CROI 2015 Program and Abstracts

Abstract Listing

Poster Abstracts

269 Rectal Tenofovir Gel Usage Is AssociatedWith Changes in the Mucosal Proteome Adam Burgener 1 ; Laura Romas 1 ; Kenzie Birse 1 ; Kenneth H. Mayer 3 ; Irma Febo 5 ; Ross Cranston 2 ; Alex Carbarballo-Dieguez 4 ; Ian McGowan 2

1 University of Manitoba, Winipeg, Canada; 2 University of Pittsburgh, Pittsburgh, PA, US; 3 Harvard Medical School, Boston, MA, US; 4 Columbia University, New York, NY, US; 5 University of Puerto Rico, San Juan, US Background: After the success of the CAPRISA-004 clinical trial showing efficacy of vaginally-applied tenofovir (TFV)-based gels at preventing HIV infection, rectal formulations are currently being explored as a prevention option. However, further research is needed to characterize the safety of rectal microbicide gels. Standard mucosal safety tests, such as inflammatory cytokine measurements, may miss important immunological or biological side effects. Advances in systems biology tools, such as mass spectrometry, allow the monitoring of hundreds of mucosal immune factors simultaneously. Here we performed a comprehensive proteomic analysis to explore the impact of tenofovir 1% gel when applied rectally in the Project GEL study (MTN). Methods: Project GEL was a phase 1 randomized, double-blind, multi-site, placebo- controlled trial in which in which 24 participants were randomized 1:1 to receive rectal tenofovir 1% gel or (HEC) universal placebo gel. Study participants received one dose at first visit under observation, and then after a one-week recovery period, seven consecutive once-daily doses. Rectal mucosal swabs were collected on each visit. Mucosal samples were then analyzed by label-free tandem-mass spectrometry. Data was normalized and log transformed, followed by statistical analysis including paired (time-effects) and unpaired (across study arm) t- tests to identify differentially expressed proteins. Only high power (low technical variance) proteins were utilized in downstream analysis. Results: Over 249 unique proteins were identified in rectal mucosal samples. Within the TFV-arm, 7% (17/249, p<0.05) and 10% of total proteins changed (25/249, p<0.05) after 1 and 8 daily applications of TFV-gel, respectively, compared to 3% (7/249, p<0.05) and 6% (16/249,p<0.05) in the HEC arm. Biofunctional analysis utilizing DAVID gene ontology toolset indicated the TFV arm differentially expressed proteins were strongly associated with epidermal development (p=5.1E-9) and structural activity functions (p=1.5E-07), whereas HEC arm had milder associations with cell activation (p=4.3E-3) after 7 days usage. Conclusions: These preliminary studies suggest that rectal TFV-gel usage may be associated with mucosal proteome changes related to epidermal development. Further long- term studies are warranted for validation of these biomarkers including the impact of TFV-based rectal microbicide gels on the overall mucosal proteome. 270 Links Between Systemic and Mucosal Immunity in Treated HIV Infection Talía Sainz 1 ; Sergio Serrano-Villar 2 ; Gregory Melcher 3 ; Zhong-Min Ma 3 ; Christopher Miller 3 ; Netanya S. Utay 4 ; Basile Siewe 5 ; PaoloTroia-Cancio 3 ; Elizabeth Sinclair 6 ; David Asmuth 3 1 Hospital Universitario Gregorio Marañón, Madrid, Spain; 2 Hospital Ramon y Cajal, Madrid, Spain; 3 University of California Davis, Davis, CA, US; 4 University of Texas Medical Branch, Galveston, TX, US; 5 Rush University Medical Center, Chicago, IL, US; 6 University of California San Francisco (UCSF), San Francisco, CA, US Background: Recent data suggest that biomarkers of gut epithelial barrier dysfunction, bacterial translocation and inflammation predict mortality and are possibly linked to altered gut immunity. A better understanding of the underlying mechanisms might help to design interventions targetting chronic inflammation. Methods: We assessed biomarkers of inflammation (IL-6) and gut barrier (lipoteichoic acid [LTA], sCD14 and zonulin) in a clinical trial comparing three different ART initiation regimens on 33 ART naïve HIV+ subjects. Peripheral blood, rectal and duodenal biopsies were obtained at baseline and 9 months after randomization. Immunohistochemistry of biopsies was performed to determine CD4 T cells/mm 2 . Tissue was digested into single-cell suspensions for immunophenotyping. Spearman’s rank correlation coefficients were assessed at month 9 and associations between changes in continuous variables over time were analyzed using linear mixed models with random intercepts. Results: In duodenum, T-cell density correlated with plasma markers of gut barrier integrity: for CD4 T-cells and zonulin, Rho 0.59, P=0.002; for CD8 T-cells and lipoteichoic acid, Rho 0.67, P<0.001. Frommaturational subsets, we observed significant correlations between zonulin and naive (CCR7+CD45RA+) CD8-T cells in colon and duodenum (Rho -0.51, P<0.01 and Rho -0.58, P<0.01) as well as with EM (CCR7+CD45RA-) CD8-T cells in colon and duodenum (Rho -0.58, P<0.01 and Rho -0.44, P<0.01). The naive/memory CD8 T-cell ratio in colon and duodenum correlated with zonulin levels (Rho 0.52, P<0.01and Rho 0.54, P<0.01, respectively). Transitional memory (TM) (CCR7-CD45RA-CD27+CD28+) CD4 T cells correlated with IL-6 levels (Rho -0.51, P<0.01) Linear mixed models identified variations of intestinal T cell phenotypes significantly associated with changes of systemic markers: in duodenum, CD4 and CD8-T cells/mm 2 , CD8+EM T cells and naive/memory CD8-T cell ratio predicted changes in zonulin levels, while in colon changes of CD4+TM predicted decreases of IL-6.

Poster Abstracts

Conclusions: We observed a number of significant correlations between different intestinal T-cell phenotypes and systemic markers of inflammation and gut barrier integrity, suggesting that strategies aimed at modulating chronic inflammation may require targeting the gut immunity and/or gut microbial community composition.

233

CROI 2015