CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

365 LONGITUDINAL CHARACTERIZATION OF HIV PROVIRUSES IN PEOPLE ON SUPPRESSIVE ART Annukka Antar 1 , Katharine M. Jenike 1 , Sunyoung Jang 1 , Danielle Rigau 1 , Daniel B. Reeves 2 , Rebecca Hoh 3 , Jeanne C. Keruly 1 , Richard D. Moore 1 , Joshua T. Schiffer 2 , Bareng Nonyane 4 , Frederick M. Hecht 3 , Steven G. Deeks 3 , Janet Siliciano 1 , Ya-Chi Ho 5 , Robert Siliciano 1 1 Johns Hopkins University School of Medicine, Baltimore, MD, USA, 2 Fred Hutchinson Cancer Research Center, Seattle, WA, USA, 3 University of California San Francisco, San Francisco, CA, USA, 4 Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA, 5 Yale University, New Haven, CT, USA Background: Clonal proliferation of CD4+ T cells harboring HIV proviruses is a major contributor to HIV persistence despite antiretroviral therapy (ART). Most proviruses are defective, and it is unknown whether cells harboring intact and defective proviruses or those with escape epitopes experience differential rates of clonal expansion and persistence during ART. Methods: To determine whether HIV-specific cytotoxic T lymphocyte (CTL) pressure or differential rates of clonal proliferation shape the HIV provirus landscape during long-term ART, we sequenced 661 near-full length proviruses from two samples, average 7 years apart, from 8 individuals with suppressed viral loads on ART. 3 of these were elite controllers on ART. The best-defined CTL epitopes based on HLA type were identified in HIV Gag, Pol, and Nef, and each epitope in each provirus was categorized as recognized or not based on its sequence and published data. Results: We found that the provirus landscape of intact and defective proviruses does not change dramatically over time on suppressive ART, although there was a trend towards fewer intact proviruses over time when analyzed by dual primer-probe ddPCR. There was no evidence for longitudinal selection effects in HIV epitopes in Gag, Pol, and Nef. Intact proviruses appear in large clones at least as often as defective proviruses, and proviruses found in large clones are not enriched in escape/ unrecognized epitopes. Elite controllers on ART have a similar distribution of defective proviruses and similar proportions of escape and unrecognized epitopes as other individuals on ART. The proportion of proviruses present in large clones increased over time on ART in all participants. Albeit with small sample sizes, modeling of this data suggests that over time on ART a smaller number of very large infected CD4 clones come to dominate the observed provirus landscape. Conclusion: We demonstrate that the mechanisms of clonal proliferation in vivo do not activate HIV expression often enough to discern dramatic differences in the types of proviruses found in large clones due to viral cytotoxicity or CTL recognition. Our work suggests that CTL targeting of activated CD4+ T cells expressing HIV genes during long-term suppressive ART shapes the provirus landscape only subtly if at all. Our findings in elite controllers on ART indicate that the drivers of HIV persistence are similar in this population despite stronger, polyfunctional HIV-specific CTLs. 366 IDENTICAL HIV PROVIRUSES ARISE FROM CELL EXPANSION AND INFECTION BY COMMON ANCESTOR Sean Patro 1 , Aurelie Niyongabo 1 , Shuang Guo 2 , Jason W. Rausch 1 , Michael J. Bale 1 , Andrew Musick 1 , Xiaolin Wu 2 , Liliana Perez-Rodriguez 3 , Wei Shao2, Eli A. Boritz 3 , Steven G. Deeks 4 , Frank Maldarelli 1 , Stephen H. Hughes 1 , John M. Coffin 5 , Mary F. Kearney 1 1 National Cancer Institute, Frederick, MD, USA, 2 Leidos Biomedical Research, Inc, Frederick, MD, USA, 3 NIAID, Bethesda, MD, USA, 4 University of California San Francisco, San Francisco, CA, USA, 5 Tufts University, Boston, MA, USA Background: Understanding the mechanisms of HIV-1 persistence during ART is crucial for developing curative strategies. Identical proviral sequences that are observed during ART can result from the proliferation of single infected cells or from genetic bottlenecks leading to viral clones that can spread prior to ART initiation. To investigate the origins of identical proviruses in individuals on ART, we sequenced near full-length proviruses and determined their sites of integration. Methods: PBMC was obtained from 5 donors on ART and analyzed by Multiple- Displacement Amplification (MDA) Single-Genome Sequencing in which DNA is diluted to a proviral endpoint, subjected to whole genome amplification, and used for both integration site analysis and full-length proviral sequencing. 15 sets of identical sub-genomic sequences (12 in P6-PR-RT, 3 in env) were examined to determine if i) identical proviruses also had identical integration sites (cell clones), ii) identical proviruses had different integration sites (viral

364 HIV DYNAMICS AND REPOPULATION OF RESERVOIRS IN THE HUMAN BODY Antoine Chaillon 1 , Sara Gianella 1 , Simon Dellicour 2 , Stephen A. Rawlings 1 , Michelli Faria De Oliveira 1 , Caroline Ignacio 3 , Magali Porrachia 4 , Bram Vrancken 2 , Davey M. Smith 1 1 University of California San Diego, San Diego, CA, USA, 2 Katholieke University Leuven, Leuven, Belgium, 3 University of California San Diego, La Jolla, CA, USA, 4 Veterans Medical Research Foundation, San Diego, CA, USA Background: Characterizing HIV persistence and dynamics across the human body is important to develop ways to clear reservoirs. This goal has been hampered by technical difficulties and obtaining fresh tissues. Methods: Samples were obtained from 6 Last Gift participants, who provided blood ante-mortem and their whole bodies for rapid autopsy within 6 hours of death. Two voluntarily stopped their antiretroviral therapy (ART) before death and 4 remained on ART. HIV reservoirs were characterized by digital droplet PCR and single genome amplification and sequencing of full-length (FL) envelope HIV. Phylogeographic methods reconstructed HIV spread and generalized linear models (GLM) tested for associations between HIV diversity, divergence, predicted tropism, DNA level, and viral dispersal. Results: Across participants, HIV DNA levels ranged from~0 copies/10 6 cells in the occipital lobe (interquartile range: [IQR] 0-4.6) to 659 cp/10 6 cells in lymph nodes (IQR 580-753), mean=98 cp/10 6 cells (IQR:23-132) across all tissues. We recovered 603 intact FL env sequences in antemortem blood cells and across 28 anatomical sites (mean of 7 viruses/site, IQR: 5-10), including the central nervous system (CNS). Among the 2 participants stopping ART, viral diversity was lower (p<0.01) in blood plasma than in most tissue-derived HIV DNA. Most of the rebounding intact HIV RNA populations in blood (65% and 80% of variants) were clonal (>99% identical). There was also evidence of clonal expansion within tissues (Fig.), especially gut and genital tract (e.g. 30 identical proviruses observed across 8 tissues in 1 participant). While the main sources of viral dispersal were blood, gut and lymph nodes, our models also revealed viral exchanges within the CNS (Fig.) and from blood toward the CNS. The GLM models revealed that low HIV genetic divergence between sites and high HIV diversity in the recipient sites but not HIV DNA levels were associated with viral exchange. Conclusion: This study found: 1) The emergence of large, clonal, intact HIV RNA populations after stopping ART, which repopulated tissues throughout the body; 2) Multiple sites can act as hubs for dissemination of HIV within the host; 3) Viral exchanges occur within the CNS areas and between the CNS and blood; 4) viral dynamics are associated with low HIV divergence between sites and high HIV divesity at the recipient site.

Poster Abstracts

CROI 2020 126