CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

variants carrying these mutations varied within different brain areas of a same patient. Conclusion: This work showed by UDS significant inter-regional and intra- regional viral diversity in CNS reflecting viral replication. It also confirmed HIV-compartmentalization in different brain areas suggesting that there is not a single but several reservoirs within CNS.

1 Africa Health Research Institute, Mtubatuba, South Africa, 2 University of KwaZulu- Natal, Durban, South Africa, 3 Cambridge University, Cambridge, UK Background: One line of evidence that the CNS may be a source for HIV in the face of ART is cerebrospinal fluid (CSF) escape, where HIV is suppressed in the blood but detectable in the CSF. However, it is unclear if CSF HIV is CNS restricted or can transmit to other sites. For the latter, HIV requires high-level drug resistance and infection of a cell able to traffic out of the CNS. Here we investigated the cell of origin of HIV in the CSF of CSF escape study participants, as well as whether the ART concentrations found in the CSF could result in the evolution high level resistance necessary for HIV replication outside the CNS compartment. Methods: We collected blood and CSF from 122 South African participants clinically indicated for lumbar puncture. We performed a viral load assay and detected concentrations of antiretroviral drugs in the blood and CSF and chose participants on the first line regimen of efavirenz, emtricitabine, and tenofovir for further study to avoid confounding effects of regimen type. For CSF escape participants (22% of total), we used the CD26 and CD36 host cell surface markers on the virion envelope to determine the cellular source of HIV using binding to anti-CD26 and CD36 antibody columns, followed by viral load assay of bound virus. The cell type specific signature of CD26 and CD36 was determined from in vitro infected macrophages and T cells and was unambiguous for these cell types. We also examined the effect of measured ART levels of CSF escape participants on HIV replication and evolution using in vitro infection. Results: We observed that the CD26/CD36 signature on the viral surface of HIV from CSF escape was consistent with T cell origin of the CSF virus. This was also the case for CSF HIV from participants who were viremic in both compartments. ART levels of efavirenz, emtricitabine, and tenofovir were not significantly different between individuals with CSF escape and those who were fully suppressed. Furthermore, HIV replication at CSF ART levels was required in in vitro infection for the progression to high level, multidrug resistance and replication at ART levels found in the blood. Conclusion: The combination of an infected cell type able to disseminate infection and ART levels conducive to stepwise evolution of resistance implicates the CNS as a source for the spread of drug resistant virus in the face of ART. 425 HIV-1 VIRAL DIVERSITY AND RESISTANCE IN CENTRAL NERVOUS SYSTEM BY DEEP SEQUENCING Eleni Giatsou 1 , Basma Abdi 1 , Isabelle Plu 1 , Nathalie Desiré 1 , Romain Palich 1 , Danielle Seilhean 1 , Vincent Calvez 1 , Anne-Geneviève Marcelin 1 , Aude Jary 1 1 AP–HP, Hôpitaux Universitaires Pitié Salpêtrière, Paris, France Background: The central nervous system (CNS) compartment is one of several sites in which compartmentalized HIV-1 replication has been observed. Most studies assessed viral compartmentalization in the CNS via cerebrospinal fluid, however, information about tissue compartmentalization of HIV-1 is still limited. Methods: We used ultra-deep sequencing (UDS) to study viral diversity and resistance patterns in different brain areas by analyzing reverse transcriptase (RT) gene. Twelve samples from 3 patients (P1, P2 and P3) with possible or certain HIV-encephalopathy were studied and sequencing was performed on MiSeq (Illumina®). HIV proviral DNA reservoir quantification was performed with Generic HIV DNA Cell® kit and diversity by (i) phylogenic analysis with approximately-maximum-likehood phylogenic trees with Fastree 2.1, (ii) single- nucleotide polymorphism on Geneious Prime software and (iii) HIV-1 genotypic drug resistance identification with algorithms 2018 administered by ANRS. Results: HIV-proviral DNA was undetectable in all P1 samples and in P2 cerebellum and thalamus sample. P2 temporal lobe and medulla oblongata sample as well as all P3 specimens showed detectable proviral-DNA loads by increasing order: cerebellum (23 cp/ 10 6 cells), medulla oblongata (31 cp/ 10 6 cells), temporal lobe (91 cp/ 10 6 cells), substantia nigra (92 cp/ 10 6 cells), caudate nucleus (130 cp/ 10 6 cells), and frontal lobe (544 cp/ 10 6 cells). Overall, RT phylogenic analysis revealed (i) a high diversity in each site analyzed, and (ii) HIV compartmentalization within different brain areas with a majority of them harboring a distinct HIV subpopulation. However, some cerebral sites also shared HIV variants; caudate nucleus and spinal cord in P1 or caudate nucleus, cerebellum and frontal lobe in P3 [figure 1]. On the other hand, brainstem (substantia nigra and oblongata specimen) area harbored a specific subpopulation in both P2 and P3. Some non-synonymous conferring resistance to Nucleoside and Non-Nucleoside Reverse Transcriptase Inhibitors were found, specifically M41L, V90L and V106I. However, the presence or proportion of

