CROI 2019 Abstract eBook

Abstract eBook

Poster Abstracts

398 THE IMPACT OF VORINOSTAT AND THERAPEUTIC VACCINE ON GUT HIV DNA: THE RIVER GUT STUDY John P. Thornhill 1 , Carolina Herrera 1 , Jonathan Hoare 1 , Simon Peake 1 , Helen Brown 2 , Nneka Nwokolo 3 , Julie Fox 4 , Tomas Hanke 2 , John Frater 2 , Sarah Fidler 1 , for the the RIVER trial investigators 1 Imperial College London, London, UK, 2 University of Oxford, Oxford, UK, 3 Chelsea and Westminster Hospital, London, UK, 4 Guy’s and St Thomas’ NHS Foundation Trust, London, UK Background: The RIVER randomized trial examined the impact of a T-cell prime-boost vaccination with vorinostat plus ART (ART+V+V) compared with ART alone in treated primary HIV infection (PHI) on blood total HIV DNA. Tissue reservoirs such as the gut-associated lymphoid tissue (GALT) are important sites of HIV persistence and may be differentially affected by the intervention. This RIVER sub study compares HIV DNA, markers of immune activation & exhaustion in GALT, and microbial translocation by study arm. Methods: ART was commenced within 4 weeks of confirmed PHI diagnosis at enrolment. At week 24, when plasma HIV-RNA was suppressed, participants were randomized (1:1) to receive either ART or ART plus a prime-boost T-cell vaccination (ChAdV63.HIVconsv and MVA.HIVconsv) followed by 10 doses of 400mg of vorinostat (ART+V+V). Following completion of the RIVER study protocol individuals (from each arm) consented to the gut sub study; individuals underwent colonoscopy, with terminal ileum and rectal biopsies taken for HIV DNA quantification (qPCR) and assessment of immune activation and exhaustion (PD-1 and HLA-DR/CD38 expression on CD4+ cells) by flow cytometry. Plasma microbial translocation markers (sCD163 & sCD14) were measured in plasma using Luminex. P24 antigen was measured in stimulated tissue explant supernatants by ELISA. Results: Eleven men were enrolled in the RIVER gut study, five in the ART-only arm and six in the ART+V+V arm, all were male. The median total HIV DNA in the terminal ileumwas 2.8 (ART+V+V) and 3.1 (ART) log 10 copies per 10 6 gut cells (P=0.25), and in the rectum 2.8 (ART+V+V) and 3.0 (ART) log 10 per 10 6 gut cells. (P=0.14). No significant differences in expression of PD-1 and HLA-DR/ CD38 co-expression on CD4+ T-cells from GALT were observed between study arm, median p24 levels measured from explant supernatants (n=7) were similar in each arm. Significantly higher sCD163 (P=0.03) but not sCD14 (P=0.55) levels were observed in plasma from participants in the ART+V+V arm compared with ART only. Conclusion: These data suggest that vorinostat in combination with a T-cell prime-boost-vaccine did not impact the GALT HIV reservoir, nor measures of immune exhaustion & activation on GALT CD4 T-cells in ART+V+V treated individuals compared with ART alone during PHI. Measures of bacterial translocation appear to be increased in ART+V+V over ART-only warranting further investigation.

