CROI 2020 Abstract eBook

Abstract eBook

Oral Abstracts

Results: Single cell transcriptomes were analyzed from 31,175 CSF cells and 35,694 PBMC. CSF cells were comprised of T cells (93%), B cells (0.5%), Monocytes (3%), Dendritic cells (1.9%), and NK cells (1.7%); CSF cell subset frequencies did not differ between PWH and HIV-. Differential expression analysis identified 64 and 128 genes that were differentially expressed between PWH and HIV-, in CSF and blood T cells, respectively, with 33 genes that were differentially expressed in both blood and CSF T cells based on HIV infection (log fold change >0.1; FDR< 0.01). We next trained two logistic LASSO PBMC-based and CSF-based models to differentiate between T-cells from a HIV- or a PWH and tested them in a leave-one-out cross validation (LOOCV) approach. Expression of ~200 genes differentiated a T cell from a PWH versus a HIV- at accuracy of >0.8. 62 and 54 genes were stably selected in all LOOCV iterations for the CSF and PBMC models, respectively. Out of the 62 genes selected in the CSF model, 41 were common to the PBMC model. Ingenuity pathway analysis revealed a significant association between HIV status and signaling downstream of the pro-inflammatory cytokine IL-15 in both blood and CSF. Conclusion: By using a multimodal analysis including machine-learning of single cell gene expression data in T cells, we identified potential regulators of immune dysfunction during ART-suppressed HIV infection, including IL-15 pathways.

0.9) for persons without HIV (p=0.003). The majority (80%) of enhancement was eccentric, which did not differ by HIV status. Over half (53%) of PWH with abnormal wall enhancement did not have associated luminal narrowing on TOF MRA. The greater mean number of enhancing arterial segments in PWH remained statistically significant after adjusting for demographics and CV risk factors. In a model adjusted for age, sex, race, and statin use, PWH had an average 4.58 times as many enhancing arterial segments as persons without HIV (95% CI 1.51-13.83, p=0.007). Conclusion: PWH had a greater burden of primarily eccentric arterial wall enhancement compared with persons without HIV. Furthermore, luminal imaging with TOF MRA underestimated the burden of arterial disease in more than half of PWH. Future studies should investigate the association of arterial wall enhancement with indices of immune activation and radiologic markers of cerebrovascular disease. 29 PET IMAGING OF SYNAPTIC DENSITY IN HIV: PRELIMINARY FINDINGS FROM A PILOT STUDY Julian Weiss 1 , Rachela Calvi 1 , Mika Naganawa 1 , Takuya Toyonaga 1 , Shelli Farhadian 1 , Michelle Chintanaphol 1 , Jennifer Chiarella 1 , Ming-Qiang Zheng 1 , Jim Ropchan 1 , Yiyun Huang 1 , Richard Carson 1 , Serena S.Spudich 1 1 Yale University, New Haven, CT, USA Background: Synaptic injury, which is potentially reversible, is a pathological hallmark of HIV associated neurocognitive disorder (HAND) in people living with HIV (PLWH) on antiretroviral therapy (ART), but it has only been assessed in post-mortem studies in humans. Here we report initial results from a pilot positron emission tomography (PET) study employing the novel ligand 11C-UCB-J for synaptic vesicle protein 2A (SV2A) to measure synaptic density in virologically suppressed PLWH and healthy controls (HC). Methods: Six male PLWH and seven age-matched HC underwent 3T magnetic resonance imaging (MRI) and high-resolution PET scanning with 11C-UCB-J combined with arterial blood sampling. Distribution volume (VT, mL/cm 3 ) and binding potential (BPND), a measure of SV2A binding, were assessed in 28 regions of interest (ROIs) using the centrum semiovale as a reference region. Partial volume correction using Freesurfer was performed to correct for atrophy. Differences in VT and BPND between the groups were analyzed using a Student’s t-test. Results: There were no significant differences in age (HC: mean [SD], 59 [8]; PLWH: 61 [5]; p=0.53), race, or body mass index between the groups. PLWH were well-suppressed on ART (mean [SD], CD4 T cells 703 [194] cells/mL, duration of ART 21 [8] years), and all participants had CSF and plasma HIV RNA <20 copies/mL. VT values of the reference region were similar in both groups (HC: 4.08 [0.70]; PLWH: 4.37 [1.01]; p=0.57). PLWH had significantly lower SV2A specific binding (BPND) in eight cortical ROIs compared with HC (p<0.05), illustrated by representative cases (Figure 1). Four of these ROIs, including the precuneus (HC: 6.13 [0.87]; PLWH: 4.89 [0.85]; p=0.03) and superior parietal lobule (HC: 6.04 [0.65]; PLWH: 4.71 [1.07]; p=0.03), are within the parietal lobe, which as a whole trended toward significance (HC: 6.19 [0.88]; PLWH: 5.07 [1.04]; p=0.07). There were no significant differences in VT, though lower values in PLWH were noted in the eight cortical ROIs with significantly decreased BPND. Conclusion: This preliminary analysis of an ongoing pilot study demonstrates the potential utility of SV2A PET imaging in identifying regions of reduced synaptic density in suppressed PLWH. A larger sample is needed to draw conclusions on which ROIs are most affected, and to explore associations between synaptic density and lab and clinical parameters including neuropsychological performance. SV2A imaging may be a promising outcome measure for interventional trials of HAND.

Oral Abstracts

28 GREATER BURDEN OF INTRACRANIAL ARTERIAL-WALL ENHANCEMENT IN PERSONS LIVING WITH HIV Felicia C. Chow 1 , Andrew Callen 1 , Victor Arechiga 1 , David Saloner 1 , Jared Narvid 1 , Priscilla Hsue 1 1 University of California San Francisco, San Francisco, CA, USA Background: Persons living with HIV (PWH) are at higher risk of stroke compared with age-matched persons without HIV. However, the mechanisms underlying increased cerebrovascular risk in PWH are unclear. In particular, the contribution of intracranial arterial disease to HIV-associated stroke remains poorly defined. We compared intracranial vessel wall magnetic resonance imaging (VW-MRI), which can demonstrate atherosclerotic disease even when conventional angiography is normal, in treated, virologically suppressed PWH and persons without HIV. Methods: All participants were >40 years of age with a history of cardiovascular (CV) disease or at least one CV risk factor. PWH were on antiretroviral therapy with undetectable plasma viral load. Demographics- matched persons without HIV were friends and family of PWH or recruited through flyers. Participants underwent a time-of-flight (TOF) MR angiogram (MRA) and 3D high resolution variable flip angle black blood post-contrast VW-MRI (CUBE) on a GE 3T Discovery scanner. The primary outcome was the number of visualized arterial segments with abnormal wall enhancement. Poisson models were used to compare the mean number of enhancing segments by HIV status. Results: Of 31 participants (mean age 58 years, 97%men), 19 were PWH (median CD4 count 492 cells/mm 3 ). There were no significant differences in age, sex, race, or CV risk factors between PWH and persons without HIV. A greater proportion of PWH were on a statin (84% versus 42%, p=0.021). The mean number of enhancing arterial segments for PWH was 1.8 (SD 1.3) versus 0.4 (SD

10

CROI 2020