CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

Methods: Cell-associated SHIV DNA levels were measured in PBMCs during acute infection in 8 macaques infected during concurrent PrEP with FTC/TDF combination or single- agent tenofovir alafenamide fumarate (TAF). Macaques continued treatment with 1-2 weekly drug doses to model suboptimal drug exposure during undiagnosed HIV infection in humans. Peak SHIV DNA and area under the curve values (AUC) over 5 or 20 weeks, as well as RNA levels, were compared to the values seen in untreated SHIV infections (n = 10). SHIV DNA levels were also measured in lymphoid tissues collected from FTC/TDF or maraviroc PrEP breakthroughs after 1 year of infection. Results: PrEP breakthrough infections had reduced plasma RNA viremias relative to untreated infections both at peak and during the first 20 weeks of infection (p<0.005). SHIV DNA levels in PBMCs were also reduced in PrEP breakthrough infections both at peak and at week 5 ( p = 0.022 and p = 0.043, respectively), but not after 20 weeks of infection. At 1 year, SHIV DNA reservoirs in lymphoid tissues were similar in size among macaques that received PrEP with FTC/TDF (median = 464 SHIV DNA copies/10 6 cells; range, 40-5,346), PrEP with maraviroc (median = 1088 copies/10 6 cells; range, 554-10,090), or placebo (median = 952 copies/10 6 cells; range, undetectable-24,668) (p > 0.05). Conclusions: Antiviral drug activity due to PrEP limits acute SHIV replication but has only a transient effect on cell-associated DNA levels in PBMCs and lymphoid tissues. Our model suggests that suboptimal drug exposure in persons that are taking PrEP and become infected with HIV may not be sufficient to reduce the pool of HIV-infected cells, and that treatment intensification may be needed to sustain the virologic benefit from the PrEP regimen. 334 Germinal Centre T and B Cells in Lymph Node Fine-Needle Biopsies During HIV Infection John J. Zaunders 1 ;William Hey-Cunningham 2 ;Yin Xu 2 ; Chester Pearson 2 ; Michelle Bailey 2 ; Brad Milner 1 ; Andrew Field 1 ; David Cooper 3 ; Anthony Kelleher 2 ; Kersten K. Koelsch 4 1 St Vincent’s Hosp, Darlinghurst, Australia; 2 Univ of New South Wales, Sydney, Australia; 3 Kirby Inst, Sydney, Australia; 4 Kirby Inst, Univ of New South Wales, Sydney, Australia Background: Measurement of HIV-1 reservoirs is usually studied in peripheral blood (PB), but 98% of CD4 T cells are in lymphoid tissues and other organs. Histology has shown characteristic hyperplastic germinal centres (GC) in lymph nodes, recently associated with increases in the T follicular helper cell (Tfh) subset of CD4 T cells, and there is now recognition that GC and Tfh are an important reservoir of HIV-1. We investigated whether ultrasound-guided fine needle biopsies (FNB) from peripheral lymph nodes (LN) provide a minimally invasive means of longitudinally accessing HIV-1 reservoirs in GC . Methods: FNB of inguinal LN were performed on 10 healthy adult controls (HC), 11 HIV+ ART-naive and 10 HIV+ on ART subjects, without adverse events. FNB were analyzed for CD3+CD4+CD45RAnegPD-1hiCXCR5+ICOS+CD127dim Tfh cells and CD19+CD20highCD38highIgDnegKi-67+Bcl-6+ GC B cells, using TruCount tubes on a 4-laser LSR II. CD4 T cells, from FNB and PB, respectively, were purified by cell sorting, from 19 patients, and HIV DNA and cell-associated (CA) HIV RNA were quantified by real-time PCR. Results: Overall, the median number of CD4 T cells obtained by FNB from HC was 887,000 (IQR: 206,000-1,187,000), and from all 21 HIV+ subjects (519,000 (IQR: 228,000- 1,340,000)). However, the median Tfh cell number in FNB from ART-naïve HIV+ subjects was 46,300, significantly higher than in HC (5,600; p=0.04); HIV+ on ART subjects had a median of 15,800 Tfh cells. The median GC B cell number from ART-naïve HIV+ was 153,493 cells, significantly higher than in HC (median: 1,506; p<0.01). For HIV+ on ART, the median GC B cell number was 30,749 cells, so that 4/10 were above HC normal range. Cell sorting from HIV+ FNB yielded a median of 295,000 CD4+ T-cells (purity>96%). HIV DNA was quantified in 19 out of 19, CA HIV RNA in 13 out of 17, samples. HIV DNA copies per 10 6 CD4+ T cells were higher in CD4+ T cells purified from LN FNB compared to PB in both ART-naïve (3.22 vs 2.56, p=0.02) and ART (3.21 vs 3.08, p=0.13) groups. Unspliced HIV RNA copies per 10 6 CD4+ T cells were similarly elevated in LN FNB samples from both the ART-naïve (3.97 vs 3.05, p<0.01) and ART (3.66 vs 2.95, p=0.13) groups. Conclusions: Ultrasound guided FNB of unenlarged LN was well-tolerated, and provided sufficient CD4+ T cells and B cells for high-dimensional flow cytometry and cell sorting for PCR, thereby providing access to Tfh and GC B cells for HIV reservoir, pathogenesis, therapy and vaccine studies. 335 Tcm CD4 T-Cell Infection Associates With Immune Failure and HIV Persistence on ART Vincent C. Marconi 1 ; Emily Ryan 2 ; Luca Micci 3 ; Amelie Pagliuzza 4 ; Sol Aldrete 1 ; Keith Delman 1 ; Jacob D. Estes 5 ; Nicolas Chomont 6 ; Rafick Sekaly 7 ; Mirko Paiardini 2 1 Emory Univ Sch of Med, Atlanta, GA, USA; 2 Yerkes Natl Primate Rsr Cntr, Emory Univ, Atlanta, GA, USA; 3 Emory Univ, Atlanta, GA, USA; 4 CRCHUM, Montreal, QC, Canada; 5 Frederick Natl Lab, Leidos Biomed Rsr, Frederick, MD, USA; 6 Univ de Montréal, Montreal, QC, Canada; 7 Case Western Reserve Univ, Cleveland, OH, USA

