CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

POSTER ABSTRACTS

167 RHESUS PRIMARY CELL IN VITRO SHIV INFECTION ASSAY CORRELATES WITH IN VIVO REPLICATION Adrienne E. Swanstrom 1 , Sean O’Brien 1 , Paul Bieniasz 2 , Theodora Hatziioannou 2 , Mario Roederer 3 , Brandon Keele 1 , Gregory Q. Del Prete 1 , Jeffrey D. Lifson 1 1 AIDS and Cancer Virus Program, Frederick, MD, USA, 2 The Rockefeller University, New York, NY, USA, 3 Vaccine Research Center, NIAID, Bethesda, MD, USA Background: Chimeric simian-human immunodeficiency viruses (SHIVs) containing an HIV-1 Envelope (Env) in an SIV backbone allow direct assessments of HIV-1 Env-based antibody and vaccine strategies in nonhuman primates (NHP) that are not feasible in SIV infection systems. However, it has been difficult to identify SHIVs able to consistently replicate in NHP at levels comparable to HIV-1 in humans or pathogenic SIVs in NHP. Recent publications have shown enhancement of rhesus CD4 use by HIV-1 Envs through distinct mutations at two positions within Env - 281 and 375 - often lead to improved SHIV replication in NHPs. While alterations at these two sites have consistently enhanced RhCD4 engagement by HIV-1 Envs of various clades, there are multiple possible amino acid substitutions at each site. Data to date suggest that the amino acid substitution at either 281 or 375 that yields the highest level of in vivo replication is unpredictably specific to each Env clone, resulting in multiple candidate clones to screen for Env-SHIV development. Given the cost and time required for such in vivo studies, a more efficient process is needed to identify SHIVs with potentially good replicative fitness in rhesus macaques. Methods: We have developed an in vitro infection assay that can be used to screen and downselect candidate SHIVs for in vivo studies. Enriched rhesus primary CD4+ T cells pooled from 3 naïve macaques were stimulated for 3 days with αCD2/CD3/CD28 beads and IL-2, then the beads were washed out and 1x10^6 cells were spinoculated with test SHIVs (MOI of 0.01). Input virus was washed out and cultures were maintained in IL-2 for up to 15 days with culture supernatants sampled every 2-4 days for p27 quantification by ELISA. Results: Results from this in vitro assay and results from co-inoculation in vivo competition fitness studies in macaques showed comparable rank order performance of different mutant versions of multiple SHIVs tested, identifying variants exhibiting ≥ one log lower p27 production as less fit for in vivo replication. Conclusion: As our understanding of how to manipulate HIV-1 Envs to enhance replication in NHPs improves, this assay will be useful to screen SHIVs for prioritization for in vivo studies. 168 FIRST REPORT OF A SUBTYPE B CONTAINING HIV-1 RECOMBINANT OF SUB-SAHARAN AFRICA ORIGIN Gustavo Kijak 1 , Erik Billings 1 , Eric Sanders-Buell 1 , Anne Marie O’Sullivan 1 , Nicaise Ndembi 2 , Sylvia Adebajo 3 , Afoke Kokogho 4 , Mark Milazzo 5 , Merlin L. Robb 5 , Nelson L. Michael 6 , Sodsai Tovanabutra 1 , Man Charurat 2 , Trevor Crowell 5 , Julie Ake 6 1 US Military HIV Research Program, Silver Spring, MD, USA, 2 Institute of Human Virology Nigeria, Abuja, Nigeria, 3 Population Council, New York, NY, USA, 4 Walter Reed Program–Nigeria, Abuja, Nigeria, 5 US Military HIV Research Program, Bethesda, MD, USA, 6 Walter Reed Army Institute of Research, Silver Spring, MD, USA Background: Subtype B is the major HIV-1 subtype in Western Europe, the Americas, and Australia, and also circulates in Southeast Asia and China. In all of these regions, co-circulation with other subtypes has led to the emergence of unique recombinant forms (URFs) and circulating recombinant forms (CRFs). Data published to date indicates a limited dissemination of subtype B, and B-containing URFs and CRFs in sub-Saharan Africa. Here we describe the circulation of recombinant strains containing subtype B and CRF02_AG among a cohort of men who have sex with men (MSM) in Lagos, Nigeria. Methods: Between 2013 and 2016, the Lagos arm of the RV368/TRUST study enrolled 672 Nigerian MSM. Within this group, the HIV-1 sero-prevalence was 59%. HIV-1 pol sequences (HXB2: 2273-3869) were obtained from 150 plasma samples with VL ≥ 1000 copies/mL. Samples with subtype B genetic

