CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

weeks before and median 37 [IQR 24-48] weeks after the switch. PHQ-9 scores were higher in 48% of participants, lower in 31%, and unchanged in 21% after switching. The proportion of participants with at least moderate depression symptoms (PHQ-9≥10) rose from 9% to 16% (p=0.007), while the percentage of those with moderately severe symptoms (PHQ-9≥15) did not change (3% vs. 3%). The PHQ-9 sub-scores of somatic symptoms (sleep/appetite/energy level) had a more significant increase than that of cognitive/affective symptoms (p=0.005 vs. p=0.052). Multivariate analysis showed that viral suppression (Mean difference -2.9, 95%CI [-0.9 to -5.0], p=0.005) and higher PHQ-9 scores (Mean difference -2.7, 95%CI [-1.2 to -4.2], p<0.001) prior to DTG were linked to decrease in PHQ-9 score after DTG. NPZ-4, CD4+ T-cell counts and CD4/CD8 ratio improved after DTG (Table). Conclusion: DTG-associated NP adverse effects in this cohort were primarily related to somatic symptoms including insomnia, whereas there was no change in the prevalence of severe depressive symptoms or major depression.

participants with viral suppression prior to switch, neuropsychological and CES-D scores did not change significantly after switching off EFV. In Analysis 2, trends in neuropsychological and CES-D scores in the three different groups did not show significantly differences during a median of 3.2 years of follow up. Conclusion: Discontinuation of EFV is not associated with changes in neuropsychological performance or severity of depression in men. Furthermore, we did not observe differences among participants who were never on EFV, on EFV and then switched off and continuously on EFV.

442 MEMORY AND LEARNING DYSFUNCTION WITH INTEGRASE STRAND TRANSFER INHIBITORS USE Jane A. O’Halloran 1 , Sarah A. Cooley 1 , Jeremy Strain 1 , Robert Paul 2 , Rachel Presti 1 , Beau Ances 1 1 Washington University in St Louis, St Louis, MO, USA, 2 University of Missouri, Columbia, MO, USA Background: Integrase strand transfer inhibitors (INSTIs) have been associated with neuropsychiatric symptoms in post marketing analysis. However, limited data exists on the effect of these drugs on neurocognitive function. We assessed neurocognitive function and neuroimaging in people living with HIV (PLWH) on INSTI-based regimens Methods: We performed a cross-sectional analysis of PLWH on ART aged >18 years. PLWH with<=”” div=””> Results: Of 202 PLWH, median (IQR) age 55 (48, 60) years, 152 (75%) male, 136 (67%) black, median recent CD4+ T cell count 576 (401, 818) cells/mm3, 96% HIV RNA <200 copies/ml, 99 (49%) were on INSTI-based ART (40.4% raltegravir, 29.3% elvitegravir, 30.3% dolutegravir), while 103 (51%) were on non-INSTI based ART. No between group difference was identified in age, sex, race, education, smoking status, depression scores, psychiatric medication use, HIV infection duration or nadir and recent CD4+ T cell count. On neuropsychological assessment, PLWH in the INSTI group had worse learning/memory domain scores compared with those on non-INSTI based ART (1.25 (0.25, 2) versus 0.25 (0, 1.8); p=0.02) (figure a). This remained significant when corrected for nadir CD4+ T cell count, current ART regimen duration, current NNRTI or PI use, current psychiatric medication use or having previously received another ART regimen (p=0.04). A composite volumetric measurement for learning/memory brain regions was significantly smaller in the INSTI group compared to non-INSTI group (figure b). Functional connectivity was reduced in the memory resting state network in the INSTI group compared to the non-INSTI group (figure c). Conclusion: We demonstrated worse learning/memory neuropsychological performance; smaller volumetric measurement in regions associated with learning/memory and decreased functional connections in the resting state memory network in those on INSTI-based regimens. Our results suggest a dysfunction in the learning/memory network but prospective studies are required to explore this further.

Poster Abstracts

441 NEUROPSYCHOLOGICAL CHANGES IN EFAVIRENZ SWITCH REGIMENS IN MACS Yijia Li 1 , Zheng Wang 1 , Yu Cheng 1 , James T. Becker 1 , Eileen Martin 2 , Andrew Levine 3 , Leah H. Rubin 4 , Ann B. Ragin 5 , Ned Sacktor 4 , University o. Pittsburgh 1 1 University of Pittsburgh, Pittsburgh, PA, USA, 2 Rush University, Chicago, IL, USA, 3 University of California Los Angeles, Los Angeles, CA, USA, 4 Johns Hopkins University, Baltimore, MD, USA, 5 Northwestern University, Chicago, IL, USA Background: Efavirenz (EFV) has been associated with side effects involving the central nervous system including depression, vivid dreams and sleep impairment. However, it remains largely unknown whether switching off EFV improves neuropsychological performance and depression. Methods: We utilized data from the Multicenter AIDS Cohort Study (MACS). Participants were categorized by their use of EFV: never on EFV (No EFV), on EFV then switched off (Switch-OFF) and continuously on EFV (No Switch-OFF). Baseline time point was defined as visits when first neuropsychological data were available. In the first analysis, we compared neuropsychological and Center for Epidemiologic Studies Depression (CES-D) scores before and after EFV switch in Switch-OFF group, aligning participants at the time of switch. Linear mixed effects models were fitted to examine the change of each of scores before and after switching. In the second analysis, we compared the trajectory of neuropsychological/CES-D scores among the three groups. Results: This analysis included 1,989 HIV-seropositive participants with neuropsychological data, with 1,675 participants in No EFV group, 270 in Switch-OFF group, and 44 in No Switch-OFF group. Participants at baseline had a median age of 37 years, median CD4 cell count 442 cells/μL, and 22.9% of viral suppression rate. Prevalence of HIV-associated dementia was 3.24%, 4.09% and 2.78% in No EFV, Switch-OFF, and No Switch-OFF groups (P=0.197). Baseline CES-D scores were 9, 9 and 6 in No EFV, Switch-OFF, and No Switch-OFF groups (P=0.16). In Analysis 1, viral suppression rate before switch was 62.8% and median CD4 cell count was 548 cells/μL in Switch-OFF group; neuropsychological T scores and CES-D scores did not show clinically significant changes over 1.5 years prior to and 1.5 years after switch for each of neuropsychological domains (Table). In a sensitivity analysis only including

CROI 2019 160