CROI 2024 Abstract eBook

Abstract eBook

Poster Abstracts

further evaluation of the described antibody combination for epidemic/ pandemic response and preparedness. Safety and Immunogenicity of Month 30 Boost of ALVAC+gp120/MF59 Preventive HIV Vaccines Vimla Naicker 1 , Fatima Laher 2 , Kelly Seaton 3 , Stephen C. De Rosa 4 , Lynn Morris 5 , Nonhlanhla N. Mkhize 6 , Linda-Gail Bekker 7 , Mookho Malahleha 8 , Kathy T. Mngadi 9 , Jack R. Heptinstall 3 , David C. Montefiori 10 , Juliana Mc Elrath 4 , Georgia D. Tomaras 10 , Zoe Moodie 11 , for the HVTN 100 Study Team 1 South African Medical Research Council, Durban, South Africa, 2 Perinatal HIV Research Unit, Soweto, South Africa, 3 Duke University, Durham, NC, USA, 4 Fred Hutchinson Cancer Center, Seattle, WA, USA, 5 University of the Witwatersrand, Johannesburg, South Africa, 6 National Institute for Communicable Diseases, Johannesburg, South Africa, 7 Desmond Tutu HIV Foundation, Cape Town, South Africa, 8 Synergy Biomed Research Institute, East London, UK, 9 The Aurum Institute, Johannesburg, South Africa, 10 Duke University School of Medicine, Durham, NC, USA, 11 Fred Hutchinson Cancer Research Center, Seattle, WA, USA Background: HVTN 100, a phase 1-2 preventive HIV vaccine trial in South Africa, administered subtype C- containing ALVAC-HIV (vCP2438) at months 0 and 1, and ALVAC-HIV with bivalent subtype C gp120/MF59 at months 3, 6 and 12 in Part A. IgG binding antibody and T-cell responses were similar or greater at month 12.5 compared to month 6.5 then waned by month 18. In Part B we boosted after an 18-month interval at month 30. Methods: Part B vaccinations were administered to eligible Part A participants, randomized to either ALVAC- HIV + gp120/MF59(n=32), or gp120/MF59 alone(n=31) and placebo group was re-administered placebo(n=7). At months 30, 30.5 and 36, we measured envelope (Env)-specific serum binding antibodies by binding antibody multiplex assay (BAMA) and HIV-specific CD4 T-cell responses by intracellular cytokine staining assay, and neutralization by reductions in Tat-regulated luciferase reporter gene expression in TZM-bl cells. Results: Vaccine groups had an acceptable safety profile based on reactogenicity, adverse events and laboratory tests. There were no statistically significant differences in IgG binding antibody response rates or magnitudes between the two vaccine groups for any gp120, gp140, or V1V2 antigens at any timepoint. Vaccine groups had positive IgG responses to all V1V2(n=27), gp120(n=11) and gp140 antigens(n=9) at all timepoints but were more common for gp120 and gp140 antigens and highest at month 30.5. Booster vaccination restored the magnitude-breadth IgG response to V1V2 antigens at Month 30.5 which waned by month 36 with median area under the curves(AUCs) for combined vaccine groups 0.13, 2.55 and 0.40 for months 30, 30.5 and 36 respectively (Fig 1). Month 30 boosting increased magnitude and durability of tier 1A neutralization responses but did not induce tier 2 neutralization responses. Month 30.5 tier 1A response magnitudes for the pooled vaccine group increased significantly compared to month 12.5, median 331 versus 756, p<0.0001 for TV1c8.2; 886 vs. 1763, p<0.0001 for MW965.26). CD4+ Env responses, rate and magnitude, to vaccine-matched antigens were seen at month 30, boosted at month 30.5, waned by month 36, with no significant differences between vaccine groups. Conclusion: Booster vaccination with gp120/MF59 given alone or with ALVAC after an 18-month interval was safe and induced binding, tier 1A neutralization and CD4+ T-cell responses similarly in both vaccine groups. Late boosting may increase breadth of responses and restore V1V2 binding antibody responses.

