CROI 2025 Abstract eBook

Abstract eBook

Poster Abstracts

HIV restriction factor APOBEC3G (A3G) in newly infected cells. Here we show that inhibiting neddylation via MLN reduces HIV reactivation from latency after treatment with latency reversing agents (LRAs). Methods: Latently infected cell lines (ACH2, J-Lat 6.3) were treated with MLN prior to, and during, exposure to three mechanistically distinct LRAs: TNFa, PMA/ionomycin, and JQ1. Following LRA treatment +/- MLN, we quantified extracellular p24 via ELISA and assessed intracellular p24 and apoptosis levels via flow cytometry. To assess provirus transcription, HIV gag, TAR, and long LTR transcripts were quantified by RT-qPCR. Infectivity of virions reactivated from replication competent, Vif+, latent HIV was assessed via infection of TZM-bl cells and immunoblots revealed A3G virion incorporation levels. Lastly, tat/rev multiply-spliced RNA (msRNA) was measured in CD4 T cells from ART suppressed people living with HIV (PLWH) stimulated ex vivo to assess MLN impact on reactivation in a primary cell model of HIV latency. Results: In ACH2 cells treated with TNFa, 200nM MLN reduced: gag RNA 4.79-fold (p=0.0029, n=3), the percentage of cells expressing intracellular p24 by 44.3% (p=0.0012, n=4), and extracellular p24 production by 2.36-fold (p=0.0102, n=5) relative to DMSO. PMA/ionomycin and JQ1 stimulation of ACH2 yielded parallel results, as did TNFa treatment of J-Lat 6.3. MLN reduced TAR and long LTR transcripts in ACH2 and J-Lat 6.3, indicating decreased initiation of proviral transcription. ACH2 virion infectivity was reduced by 93.3% compared to DMSO (n=3), and Vif-positive virion A3G increased. MLN treatment also induced apoptosis in ACH2 and uninfected CD4 T cells (Fig. A). In CD4 T cells from three ART-suppressed PLWH, 200nM MLN following PMA/ionomycin stimulation reduced tat/rev msRNA by 35%, 67%, and 43% respectively (n=3) (Fig. B). Conclusions: Neddylation inhibition limits provirus reactivation by reducing HIV transcription, antigen and virus production and infectivity after LRAs. Our work highlights neddylation as a targetable mechanism to modulate proviral reactivation by LRAs, and MLN as a method to diminish reservoir replenishment when combined with other HIV cure strategies during or after ART. The figure, table, or graphic for this abstract has been removed. Venetoclax Decreases Intact Proviral DNA Frequency in SIV-Infected, ART-Suppressed Rhesus Macaques Sydney N. Bergstresser 1 , Tomas Raul Wiche Salinas 2 , Diane Carnathan 2 , Hong Wang 1 , Yohannes M. Abraham 1 , George Pavlakis 3 , Barbara Felber 3 , Mario Roederer 4 , Christine Fennessey 5 , Brandon Keele 5 , Mirko Paiardini 2 , Deanna Kulpa 2 , Guido Silvestri 6 1 Yerkes National Primate Research Center, Atlanta, GA, USA, 2 Emory University, Atlanta, GA, USA, 3 National Cancer Institute at Frederick, Frederick, MD, USA, 4 Vaccine Research Center, Bethesda, MD, USA, 5 Frederick National Laboratory for Cancer Research, Frederick, MD, USA, 6 Emory National Primate Research Center, Atlanta, GA, USA Background: Understanding the mechanisms responsible for the persistence of the virus reservoir under ART is essential to develop novel HIV cure approaches. Upregulation of the anti-apoptotic protein B cell lymphoma-2 (BCL-2) has been observed in HIV-Gag+CD4+ T cells from ART treated people living with HIV, along with an enrichment for intact proviral DNA in cells with high BCL-2 expression. These data suggest high levels of BCL-2 may help HIV Gag+CD4+ T cells with intact provirus to evade apoptosis and enter long-term survival. Here we used BCL-2 inhibition in ART-suppressed SIV-infected Rhesus macaques (RM) to promote reservoir clearance in vivo . Methods: 29 adult RMs were infected with SIVmac239M for 8 weeks followed by daily ART for >90 weeks. To induce SIV expression under ART, we employed the previously established model of CD8a+ T cell depletion combined with IL-15 treatment. 23 RMs were given the CD8a depleting antibody MT807R1 followed by 7 doses of heterologous IL-15 across 21 days. 17 CD8a depleted RMs were administrated 10 doses of Venetoclax (300 mg) in the same period. 9 Venetoclax treated RM were given 4 SIV env protein monoclonal antibodies (RhmAbs) one day prior to CD8a depletion. All RM were followed until elective necropsy at 84 days post CD8a depletion. Results: On day 7, following the first full cycle of Venetoclax, RM were observed to have elevated frequency (p<0.0001) of Caspase-3+CD4+ cells. We observed a significant decrease in the frequency of intact proviral DNA from this same time point (p=0.0267), followed by a further reduction 35 days after the last intervention (day 56; p<0.0001). A decline in total CD4+ T cells was observed in all MT807R1 treated RMs by day 7 post depletion concurrent with a rise in plasma viremia, however the CD4+ T cell compartment fully reconstituted in venetoclax treated RMs and even expanded by day 56. Adjusting for changes in