Poster Abstracts

' 426 HIV DIVERSITY IN CSF AND PLASMA OF INDIVIDUALS WITH HIV AND CRYPTOCOCCAL MENINGITIS Nametso Kelentse 1 , Sikhulile Moyo 2 , Mompati L. Mogwele 2 , Kwana Lechiile 2 , Kaelo Seatla 2 , Natasha O. Moraka 3 , Kesaobaka Molebatsi 1 , Tshepo B. Leeme 4 , David Lawrence 5 , Rosemary Musonda 2 , Ishmael Kasvosve 1 , Thomas S. Harrison 6 , Joseph N. Jarvis 5 , Simani Gaseitsiwe 2 1 University of Botswana, Gaborone, Botswana, 2 Botswana Harvard AIDS Institute Partnership, Gabarone, Botswana, 3 Stellenbosch University, Tygerberg, South Africa, 4 Botswana–UPenn Partnership, Gaborone, Botswana, 5 London School of Hygiene & Tropical Medicine, London, UK, 6 St. George's University of London, London, UK Background: HIV-1 can compartmentalize in reservoir sites e.g. the central nervous system (CNS) and this is a barrier to complete HIV eradication. We compared cerebrospinal fluid (CSF) and plasma viral load (VL), drug resistance mutations (DRMs) and co-receptor usage in HIV-1 strains from individuals co- infected with HIV-1C and Cryptococcal meningitis (CM) in Botswana Methods: This was a cross-sectional study utilizing CSF and plasma paired samples from 60 participants enrolled in a clinical trial evaluating the early fungicidal activity of 3 short-course, high-dose liposomal amphotericin B regimens for CM between 2014-2016. HIV VL was measured in 38/60 (63%) paired samples. Viral escape was defined as HIV-1 RNA ≥0.5 log 10 in CSF than plasma and HIV-1 VL discordance as CSF/Plasma ratio >1. HIV-1 protease, reverse transcriptase and envelope were sequenced using big dye sequencing chemistry and analyzed for DRMs using Stanford HIV drug resistance database. Geno2pheno was used for prediction of co-receptor usage. Results: A total of 34/38 participants (89.5%) had detectable VL in plasma and CSF with medians of 5.1 (IQR:4.7-5.7) and 4.6 (IQR:3.7-4.9) log 10 copies/ ml, respectively (p≤0.001). The prevalence of CSF viral escape was 1/34 (2.9%) [95% CI: 0.07-15.3]. HIV-1 VL discordance was observed in 7/34 (21%) pairs. Discordance was not associated with CD4 count, ART status, duration or regimen, abnormal mental status, or mortality. A total of 26/45 (58%) pairs were sequenced and 14%were on ART. Frequency of DRMs in the plasma and CSF was 9 and 11, respectively. The most predominant DRM in the plasma was K101E (n=2) whilst the other mutations occurred at equal frequency of 1 in

CROI 2020 148