HIV Clinical Research Center, University of Gothenburg Background: HIV infects cells in the central nervous system (CNS), mainly microglia and perivascular macrophages, and induces a chronic intrathecal immune activation. Similar to its effect outside the CNS, antiretroviral treatment (ART) substantially decreases CNS inflammation and CD4+ T-cell trafficking through the cerebrospinal fluid (CSF) is often reduced to near normal levels. Yet, CSF levels of neopterin, a pteridine marker of primarily macrophage/microglia activation, have been found to be stably increased in the majority of persons living with HIV (PLWH) who begin treatment during the chronic phase of HIV infection when the immune function is impaired. By contrast, CSF neopterin is essentially normalized when ART is initiated early, during acute HIV infection (AHI). The aim of this study was to evaluate if CSF immune activation biomarkers normalize to a larger extent in PLHIV with chronic HIV who start ART at high, as compared to starting treatment at low CD4-cell counts. Methods: 176 neuroasymptomatic patients who started ART during chronic HIV were retrospectively included from the longitudinal prospective Gothenburg CSF cohort study and followed for in median 5.0 years (mean 6.1 years). Lumbar punctures were performed at baseline before ART, after 1, and >3 years. Twenty-five participants had a CD4 nadir <50; 52 between 50 and 199; 61 between 200 and 349; 22 between 350 and 499; and 16 ≥500 cells/ µL. Neopterin concentrations were measured using a commercially available immunoassay (NEOPT-SCR.EIA 384 Det., Thermo Fisher Scientific – BRAHMS GmbH, Henningsdorf, Germany) with an upper normal reference value of 5.8 nmol/L in CSF. Results: A significant inverse correlation between CD4 cell count and CSF neopterin was found at baseline (r = -0.25, p < 0.01) while no correlations between CD4 nadir and CSF neopterin were found after 1, or >3 years ART. 15% of participants with the highest CD4 nadir (>500) had normal CSF neopterin (<5.8 nmol/L) compared to 0% of those with the lowest CD4 nadir (<50). After >3 years of ART, 57% and 50% respectively had normal CSF neopterin. Conclusion: CSF Neopterin does not normalize in many patients initiating ART during chronic HIV. This also applies to ART-initiation at high CD4 cell counts. 438 INTEGRASE INHIBITOR START OR SWITCH IMPACTS LEARNING IN WOMEN WITH HIV Jane A. O'Halloran 1 , Kunbo Wang 2 , Dionna W. Williams 3 , Raha Dastgheyb 3 , Kathryn Fitzgerald 3 , Asante R. Kamkwalala 3 , Amanda B. Spence 4 , Anjali Sharma 5 , Pauline M. Maki 6 , Deborah Gustafson 7 , Kathleen M. Weber 8 , Adaora Adimora 9 , Margaret Fischl 10 , Yanxun Xu 3 , Leah H. Rubin 3 1 Washington University in St Louis, St Louis, MO, USA, 2 Johns Hopkins University, Baltimore, MD, USA, 3 Johns Hopkins University School of Medicine, Baltimore, MD, USA, 4 Georgetown University, Washington, DC, USA, 5 Albert Einstein College of Medicine, Bronx, NY, USA, 6 University of Illinois at Chicago, Chicago, IL, USA, 7 SUNY Downstate Medical Center, Brooklyn, NY, USA, 8 Cook County Health & Hospitals System, Chicago, IL, USA, 9 University of North Carolina at Chapel Hill, Chapel Hill, NC, USA, 10 University of Miami, Miami, FL, USA Background: In recent years, the integrase strand transfer inhibitor (INSTI) class of antiretroviral therapy (ART) has become an integral component of HIV treatment. Despite concerns regarding neuropsychiatric adverse events there are limited data on cognitive side effects, particularly in women with HIV (WWH). Methods: WWH enrolled in the Women’s Interagency HIV Study (WIHS), who started or switched to INSTI-based ART and had completed one comprehensive neuropsychological (NP) test battery before and after the start/switch, were included. The NP battery assessed learning, memory, fluency, attention/ 1 Sahlgrenska University Hospital, Gothenburg, Sweden, 2 Innsbruck Medical University, Innsbrusk, Austria, 3 Sahlgrenska Academy at the University of Gothenburg, Gothenburg, Sweden