decades. We set out to address a critical barrier to the clinical use of AAV/ bnAb vectors: the high prevalence of pre-existing anti-AAV nAbs in humans. One way to maximize the impact of AAV/bnAb therapies is to determine the seroprevalence of muscle-tropic AAV capsids in areas with high incidence of HIV, so that people living with HIV (PLWH) who are seronegative for these AAV capsids can potentially benefit from AAV/bnAb therapies without any additional intervention. AAV epidemiology is well documented in developed countries but nearly absent for sub-Saharan Africa, home of most PLWH. Methods: To address this deficit, we partnered with the HVTN (HIV Vaccine Trials Network) and ACTG (AIDS Clinical Trials Group) to establish the prevalence and titers of anti-AAV nAbs in South Africa and Zimbabwe. We used HEK293T cells and luciferase-expressing vectors to screen sera from 300 healthy adult donors (HD) and 277 PLWH for nAbs against AAV-1, -8, and -9. Results: First, we screened sera at a 1:20 dilution and found that 15%, 79%, and 77% of HD samples displayed <50% neutralization of AAV-1, -8, and -9, respectively. We found no significant sex-specific differences in the prevalence of anti-AAV nAbs in this cohort. Among PLWH, 17%, 72%, and 73% of samples exhibited <50% neutralization of AAV-1, -8, and -9, respectively. Next, we determined the midpoint titers of anti-AAV-9 nAbs in the samples that exhibited >50% neutralization at a 1:20 dilution (i.e., AAV-9+). We selected AAV-9 for this in- depth analysis because of its superior performance to AAV-1 and -8 in promoting persistent mAb expression in nonhuman primates following intramuscular (IM) administration. Among HD, 75% of AAV-9+ donors exhibited titers of anti-AAV-9 nAbs below 1:160, a level that is not thought to impair IM AAV transduction in primates. We are currently determining the titers of anti-AAV9 nAbs among the AAV-9+ PLWH. Conclusion: In summary, nearly three quarters of HD and PLWH in South Africa and Zimbabwe could be eligible to participate in clinical trials of bnAb-encoding AAV-9 vectors delivered by the IM route. Machine Learning-Guided Generation of a Combination of Broadly Neutralizing Sarbecovirus Antibodies Grace Marden , Kimberly Schmitt, Hongru Li, Alex Ramos, Eric Carlin, Nadine Shaban, Geetika Sharma, Monica Menzenski, Anne Jecrois, Gevorg Grigoryan, Adam Root, Heather Van Epps, Kristen Hopson, Daria Hazuda, Francesco Borriello Generate Biomedicines, Somerville, MA, USA Background: Sarbecoviruses represent a subgenus of coronaviruses including strains mostly circulating in bats and at risk of zoonotic spillover as well as strains that can infect humans (SARS-CoV-1 and SARS-CoV-2). Over the past 20 years, SARS-CoV-1 was responsible for an outbreak in Asia while SARS-CoV-2 initiated a global pandemic and remains a continuous health threat especially for vulnerable populations. There is still the need to develop therapeutic measures that can effectively control human sarbecoviruses and potential zoonotic spillover events, thereby preventing future epidemics/pandemics. Several monoclonal antibodies have been discovered with broad neutralizing activity against sarbecoviruses by targeting conserved regions of the spike proteins such as class 3 and 4 RBD regions and S2 fusion machinery. Viral escape mutations accrued in the SARS-CoV-2 spike protein have rendered most of the antibodies targeting RBD regions ineffective. In addition, antibodies targeting the S2 fusion machinery have relatively low neutralization potency compared to anti-RBD antibodies and therefore their therapeutic utility has not been realized. Methods: Here we used a machine learning-guided protein engineering approach to optimize broadly neutralizing antibodies against sarbecoviruses. We screened computationally designed sequence sets for binding to spike proteins, neutralization of pseudoviruses and developability parameters to select lead molecules. Additional characterization of lead molecules as single agents and combination included: epitope mapping through cryo-EM and X-ray crystallography, assessment of binding affinities to spike proteins, neutralization of pseudoviruses and live viruses, in vitro escape experiments and in vivo hamster challenge with SARS-CoV-2 BA.2. Results: We rescued the neutralizing activity of a previously described class 4 anti-RBD antibody against Omicron variants. We also improved neutralization potency and efficacy of a previously described antibody with pan-sarbecovirus activity targeting the S2 stem helix peptide. Finally, we demonstrate that the combination of these optimized antibodies improves both neutralization of SARS-CoV-2 variants in vitro and suppression of viral replication in a hamster challenge model. Conclusion: Our work highlights the successful use of machine learning-guided protein engineering for optimization of anti-viral antibodies and supports

377

Poster Abstracts

376

88

CROI 2024