material were further characterized by full–genome sequencing. HIV-1 subtype assignment and recombinant analysis were performed combining phylogenetic analysis, visual inspection, Simplot (v3.5.1), and subtyping tools (jpHMM, NCBI genotyping). A genetic distance of ≤ 1.5% and bootstrap supported maximum- likelihood branching were used to identify linked transmission networks. Results: The predominant subtype was CRF02_AG (57%), followed by CRF02/B recombinants (15%), subtypes G (13%), A1 (3%), B (2%), and other recombinants containing CRF02, A1, or G. Transmission networks involved 24 participants in 9 clusters containing 2 to 5 participants per cluster. Within the main CRF02/B cluster, 5 strains were identified and subjected to near-full length sequence analysis. Three of the full-length genomes constitute the new CRF95_02B, containing RT and int genes from subtype B with the remainder comprised of CRF02_AG. Conclusion: The newly isolated CRF95_02B and the large cluster of strains associated with it, demonstrate the need for increased epidemiological monitoring of all groups engaged in high-risk activities. With high HIV-1 prevalence, a new CRF, and multiple transmission networks, this cohort of Nigerian MSM represents a previously hidden reservoir of HIV-1 strains that will need to be considered during vaccine immunogen selection and development. The views expressed are those of the authors and should not be construed to represent the positions of the U.S. Army or the Department of Defense. 169 PHYLOGENETIC CLUSTERS OF HIV-1 REVEAL POTENTIAL VIRAL GENETIC IMPACT ON COMORBIDITIES Katharina Kusejko 1 , Dominique L. Braun 1 , Manuel Battegay 2 , Enos Bernasconi 3 , Alexandra Calmy 4 , Matthias Cavassini 5 , Matthias Hoffmann 6 , Jürg Böni 7 , Sabine Yerly 4 , Thomas Klimkait 8 , Matthieu Perreau 5 , Andri Rauch 9 , Roger Kouyos 1 , Huldrych F. Günthard 1 1 University Hospital Zurich, Zurich, Switzerland, 2 University Hospital Basel, Basel, Switzerland, 3 Servizio di Malattie Infettive, Lugano, Switzerland, 4 University Hospitals of Geneva, Geneva, Switzerland, 5 Lausanne University Hospital, Lausanne, Switzerland, 6 St. Gallen Cantonal Hospital, St. Gallen, Switzerland, 7 University of Zurich, Zurich, Switzerland, 8 University of Basel, Basel, Switzerland, 9 University Hospital of Bern, Bern, Switzerland Background: Understanding and addressing comorbidities, co-infections, and AIDS-defining illnesses in HIV-infected patients remains one of the major challenges in managing HIV-infection in the era of highly active antiretroviral therapy. Here we tested at the population level whether viral genetic data can be informative to discern the differences in occurrence of comorbidities respective co-infections based on similarity of viral genomes in addition to demographic and clinical parameters. Methods: Using HIV-1 pol-sequences of ~11,000 patients of the drug resistance database of the Swiss HIV Cohort Study (SHCS) as well as 240,000 Los Alamos background sequences, we identified phylogenetic clusters of SHCS patients. The occurrence of comorbidities, co-infections and AIDS- defining illnesses in these patients was then analyzed with respect to those demographical and clinical confounders, which revealed a clustering pattern in the HIV-1 transmission network, by applying mixed effects logistic models. Results: Overall, HIV-related thrombocytopenia, Kaposi’s sarcoma and HIV- associated encephalopathy exhibited a significant (see Table ) phylogenetic signal after adjusting for confounders suggesting a potential role of viral genetic factors for these diseases. In addition, the co-infections Hepatitis C, Hepatitis B and Cytomegalovirus revealed a strong phylogenetic clustering after adjusting for confounding suggesting shared transmission routes for these infectious conditions. The phylogenetic signal for diabetes mellitus, cardiovascular diseases, neoplasms, syphilis and candida stomatitis could be explained by the clustering of those demographical and clinical confounders which showed a high within-cluster similarity, such as age, sex, risk group, ethnicity, length of HIV infection and antiretroviral treatment. Conclusion: This new type of analysis of combining viral sequences and well- defined clinical endpoints could be useful in triggering targeted pathogenesis

Poster Abstracts

60

CROI 2018