targeted killing of PD-1+ CD4 T cells. Isolated CD4 T cells from eight untreated and five ART treated HIV-1 donors were incubated with either unconjugated αPD-1 Ab or ADC in a 5-day assay. Cells undergoing apoptosis and cell death were assessed by flow cytometry and culture supernatants were analyzed for viral p24 by ECL COBAS HIV Ag assay. Cells from ART treated donors were tested in a quantitative viral outgrowth assay (qVOA) to evaluate the frequency of cells harboring replication competent and infectious viruses following Ab or ADC treatment. Results: In HIV-1 viremic donors, α-PD-1 ADC compared to α-PD-1 Ab efficiently depleted PD-1+ CD4 T cells as indicated by the increase frequency of apoptotic and dead cells in cultures treated with α-PD-1 ADC (4.4- and 5.8-fold increase in apoptotic and dead cells, P<0.0001). Of note, levels of viral p24 were strongly suppressed by 79% (P<0.0001) in culture supernatants of α-PD-1 ADC treated cultures compared to α-PD-1 Ab. In ART treated aviremic subjects, α-PD-1 ADC treatment also efficiently depleted PD-1+ CD4 T cells compared to α-PD-1 Ab (2.4/2.5 fold increase in apoptotic and dead cells, P≤0.0025). Importantly, HIV-1 p24 was undetectable in all culture supernatants from the five ART treated patients studied. Cumulative qVOA data from ART treated donors showed that α-PD-1 ADC treated CD4 T cells had a 92% reduction in the levels of HIV RNA (P<0.0001; n=4) and undetectable levels of infectious virus (p24 levels, P<0.0001; n=4) compared to α-PD-1 Ab treated cells. Conclusion: These results indicate that depletion of PD-1+ CD4 T cells by α-PD-1 ADC represents a novel potential intervention towards functional HIV cure. 397 CHARACTERIZATION OF HIV-1 TRANSCRIPTION PROFILE AFTER ROMIDEPSIN THERAPY IN VIVO Sara Moron-Lopez 1 , Peggy Kim 2 , Ole S. Søgaard 3 , Martin Tolstrup 3 , Joseph K. Wong 2 , Steven A. Yukl 2 1 University of California San Francisco, San Francisco, CA, USA, 2 San Francisco VA Medical Center, San Francisco, CA, USA, 3 Aarhus University Hospital, Aarhus, Denmark Background: Reversing HIV-1 latency has been suggested as a strategy to eradicate HIV-1. We investigated the effect of romidepsin on the HIV transcription profile in participants from the REDUC part B clinical trial. Methods: Seventeen participants on suppressive antiretroviral therapy were vaccinated with six doses of the therapeutic vaccine Vacc-4x followed by treatment with three doses of romidepsin. Samples from nine study participants were available for HIV transcription profile analysis. Read-through, total (TAR), elongated (longLTR), polyadenylated (polyA) and multiply-spliced (TatRev) HIV transcripts and total HIV DNA were quantified at baseline (visit1) and 4 hours after the second (visit 10b) and third (visit 11b) romidepsin infusions, using droplet digital PCR. Results: We observed a significant increase in read-through (1.7-fold, p=0.02), total (1.9-fold, p<0.01), elongated (2.4-fold, p<0.01) and polyadenylated (1.9- fold, p=0.03) HIV RNA/10 6 PBMCs after the second romidepsin infusion (visit 10b), and a 1.9-fold increase in elongated transcripts after the third romidepsin infusion (visit 11b) (p<0.01). No significant changes were observed in multiply- spliced HIV RNA or HIV DNA. No change was observed in the ratio of read- through/total HIV transcripts. The ratio of elongated/total HIV RNA increased after both the second and third romidepsin infusions (p=0.02), while the ratio of polyadenylated/elongated HIV decreased after the third infusion (p=0.02). A strong negative correlation was observed between HIV DNA and the time to rebound (VL>50copies/ml) at visit 1, 10b, and 11b (Rho=-0.81, -0.88, and -0.91; p=0.02, p<0.01, and p<0.01, respectively). Levels of all HIV RNAs tended to correlate negatively with the time to rebound. This association was strongest for the comparison between elongated transcripts and time to VL>1,000copies/ml after romidepsin administration (Rho=-0.78, p=0.03 at visit 10b; Rho=-0.77, p=0.03 at visit 11b). Conclusion: In these patients, romidepsin increased early events in HIV transcription (initiation and especially elongation), but had less effect on later stages (completion, multiple splicing) that may be required for comprehensive latency reversal and cell killing. Without cell death, increased HIV transcription before or after latency reversal may hasten viral rebound after therapy interruption.

Poster Abstracts

CROI 2019 143