CROI 2019 Abstract eBook

Abstract eBook

Oral Abstracts

antiretroviral therapy (ART). However, the dynamics and evolution of the proviral reservoir in these individuals are largely unknown. Methods: 68 HIV-1 ECs with undetectable viral loads (viral load <50), 38 viremic controllers (VCs, viral load 50-2000), and 34 chronically ART-treated patients were included in this study. Genomic DNA was extracted from PBMCs and diluted to single genome levels for HIV-1 near full-length next generation viral sequencing. Quantitative viral outgrowth assays (QVOA) were performed with autologous CD4+ T cells; outgrowing virus was subjected to HIV-1 near full-genome sequencing. Results: We obtained 1066, 1385, and 1601 individual proviral sequences in ECs, VCs, and ART-treated patients, respectively. The median frequency of proviral species in ECs was significantly lower than in VCs (p=0.0009) and ART- treated patients (p<0.0001). The relative number of genome-intact sequences in ECs was also significantly lower when compared to ART-treated patients (p<0.0001), but was not different from VCs (p=0.2740). Among intact proviral genomes in ECs, 46%were clonally-expanded, a proportion considerably higher than in VCs (8%) but similar to ART-treated patients (31%). Notably, we identified 2 subgroups of ECs with markedly different intact reservoir sizes: one group of ECs had high proportions of intact proviral genomes within all detected proviral species, ranging from 13% to 100%; among these intact proviral genomes, very high proportions of clonal sequences were identified by full-genome sequencing (36%-80%) that frequently were entirely identical to sequences isolated from QVOA. In contrast, we observed 3 ECs in whom no intact proviral sequences were observed after assaying 52-76 million PBMCs for near full-genome sequencing and another 31-67.5 million PBMCs for QVOA, suggesting that these 3 patients may approximate a sterilizing cure of HIV-1 infection. Conclusion: This detailed analysis suggests that ECs can be distinguished into 2 different subgroups - ECs with unusually high proportions of intact proviral genomes and very few defective proviral sequences, and ECs with no intact proviral genomes detectable in large numbers of cells. Exploring the reasons for differential viral reservoir dynamics in these patients may allow us to identify mechanisms enabling a drug-free remission of HIV-1 infection. 136 INTERFERON Α2B REDUCES INDUCIBLE CD4-ASSOCIATED HIV IN ART- SUPPRESSED INDIVIDUALS Livio Azzoni 1 , Emmanouil Papasavvas 1 , Jay Kostman 2 , Pablo Tebas 3 , Karam Mounzer 2 , Ian Frank 3 , Kenneth M. Lynn 3 , Linden Lalley-Chareczko 2 , Rui Feng 3 , Scott Appel 3 , Bonnie J. Howell 4 , Daniel Holder 4 , Shih Lin Goh 4 , Guoxin Wu 4 , Luis Montaner 1 1 Wistar Institute, Philadelphia, PA, USA, 2 Philadelphia FIGHT, Philadelphia, PA, USA, 3 University of Pennsylvania, Philadelphia, PA, USA, 4 Merck & Co, Inc, West Point, PA, USA Background: In prior studies treatment of ART-suppressed individuals with pegylated interferon alpha (pIFNα) resulted in a decrease in PBMC-associated integrated HIV DNA. We sought to confirm the effect of pIFNα on latent HIV by measuring inducible HIV expression from CD4 cells isolated from chronically infected, ART-suppressed individuals receiving pIFNα-2b immunotherapy in a randomized clinical trial (NCT02227277). Methods: We enrolled 54 HIV-infected individuals receiving suppressive ART (HIV VL < 50 copies/ml) and with CD4 count > 450/μl who were randomized 1:1:1 to 3 treatment arms: 1. 1 μg/kg of pIFNα-2b (Pegintron, Merck) for 20 weeks, with ART interruption (start at week 4, resume ART upon confirmed VL > 50 copies/μl or at week 20) 2. 1 μg/kg of pIFNα-2b added to ART 3. ART only (control) All subjects were sampled at baseline and week 20. CD4+ T cells were isolated from PBMC and cultured (2-10 replicates) for 16-hour with medium or PMA + Ionomycin. HIV p24 production was measured in cell pellets using single molecule array (SIMOA) with a limit of quantification (LOQ) of 7 fg/106 cells. Analysis approach: modified intention to treat. Measurements < LOQ were censored at the LOQ; p24 levels were log-transformed. The changes in PMA-induced SIMOA HIV p24 levels at endpoint from baseline were compared between first two treatment and placebo arms, using a linear mixed-effect model, adjusted for site effect and within-subject correlation. Results: Of 54 enrolled subjects, 46 completed the protocol for primary analysis. We observed 6 treatment-emergent AEs grade ≥ 3 (4 treatment- related). All subject in Arm 1 achieved HIV VL < 50 c/ml after resuming ART. The median baseline cell-associated p24 was 22.88 fg/10 6 CD4+ T cells for Arm 1, 12.47 for Arm 2 and 34.12 for Arm 3. At week 20, the median p24 was 0.56 for Arm 1, 7.95 for Arm 2 and 57.69 for Arm 3. Compared to Arm 3 (ART only control),