CD4+ T cell count in whole blood showed the decreased copy number of intact proviral DNA in venetoclax treated RMs was maintained (p=0.0129). Conclusions: In summary, BCL-2 antagonism decreased the frequency of peripheral CD4+ T cells with intact provirus in ART-treated SIV-infected RM after induction of virus expression. The reservoir reduction was maintained after CD4+ T cell re-population at day 56 post CD8 depletion, suggesting preferential loss of SIV-infected cells. These data support further exploration of BCL-2 antagonism as a tool to reduce reactivated virus reservoir under ART. AZD5582 Inhibits Vaccine-Elicited CD8+ T-Cell Responses in SIV-Infected Rhesus Macaques on ART Benjamin D. Varco-Merth 1 , Alejandra Marenco 1 , Omo Fadeyi 1 , Paula Armitage 1 , Robert W. Omange 2 , William Goodwin 1 , Derick Duell 1 , Manuel Medina 1 , Michael Axthelm 1 , Brandon Keele 3 , Jeffrey D. Lifson 3 , Janina Gergen 4 , Susanne Rauch 4 , Louis Picker 1 , Afam Okoye 1 1 Oregon Health and Science University, Portland, OR, USA, 2 Gilead Sciences, Inc, Foster City, CA, USA, 3 Frederick National Laboratory for Cancer Research, Frederick, MD, USA, 4 CureVac, Tübingen, Germany Background: At CROI 2024 (abstract 180), we reported that mRNA/SIV-Gag vaccination during ART and with boost prior to ART interruption (ATI) delayed time to rebound and induced early virologic control in SIV-infected rhesus macaques (RM). In this study, we explored whether the latency-reversing agent AZD5582 can be used to enhance efficacy of the mRNA/SIV-Gag vaccine. Methods: All RM were intravenously (IV) inoculated with 5000 IU SIVmac239M and started ART 9 days post-infection (dpi). After ART suppression (>200 dpi), RM were divided into treatment groups: ( A ) mRNA/SIV-Gag alone (n=8), ( B ) mRNA/control alone (n=8), ( C ) mRNA/SIV-Gag + AZD5582 (n=6) and ( D ) mRNA/Control + AZD5582 (n=5). Vaccine was administered intramuscularly (100ug/dose) at weeks 0, 3, 6, 20, plus a final immunization 2 weeks before ATI (week 34+). AZD5582 was administered IV as 15 weekly doses of 100ug/kg each between weeks 10-19 and 24-28. SIV-specific T cell responses were assessed by intracellular cytokine stimulation (ICS) assay and plasma viral loads (pVL) were quantified by ddPCR. Results: Prior to AZD5582, Gag-specific CD8+ T cells in bronchoalveolar lavage (BAL) peaked at week 6 (prior to AZD5582) to a mean of 18.1 ± 6% in group C and 25.0 ± 4.4% in group A (p=0.35) ( Figure 1A ). However, at week 15, after 5 doses of AZD5582, responses sharply declined to a mean of 0.2 ± 0.2% in group C compared to 7.2 ± 2.6% in group A (p=0.0013). By week 18, lymph node responses were also lower in group C compared to group A (mean 0.10 ± 0.09% vs 0.56 ± 0.03%, respectively, p=0.02). After the 4th vaccination, responses in BAL (AUC weeks 21-26) were lower in group C compared with group A (p=0.048). Upon ATI, viral rebound was delayed in group A compared to group B controls (median 20 vs 11.5 days, p=0.005) ( Figure 1B ). However, there was no difference in time to SIV rebound between groups C and D (median 13 vs 11.5 days, p=0.86). While post-ART peak pVL and viral burden (pVL AUC 0-8 weeks post-rebound) were lower in group A compared to group B (p=0.002 and p=0.049, respectively), no difference was observed between groups C and D (p=0.29 and p=0.55, respectively). Conclusions: Collectively, these data suggests that AZD5582 diminished mRNA/SIV-Gag vaccine-induced CD8+ T cell responses and abrogated vaccine enhanced early intercept of rebounding infections following ATI. This highlights the need for novel latency-reversal agents that do not disrupt the effector activity of CD8+ T cell responses.

530

Poster Abstracts

529

CROI 2025 137