Background: Morbidity and mortality especially from severe non-AIDS conditions are increased for individuals with persistently low CD4 T cell counts and/or persistent immune activation despite virologic suppression on antiretroviral therapy (ART). Current treatments to increase CD4 counts or reduce immune activation have been unable to reduce morbidity or mortality. It is unclear if differential infection of CD4 T cell subsets could further explain divergent CD4 response to ART and maintenance of the latent HIV reservoir. Methods: Thirty HIV-infected subjects were enrolled in a cross- sectional cohort based upon their CD4 response to ART-mediated virologic suppression. Immunologic responders (IR) were defined as having CD4 counts >500 cells/µL < 2 years after ART initiation. Immunologic non-responders (INR) were defined as having CD4 counts <350 cells/ µL > 2 years after ART initiation. The frequency of cells harboring total and integrated HIV DNA were measured in sorted naïve (N), central memory (CM), transitional memory (TM), and effector memory (EM) CD4 T cells for INR and compared to IR. Immunological parameters - including frequencies of CD4 and CD8 T cells and of their N, CM, TM, and EM subsets, their levels of activation/ proliferation, and the expression of immune checkpoint molecules - were analyzed by flow cytometry. Results: Median age was 46, 90%were male, 82.8%were African- American. The frequency of cells harboring total and integrated HIV DNA were lowest for N CD4 cells while the other subsets were equivalent. Total and integrated HIV DNA levels were highly correlated (r=0.9240; P<0.0001) and significantly higher for INR

Poster Abstracts

than IR in all CD4 T cell subsets (P<0.01). Total/integrated (T/I) HIV DNA ratio was lowest for EM CD4 cells and highest for N CD4 cells, consistent with a higher susceptibility to HIV infection of EM CD4 T cells. Interestingly the T/I ratio was significantly lower in INR than IR only for CM CD4 cells (p=0.0149). Levels of proliferation (Ki-67+) and expression of immune checkpoint molecules previously associated with HIV persistence (PD-1 and TIGIT) were significantly higher in both bulk and CM CD4 T cells of INR. Conclusions: INR have higher levels of HIV DNA in all subsets when compared to IR, thus highlighting a link between the lack of immune reconstitution and viral persistence. Furthermore, INR having a lower T/I ratio only in central memory CD4 cells as compared to IR suggests a higher potential for proviral integration and productive infection of the central memory CD4 cells as an important virologic feature of INR.

127

CROI 2016