that had a significant increase in p24 from baseline to week 20 (118%, p < 0.05), we detected a significant decrease in both treatment arm 1 (-73%) and arm 2 (-61%). Conclusion: Consistent with pilot trial results, a 20-week course of pIFNα-2b resulted in a significant decrease in levels of CD4 T cell-associated inducible HIV compared to ART alone. This effect was independent of ART interruption. This randomized study provides a strong rationale for the use of IFNα immunotherapy as a component of cure-directed strategies. 137 NEF-STOP REPAIR DYNAMICS, BUT NOT ANTI-Α4Β7, INFLUENCE POSTTREATMENT VIRAL CONTROL Michele Di Mascio 1 , Sharat Srinivasula 2 , Paula DeGrange 1 , Brandon F. Keele 2 , Michael Proschan 1 , Jeffrey D. Lifson 2 , H. Clifford Lane 1 1 National Institute of Allergy and Infectious Diseases, Bethesda, MD, USA, 2 Leidos Biomedical Research, Inc, Frederick, MD, USA Background: In contrast to published data, we have recently reported (AIDS 2018) that treatment with an antibody to α4β7 integrin in rhesus macaques infected with SIVmac239 having a stop codon in nef (SIVmac239nefstop) was not associated with prolonged significant post-treatment suppression of viremia. The only other clear difference was the rate of peripheral blood (PB) CD4 decline between the control groups of the two studies. The current analysis attempts to identify additional factors potentially responsible for the different outcomes observed Methods: Twenty-two Mamu-A001, B008 and B017 negative Indian rhesus macaques were infected i.v. with 200 TCID50 SIVmac239nefstop. At 5 weeks post-infection (wpi), combination anti-retroviral therapy (cART) was started and 4 weeks later, animals received a total of 8 infusions every 3 weeks of α4β7 antibody (n=12) or control antibody (n=10); cART was stopped at 18 wpi and animals were followed for additional ~7 months (set-point average 45 and 48 pwi, PVL-sp). In addition to plasma SIV RNA viral loads (PVL) and PB CD4 counts, levels of cell associated SIV RNA and DNA viral load (CAVL), were measured during cART administration and ~ 3 months post cART interruption in LN and rectal gut (RAL) biopsy samples Results: PVL peaked at week 2-5 wpi without a significant drop in PB CD4 counts by 2 wpi ( ~8% decline, n.s., n=22). A positive correlation was found between the frequency of nef-open restored viruses (FRV) and PVL at 2 wpi (r=0.66 P<0.001) as well as between FRV at 2 wpi and LN SIV-DNA CAVL (r=0.67, P<0.001), but not PVL or RAL SIV-DNA CAVL, at the time of cART initiation [when FRV was 100% in all animals]. PVL-sp was associated with LN SIV-DNA CAVL (r=0.62, P=0.002), but not with rectal SIV-DNA CAVL (r=0.07, n.s.) at the time of cART interruption. ~7 months following antiretroviral treatment completion mean PVL (~ 10^4 copies/ml), PB-CD4 T cell counts (~ 900 cells/ul), LN or rectal SIV CAVL were not significantly different between the two groups Conclusion: While a lack of exactly corresponding analyses from the published study precludes direct comparison, the current analysis suggests that differential rates of repair of the nef mutation may have contributed to the observed different outcomes between the two studies. Following SIV infection, faster viral dissemination in LNs appears to be facilitated by restoration of the virus to nef-open and predicts higher virologic set-point following cART interruption 138 SEARCH INTERVENTION REDUCES MORTALITY AT A POPULATION LEVEL IN MEN WITH LOW CD4 COUNT Moses R. Kamya 1 , Maya L. Petersen 2 , Dalsone Kwariisima 3 , James Ayieko 4 , Norton Sang 4 , Jane Kabami 3 , Tamara D. Clark 5 , Edwin D. Charlebois 5 , Laura B. Balzer 6 , Craig R. Cohen 5 , Vivek Jain 5 , Elizabeth A. Bukusi 4 , Gabriel Chamie 5 , Diane V. Havlir 5 1 Makerere University, Kampala, Uganda, 2 University of California Berkeley, Berkeley, CA, USA, 3 Infectious Diseases Research Collaboration, Kampala, Uganda, 4 KEMRI- UCSF, Kisumu, Kenya, 5 University of California San Francisco, San Francisco, CA, USA, 6 University of Massachusetts Amherst, Amherst, MA, USA Background: HIV Test-and-Treat has the potential to reduce mortality of HIV+ persons with low CD4+ counts on a population level by rapidly initiating ART among “late presenters” not previously in care and among persons disconnected from prior care. We evaluated the effect of streamlined ART delivery for HIV+ persons with CD4<350 cells/uL after population-wide HIV testing during the SEARCH study (NCT:01864603). Methods: At baseline (2013-2014), HIV testing at multi-disease health fairs and in homes reached 91% of 143,870 adult stable residents in 32 communities

Oral Abstracts

51

CROI 2019