working memory, executive function, processing speed, and motor function. The primary NP outcomes were demographically-corrected T-scores (M=50, SD=10) for each cognitive domain. Linear mixed effects models adjusted for relevant covariates (e.g., age, race, education, income, substance use, body mass index, HIV RNA) were used to examine the effect of start/switch of any INSTI as well as each individual drug within the INSTI class on NP function. Results: 628 WWH, median age 48 (interquartile range 36, 60) years, 65% black non-Hispanic, had NP data before and after INSTI start/switch. While 14% started INSTI-based ART, the remainder switched primarily from protease inhibitor (PI)-based ART (51%) or a non-nucleoside reverse transcriptase inhibitor (NNRTI)-based ART (27%). Raltegravir (RAL), elvitegravir (EVG), and dolutegravir (DTG) were introduced in 38%, 24% and 38% of WWH, respectively. Overall, any INSTI use was associated with poorer performance in learning after start/switch (p<0.001). Specifically, use of EVG (p=0.02) and DTG (p=0.002), but not RAL, was associated with poorer learning. In analyses restricted to INSTI switch, any INSTI use was associated with poorer performance in learning (p<0.009), as was use of DTG specifically (p=0.004). INSTIs and DTG remained associated with poorer learning among those switching from a PI-based regimen. DTG also remained associated with poorer learning among those switching from an NNRTI (p’s<0.05). Switching from an NNRTI to an INSTI was also associated with better processing speed. Conclusion: Switching or starting an INSTI was primarily associated with poorer performance in learning among WWH. These changes were mainly observed in EVG and DTG users, and not with RAL, indicating that the impact of INSTI on cognition in WWH may not be a class effect. Oluwakemi Okwuegbuna 1 , Albert Anderson 2 , Jennifer Iudicello 1 , Josue Perez-Santiago 3 , Florin Vaida 1 , Rachel Schrier 1 , Allen McCutchan 1 , Ronald J. Ellis 1 , Scott L. Letendre 1 1 University of California San Diego, San Diego, CA, USA, 2 Emory University, Atlanta, GA, USA, 3 University of Puerto Rico, San Juan, Puerto Rico Background: HIV RNA in blood substantially differs between individuals without antiretroviral therapy (ART) due, in part, to differences in the immune response, such as endogenous interferons. Elite controllers suppress HIV RNA without ART but are at greater risk for vascular and central nervous system (CNS) complications than PWH on suppressive ART, possibly due to their robust immune response to HIV. HIV RNA in CSF also substantially differs between PWH without ART but relatively little is known about the effects of the antiviral immune response on CNS health trajectory. Methods: The project aimed to determine a) the correlates of HIV RNA in CSF in 1,008 PWH without ART and b) the association between HIV RNA in CSF without ART and cognition or depression over time with ART (1,555 assessments in 300 PWH). All participants had plasma HIV RNA ≤ 200 copies/mL and were comprehensively assessed with neuropsychological (NP) testing, Beck depression inventory (BDI), and lumbar puncture. Statistical methods included univariable and stepwise multivariable regression using Bayesian Information Criterion and false discovery rate correction, recursive partitioning, and mixed models. Results: Participants were mostly middle-aged (mean 39 years), European ancestry (50.4%) men (83.1%) with a mean duration of HIV of 7.5 years. Without ART, HIV RNA in CSF was ≤ 50 copies/mL in 161 (16.0%) and was less than HIV RNA in blood in 95% (median difference -1.4 log 10 copies/mL, range -4.8 to +1.3). Multivariable regression identified that higher HIV RNA in CSF was associated with higher HIV RNA in blood, higher CSF leukocyte count, fewer CD4+ T-cells, higher CD4+ and CD8+ percent, lower serum albumin, higher total protein in CSF and blood, and lower CSF glucose (model R2=0.27, p<0.0001). Recursive partitioning identified that four variables explained 50% of the variance in HIV RNA in CSF (Figure). PWH who had lower HIV RNA in CSF without ART had worse BDI values (p=0.034) over time while on ART (but not worse NP performance), even after accounting for demographic, disease, and treatment covariates (model p<0.0001). Conclusion: The relationship between HIV RNA in CSF and blood is highly variable with 1 in 6 having undetectable HIV RNA in CSF without ART and 1 in 20 having HIV RNA in CSF higher than HIV RNA in blood. PWH who better control HIV RNA in CSF without ART have more depressive symptoms on ART, which could reflect bystander injury from a more effective antiviral immune response.

Poster Abstracts

439 HIV RNA IN CEREBROSPINAL FLUID OFF ART PREDICTS MORE DEPRESSIVE SYMPTOMS ON ART

CROI 2020 153