CROI 2025 Abstract eBook

Abstract eBook

Poster Abstracts

(p=0.028, r=0.76) and was consistently associated with a trend of greater BPND decline in all GM ROIs. Conclusions: In this preliminary analysis of 8 PWH, a significant and widespread decline in BPND was observed in GM ROIs over 2.7 years despite suppressive ART. A larger sample size, plus a longitudinal analysis of PWoH, cognitive assessments and neuronal injury markers will further validate and help understand the nature of BPND reduction.

597

Inflammatory Biomarker Profiling to Predict Neuroinflammation and Depression in Adolescents With HIV Shalena Naidoo 1 , Arish Mudra Rakshasa-Loots 2 , Kaylee van Wyhe 1 , Frances Robertson 3 , Ernesta Meintjes 3 , Mark Cotton 4 , Barbara Laughton 1 , Richard Glashoff 1 1 Stellenbosch University, Cape Town, South Africa, 2 University of Edinburgh, Edinburgh, UK, 3 University of Cape Town, Cape Town, South Africa, 4 Family Clinical Research Unit, Tygerberg, South Africa Background: Despite early antiretroviral therapy (ART), individuals with perinatally acquired HIV often show neuroinflammation and a heightened risk of depression as they approach early adulthood. Longitudinal monitoring of inflammatory markers may provide insights into these outcomes, aiding in the prediction of neuroinflammatory processes and mental health risks. We examined these associations longitudinally in cohorts from birth to early adulthood. Methods: Thirty-nine adolescents with HIV (56% female) from the Children with HIV Early Antiretroviral Therapy (CHER) trial were included. The median age was 16 years (IQR: 15–16), with 89.4% achieving virological suppression. Depression was assessed using the nine-item Patient Health Questionnaire (PHQ-9). Magnetic resonance spectroscopy (MRS) measured brain metabolites (choline, myo-inositol) in the basal ganglia (BG), midfrontal gray matter (MFGM), and peritrigonal white matter (PWM). Plasma immune biomarkers (41 total) were measured at six time points: pre-ART, 1, 2, 3-4, 4-5 years post-initiation, and at 8 years. A subset was also assessed for integrase cell-associated HIV-1 DNA (iCAD), a marker of HIV-1 persistence. Correlations between biomarkers, brain metabolites, and depression scores were explored using Pearson and Spearman correlations. Results: Pre-ART hsCRP levels positively correlated with depression scores in adolescence (r = 0.56, p < 0.01). One year post-ART, MCP-1 (CCL2) (r = -0.65, p=0.01), IL-2 (r = -0.62, p=0.02) negatively correlated with depression, while CD4 levels showed a positive correlation (r = 0.59, p=0.03). At two years, sCD163 (r = -0.57, p=0.04) and TGF-β2 (r = 0.63, p=0.02) correlated significantly, and RANTES/CCL5 (r = -0.93, p=0.02) at 4-5 years. iCAD levels correlated with YKL-40 at 16 years (r = 0.91, p=0.03). Brain metabolites pre-ART and neuroimaging at 16 years revealed Choline in the BG positively correlating with IL-17F, TNF-β, and MFGM choline. CD40L, FGF showed positive, while IL-18 and IL-8/CXCL8 showed negative relationships with immune markers. Myo-inositol exhibited similar complex associations across brain regions. Conclusions: Our findings underscore a dynamic interaction between immune biomarkers and brain metabolites over time, linking inflammation to depression in adolescents with perinatal HIV. Early hsCRP was notably predictive of depression risk at 16 years. Further investigations, including cerebrospinal fluid biomarker analysis and advanced multi-parameter predictive models, are warranted.

596

Cannabis Use Is Associated With Reduced Brain Integrity in Persons With HIV Sarah Cooley 1 , Patricia Reid 1 , June Roman 1 , Kalen J. Petersen 1 , Robert Paul 2 , Jodi Gilman 3 , Tricia Burdo 4 , Beau Ances 1 1 Washington University in St Louis, St Louis, MO, USA, 2 University of Missouri St Louis, St Louis, MO, USA, 3 Massachusetts General Hospital, Boston, MA, USA, 4 Rutgers Robert Wood Johnson Medical School, Piscataway, NJ, USA Background: Cannabis (CB) use is increasing nationwide due to legal changes and commercialization. People with HIV (PWH) are more likely to use CB versus people without HIV (PWoH). Data on the effects of CB use on brain structure and cognition in PWH are mixed, with few studies including measures of structural brain integrity. Methods: In a 2x2 design, participants were classified as CB user (CB+; PWH n=49;PWoH n=37) or non-user (CB-; PWH n=38;PWoH n=49) based on self-report (CB+=average of ≥3 days/week for past ≥12 months; CB-=no use ever, or <10 times total). Participants completed questionnaires assessing lifetime and recent (past 30 days) CB use, cognitive testing, a urine sample to measure metabolites of CB, and a magnetic resonance imaging scan to get regional brain volumes and cortical thickness (CT). Effects of HIV status and CB use, and their interaction, were assessed for cognition, brain volumes, and CT. Regression analyses assessed relationships between lifetime (LT) or recent use, urine metabolites, cognition, and imaging metrics in PWH compared to PWoH. Analyses were adjusted for age, sex, race, education, and other substance use. Results: PWoH/CB- were younger, more educated, and had less LT use of alcohol and tobacco compared to other groups. CB use amount, duration, and frequency did not differ between PWH/CB+ and PWoH/CB+. Frontal lobe volume was significantly smaller in both CB+ groups compared to CB- ( p =.01). HIV and CB significantly interacted in the frontal lobe (Figure 1), temporal lobe, cingulate, and global CT; CB use was associated with cortical thinning in PWH but not PWoH (all p <.01). Regression analyses indicated that greater LT CB use associated with worse delayed recall across all participants ( p =.02); smaller frontal, temporal, hippocampal and gray matter volumes in only PWH; and globally reduced CT in only PWH ( p <.05). Greater recent use associated with reduced volume and CT in multiple cortical regions, particularly in the frontal lobe, in only PWH ( p -values<.05). Higher urine levels of THC-COO-gluc associated with reduced frontal volumes for all participants ( p =.03). Conclusions: LT and recent CB use were detrimental to structural brain integrity, particularly in frontal and temporal regions, with greater changes observed in PWH compared to PWoH. While recent CB use did not impact cognitive performance, greater LT use was associated with worse recall regardless of HIV status. PWH should be cautioned about the potential long term effects of CB use on brain health.

Poster Abstracts

CROI 2025 160