CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

Follow up plasma and CSF VL were available for 18 patients: 16(88.9%) had undetectable plasma VL. Of these,13(81.2%) had undetectable CSF VL while 3 had CSF VL between 20-100 copies/ml. In 2 subjects,AZT intensification failed with non-suppression of both plasma and CSF VL and triple class DRMwere observed in CSF. After PI/INSTI intensification,complete resolution of symptoms occurred in 18(85.7%) patients. Follow up plasma and CSF VL were available for 20 patients: all had undetectable plasma VL. 16 patients(80%) had undetectable CSF VL while 4 had CSF VL between 20-100 copies/ml. PI/INSTI intensification successfully suppressed plasma and CSF VL in all subjects. Conclusion: Despite potential selection bias(lack of GRT in Group 1) this is a unique cohort of patients with sCVE with homogeneous treatment interventions. AZT intensification was effective in improving symptoms and reducing plasma and CSF VL in majority of subjects. Additional studies including GRT,pharmacokinetics and adherence measurements are needed to select the most appropriate treatment for sCVE in LRCs.

likelihood phylogeny (figure) suggests that HIV compartmentalization patterns differ between donors, with four donors showing evidence of HIV DNA compartmentalization (p<0.05). Overall, 143 FL HIV-env sequences were genetically intact, while 37 sequences were non-functional, with major deletions, frameshift and stop codon mutations (figure). For one donor, we found 23 clonal sequences with a frameshift mutation that was present in both brain and spleen, suggesting migration of cells with clonal provirus between tissue compartments. Conclusion: HIV DNA was detected in brain and lymphoid tissues despite long- term ART. Most HIV DNA populations in brain and lymphoid tissues appeared to have intact env genes and were often compartmentalized. Characterizing the composition of the HIV reservoirs in anatomic compartments is crucial for future HIV cure strategies.

Poster Abstracts

452 PRESENCE OF INTACT HIV DNA VARIANTS IN THE BRAIN AND LYMPHOID TISSUES DURING ART Michelli Faria de Oliveira , Benjamin Murrell, Andrej Vitomirov, Thomas Vollbrecht, Venkatesh Kumar, Nikesh Kumar, Ben Gouaux, David J. Moore, Alan F. Jarmusch, Fernando Vargas, Pieter Dorrestein, Ronald J. Ellis, Davey M. Smith, Sara Gianella University of California San Diego, San Diego, CA, USA Background: Although antiretroviral therapy (ART) reduces HIV RNA below the detection limit in blood plasma, HIV reservoirs persist in cellular compartments. Here, we characterize the size and composition of the HIV DNA reservoirs in brain and lymphoid tissues. Methods: We evaluated post-mortem brain and peripheral lymphoid tissues from 12 persons living with HIV (PLWH) obtained from the National NeuroAIDS Tissue Consortium. All donors died between 2001-2014, with virologic suppression on ART (<50 or 400 copies/ml, assay-dependent), and without evidence of central nervous system opportunistic disease. Presence of ART in post-mortem brain was confirmed by mass spectrometry. Total DNA was extracted from each tissue sample and levels of HIV DNA (gag) were quantified by droplet-digital PCR. The genotypic composition of the HIV DNA populations was evaluated by high-throughput single genome amplification using the PacBio platform to sequence Full-length HIV envelope (FL HIV-env). Results: We evaluated post-mortem tissues from 9 men and 3 women with a median age of 52 years (range: 40-66). Donors were on ART at the last visit, which occurred a median of 3 months prior to death (range: 1-4). Presence of Tenofovir or Lamivudine was confirmed in 6 out of 8 donors’ brain tissues by mass spectrometry. All donors had detectable HIV DNA in brain (frontal [FC] or occipital [OCC] cortex) and lymphoid tissues (lymph node [LN] or spleen [SP]). A total of 180 individual FL HIV-env sequences were obtained across donors (FC, n= 7 donors; OCC, n=3; LN, n=3; SP, n=8). For 10 donors, FL HIV-env sequences were obtained from paired brain and lymphoid tissues. Maximum

453 ASSOCIATION OF CEREBRAL SMALL VESSEL DISEASE WITH THE BRAIN IN HIV INDIVIDUALS Ryan Sanford 1 , Jeremy Strain 2 , Mahsa Dadar 1 , Josefina Maranzano 1 , Nancy Mayo 3 , Susan Scott 3 , Lesley K. Fellows 1 , Beau Ances 2 , Louis Collins 1 1 Montreal Neurological Institute, Montreal, QC, Canada, 2 Washington University in St Louis, St Louis, MO, USA, 3 McGill University, Montreal, QC, Canada Background: Emerging evidence has suggested that people living with HIV (PLWH) have increased risk of developing cerebral small vessel disease (CSVD). This may account for some of the cognitive impairment that continues to be common even in those with good viral suppression. In this study, we investigated whether PLWH had worse CSVD compared to demographically similar controls (CTL), and provide evidence of the impact CSVD has on brain volumetrics and cognition. Methods: Virologically suppressed PLWH on combination antiretroviral therapy (cART) and CTL participants underwent MRI and comprehensive neuropsychological testing. The total volume of white matter hyperintensities (WMH) on MRI was used as a surrogate marker for CSVD severity. Tensor-based morphometry and cortical modeling estimated regional brain volumes and cortical thickness, respectively. Rasch measurement theory was applied to the cognitive test scores, yielding an estimate of overall cognitive ability. Linear models were used to compare the WMH load, brain volumes and cognition between the two groups. These models controlled for age and sex. In addition, separate linear models assessed the association of brain volumes, cognition and factors commonly linked with vascular disease, including hypertension (defined as systolic blood pressure ≥140mmHg or diastolic ≥90mmHg), smoking, body mass index and waist circumference, with the WMH load. These models included all participants and controlled for HIV serostatus, age and sex.

CROI 2019 165