CROI 2020 Abstract eBook

Abstract eBook

Poster Abstracts

Conclusion: HIV-1 infection drives irreversible overexpression of TIGIT alone or co-expressed with TIM3, and LAG3 or CD39 in a CD8+ T-cell subset-dependent manner. These results point towards the targeting of TIGIT in combination with TIM3, and LAG3 or CD39 to regain CD8+ T-cell subset specific function in HIV- infected individuals on cART. 303 DYNAMICS OF HIV-SPECIFIC T CELLS ON DURABLE ART DIFFER BY ANTIGEN RECOGNIZED & BY SEX Eva M. Stevenson 1 , Adam R. Ward 2 , Thomas R. Dilling 1 , John W. Mellors 3 , Rajesh T. Gandhi 4 , Deborah McMahon 3 , Joseph J. Eron 5 , Ronald Bosch 6 , Christina Lalama 6 , Joshua C. Cyktor 3 , R. Brad Jones 1 , for the A5321 Team 1 Weill Cornell Medicine, New York, NY, USA, 2 George Washington University, Washington, DC, USA, 3 University of Pittsburgh, Pittsburgh, PA, USA, 4 Massachusetts General Hospital, Boston, MA, USA, 5 University of North Carolina at Chapel Hill, Chapel Hill, NC, USA, 6 Harvard University, Cambridge, MA, USA Background: T-cell responses to HIV decay in the early stages of ART, with a half-life of 39 weeks. We previously demonstrated a direct correlation between levels of cell-associated HIV DNA (CA-DNA) and magnitudes of HIV-specific T-cell responses targeting early gene products Nef/Tat/Rev in the ACTG A5321 cohort. These results suggested that ongoing interactions with HIV-infected cells may shape HIV-specific T-cell responses in individuals on long-term ART; however, little is known about the dynamics of these responses. Methods: We previously performed IFN-g ELISPOT assays on PBMCs from 49 participants (11 female) at study entry (on-ART timepoint 1): median (IQR) yrs on ART 7 (4, 8). We measured responses to pools of overlapping 15-mer peptides spanning the HIV gene products Gag, Env, Pol, Nef/Tat/Rev, as well as CMVpp65. Here, we applied this same assay to batched samples fromweek 24 & week 168 post-entry. Relationships were assessed between these responses and virologic/ immunologic & clinical data provided by the ACTG. Results: HIV-specific T-cell responses were stable on durable ART, with magnitudes differing by gene product & by sex (Figure). Responses exhibited long median half-lives, which also differed by sex: Gag 32.4yrs (F 75.1yrs, M 3.7yrs); Env 3.5yrs (F 2.5yrs, M 1.1yrs); Pol 12.1yrs (F no decay, M 6.1yrs); Nef/Tat/ Rev 6.9yrs (F 5.3yrs, M 6.1yrs). F vs. M participants exhibited higher magnitudes of responses longitudinally for all HIV gene products, but not for CMV, before and after controlling for pre-ART HIV RNA & CD4 count (all p<0.05). Higher levels of CA-DNA at study entry were associated with lesser decay of Nef/Tat/Rev- specific responses between weeks 24 & 168 (r=0.36, p=0.03; r=0.34, p=0.06 controlling for pre-ART HIV RNA & CD4 count). Correlations were not observed between: i) CA-DNA & T-cell responses to other HIV gene products (p>0.1), nor ii) between the slopes of decay of any HIV-specific T-cell responses and CA-RNA, plasma HIV-RNA, %CD38+HLA-DR+ T-cells, age, or PD-1 expression. Conclusion: Overall, HIV-specific T-cell responses were stable, demonstrating long half-lives, which differed by sex. Females also displayed higher magnitudes of HIV-specific T-cell responses. This result may help explain previous findings that females have a lower residual viremia in this cohort. Higher CA-DNA at study entry correlated with slower rates of decay in Nef/Tat/Rev-specific T-cell responses on long-term ART, consistent with some level of ongoing recognition of infected cells.

304 EXPRESSION PROFILING OF HIV LATENTLY INFECTED CELLS VIA NANOSTRING AND MASS CYTOMETRY Hannah S. Sperber 1 , Tongcui Ma 2 , Nadia R. Roan 2 , Satish K. Pillai 1

Poster Abstracts

1 Vitalant Research Institute, San Francisco, CA, USA, 2 Gladstone Institute of Virology and Immunology, San Francisco, CA, USA Background: The main barrier to an HIV cure is the latent HIV reservoir. Long-lived HIV latently-infected cells remain invisible to the host immune system and persist during antiretroviral therapy. In this study, we characterized latently-infected cells by implementing combined transcriptomic and proteomic profiling to identify unique expression signatures and reliable biomarkers that can be exploited to target and eliminate the latent reservoir. Methods: Primary CD4+ T cells were purified from six healthy donors and were infected with a dual-reporter HIV construct that enables the isolation of HIV latently-infected and productively-infected cells by flow cytometry. The populations were then characterized using NanoString hybridization and fluorescence-based digital counting technology allowing for simultaneous detection of 770 mRNA and 30 protein targets, and mass cytometry (CyTOF), measuring 40 surface proteins. Target expression levels were compared between populations using false discovery rate (FDR<0.1), cellular pathways were analyzed using global significance scores, and upstream regulatory networks and causal relationships were deciphered using Ingenuity Pathway Analysis. Results: The latent population displayed significant upregulation of CD73 protein and IL8 mRNA, and significant downregulation of CD39 mRNA compared to productively-infected cells and controls. Protein expression levels of T cell activation markers including CD25, PD-1, OX40, CD127, and GITR did not significantly differ between productively- and latently-infected cells, while ICOS, an inducible T cell costimulator, was significantly increased on latently- infected cells. The ‘Pathogen defense’ pathway was significantly suppressed in both HIV infected cell populations compared to uninfected controls. ‘Antigen processing’ was strongly suppressed in the latent population. Transcription factors FOXP1, FOXD1, and FOXJ1 were discovered as the top three master regulators in latent cells. Conclusion: Our data suggest that HIV latently-infected cells exhibit distinct molecular features associated with an anergic and/or hypoxic T cell state, and may subvert antigen processing to remain immunologically invisible. FOXD1 and FOXJ1 likely repress HIV transcription in latently-infected cells through inhibition of NFkB and NFAT complexes. Our results warrant validation in vivo using clinical samples from ART-suppressed HIV-infected individuals, and mechanistic exploration ex vivo using targeted gene knockouts.

CROI 2020 104