CROI 2024 Abstract eBook

Abstract eBook

Poster Abstracts

588

Neurite Orientation Dispersion and Density Imaging in People With HIV on Stable Treatment Phillip Chan , Nakul Raval, Praveen Honhar, Jennifer Chiarella, Allison Nelson, David Matuskey, Richard Carson, Ansel Hillmer, Serena Spudich Yale University, New Haven, CT, USA Background: Neurite Orientation Dispersion and Density Imaging (NODDI) is a multi-compartmental diffusion MRI technique that characterizes brain microstructures by quantifying the packing density of axons or dendrites (neurite density index, NDI), the orientational coherence of neurites (orientation dispersion index, ODI) and the free water fraction (isotropic compartment, ISO). This preliminary study compares NODDI metrics cross-sectionally between virally suppressed people with HIV (PWH, plasma HIV RNA <20 cps/ml, median duration of treatment: 22 years) and people without HIV (PWOH), and longitudinally in the PWH group over a median duration of 32 months. Methods: Eight PWH (age: 56 [IQR 53,59], 8/8 male) and eight PWOH (age: 57 [IQR 53,62], 3/8 male) underwent NODDI scans in a 3T Siemens Prisma scanner, employing a two-shell diffusion protocol tailored for NODDI (32,64 direction; b=700,2000 s/mm 2 ; 80 slices; TR/TE=4100/88ms). After performing distortion and eddy-current correction and FreeSurfer segmentations, NDI, ODI and ISO were measured in the primary region of interest (ROI), the frontostriatothalamic (FST) region, while NDI was measured in the secondary ROIs, including the frontal, parietal, temporal, and occipital lobes. Statistical assessments included the Mann-Whitney U Test for cross-sectional evaluations and the Wilcoxon Signed Rank Test for longitudinal changes. Results: At the 1st scan, the CD4+ and CD8+ T-cell counts of PWH were 631 [IQR 412,831] and 576 [IQR 442,840] cells/mm 3 , with a CD4/CD8 ratio of 0.96 [IQR 0.66,1.36]. Compared to PWOH, PWH had lower NDI (mean difference ∆=-13.6±6.6%, p=0.001) and ODI (∆=-12.1±3.7%, p<0.001), but ISO showed no significant difference in the FST region. PWH also had lower NDI in all secondary ROIs when compared to PWOH (Table) (p<0.05). In the longitudinal analysis, PWH showed a decrease in NDI (∆=-11.2±9.2%, p=0.02) and ISO (∆=- 17.2±7.6%, p=0.007) but statistically unchanged ODI (∆=1.5±7.8%, p=0.547) in the FST region. Moreover, NDI decreased in all secondary ROIs (p<0.05). Conclusion: Compared to PWOH, PWH with stable HIV suppression had lower NDI and ODI in the FST and lower NDI in major cortical regions, indicating a possible loss in microstructural integrity. The longitudinal decline of NDI in PWH may indicate ongoing changes at the axonal/dendritic level. Future inclusion of cognitive testing, multi-modal neuroimaging, and longitudinal PWOH controls will help validate and understand the observed NODDI changes. Associations Between a Polygenic Risk Score for Alzheimer’s Disease and Brain Integrity in HIV Anna Curtis, Laura Ibanez, Carlos Cruchaga, Beau Ances, Sarah Cooley Washington University in St Louis, St Louis, MO, USA Background: People living with HIV (PWH) continue to live to older ages making it difficult to differentiate between age-related neurocognitive disorders, such as Alzheimer disease (AD), and HIV-related neurocognitive impairment. Polygenic risk scores (PRS) allow for computing genetic risk for various diseases. This study leveraged a PRS calculated using the latest genome wide association study for AD (AD-PRS) to evaluate if it was associated with the presence of HIV disease markers, HIV-related neurocognitive impairment, and brain integrity. Methods: 115 PWH (age≥50; male n=94, female n=21) completed magnetic resonance imaging, a blood draw, and a comprehensive neuropsychological battery comprised of 15 tests covering 5 cognitive domains (learning, recall, executive functioning, psychomotor speed, and language). Participants were genotyped with an Illumina array platform (GSA), stringent quality control and imputation was performed following standard pipelines. AD-PRS was computed using the binary logarithm transformation of the reported OR from the sentinel SNPs from the latest AD GWAS, including the APOE region,

with wide-spread increases in MD, and older PWH are disproportionally affected. By contrast, E4-associated FA reduction is more localized, and PWH E4 carriers with long disease duration are at a greater risk. These findings warrant further examination of more localized WM changes along individual tracts in the rapidly aging HIV+ population.

587

HIV Clinical Factors and Cardiovascular Disease Risk Predict White Matter Hyperintensity Burden Mikaley Bolden 1 , Peyton Thomas 1 , Richard C. Gallagher 1 , Princy Kumar 1 , David J. Moore 2 , Ronald J. Ellis 2 , Xiong Jiang 1 1 Georgetown University Medical Center, Washington, DC, USA, 2 University of California San Diego, La Jolla, CA, USA Background: Cardiovascular disease (CVD) risk factors such as hypertension are highly prevalent in people with HIV (PWH) and have been shown to affect brain function and structure, including an elevated risk of white matter hyperintensity (WMH), a measure of small vessel disease. We investigated the factors associated with WMH burden (measured as total WMH volume, or WMHv) in adults with chronic HIV disease, focusing on CVD risk and HIV-clinical factors. Methods: Participants were 134 adults (age range 40-70, 34 females, 82 African American, 97 PWH). CVD percent risk was calculated using the Framingham General Cardiovascular Risk Profile, including age, BMI, SBP status and treatment, smoking status, and diabetes status. High-resolution T1w structural MRI images (1x1x1mm 3 ) were acquired. The software package CAT12 (https://neuro-jena.github.io/cat/) was used to detect WMHs and obtain WMHv and gray matter volume (GMv). The effects of HIV status, age, sex, race, education, and CVD percent risk on WMH burden were examined using non-parametric ANCOVA. For PWH, potential impacts of CD4 nadir, current CD4/CD8 ratio, and disease duration on total WMHv were examined using non parametric regressions separately for each clinical factor. Parametric multiple linear regressions investigated the effects of WMHv, HIV status, age, sex, race, education, and HIV clinical factors on total GMv. Total WMHv and GMv were normalized by intracranial volume. Results: In the full study sample, higher CVD percent risk, older age, and African American race strongly predicted higher WMHv (at least p=.001), versus a marginal effect of education (p=.049) and HIV status (p=.057), while higher WMHv, male sex, and older age were associated with lower GMv (at least p=.001). In PWH, in addition to CVD percent risk, age, and race, lower CD4/CD8 ratio (p=.016) and longer HIV disease duration (p=.002) also strongly predicted higher WMHv. Conclusion: Consistent with previous findings, WMH is prevalent in middle aged to older PWH on cART and with successful viral suppression, and WMH burden strongly correlates with CVD risk and age. In addition, WMH burden correlates with current CD4/CD8 ratio and HIV disease duration, suggesting the need to improve immune function, control chronic neuroinflammation, and develop antiretroviral regimens that are less toxic to the CNS. Also, the significantly higher WMH burden in African Americans with and without HIV calls for actions to address health disparities.

Poster Abstracts

589

CROI 2024 162