CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

736 TESAMORELIN IMPROVES FAT QUALITY INDEPENDENT OF CHANGES IN FAT QUANTITY Jordan E. Lake 1 , Kristine M. Erlandson 2 , Stefan Adrian 3 , Abanti Sanyal 4 , Ann Scherzinger 5 , Michael Dube 6 , Takara L. Stanley 7 , Steven K. Grinspoon 7 , Jean- Claude Mamputu 8 , Christian Marsolais 8 , Todd T. Brown 4 , Grace A. McComsey 9 1 University of Texas at Houston, Houston, TX, USA, 2 University of Colorado Anschutz Medical Campus, Aurora, CO, USA, 3 University of Colorado Denver, Denver, CO, USA, 4 Johns Hopkins University, Baltimore, MD, USA, 5 University of Colorado, Aurora, CO, USA, 6 University of Southern California, Los Angeles, CA, USA, 7 Harvard University, Cambridge, MA, USA, 8 Theratechnologies, Inc, Montreal, QC, Canada, 9 Case Western Reserve University, Cleveland, OH, USA Background: Fat quality may contribute to metabolic disease and inflammation as much or differently than fat quantity. HIV-infected (HIV+) adults have multiple risk factors for fat dysfunction, including HIV and antiretroviral therapy (ART). Both visceral (VAT) and subcutaneous (SAT) fat quality can be measured on computed tomography (CT) by measuring fat density (greater density=smaller, better quality adipocytes). Tesamorelin, a growth hormone-releasing hormone analogue, reduces VAT quantity in some HIV+ adults with central adiposity, but its effect on fat density is unknown. Methods: Participants were selected from two completed, randomized (2:1) trials of tesamorelin vs placebo for the treatment of central adiposity in HIV+ adults. Included participants had a clinical response to tesamorelin (defined as a VAT area decrease ≥8%, ≈70% of tesamorelin-treated participants) or were randomized to placebo. Week 0 and 26 abdominal (L4-5) CT scans were re-analyzed for VAT and SAT density (in Hounsfield Units, HU) by a central lab blinded to treatment arm. Paired t tests and linear regression models assessed 26-week, between-group differences in fat density changes. Results: Participants (193 responders, 148 placebo) were mostly male (87%) and Caucasian (83%). Arms were similar (p>0.10) at baseline in regards to sex, race/ethnicity, age, adiposity, concomitant medications, CD4+ T-cell count, HIV-1 RNA, ART, and time since HIV diagnosis. Baseline mean VAT and SAT HU were -91 and -94 in the tesamorelin arm, and -91 and -95 in the placebo arm (SAT p=0.29, VAT p=0.80). Over 26 weeks (Fig.), mean (SD) VAT density increased 6.2 (8.7) HU in the tesamorelin arm vs 0.3 (4.2) HU for placebo (p<0.0001). The effect was attenuated but persisted after controlling for baseline VAT HU and area, and VAT area change (2.3 HU, 95% CI [4.5, 7.3], p=0.001). Mean (SD) SAT density increased 4.0 (8.7) HU in the tesamorelin arm vs 0.3 (4.8) HU for placebo (p<0.0001), with no significant attenuation of effect after controlling for baseline SAT HU and area, and SAT area change (3.5 HU, 95% CI [2.3, 4.7], p<0.001). Conclusion: In HIV+ adults with central adiposity who responded to tesamorelin, VAT and SAT density increased independent of changes in fat quantity. These findings suggest that tesamorelin improves VAT and SAT quality in HIV+ adults in addition to reducing VAT quantity. Additional studies will determine whether these changes in fat density are associated with improvements in cardiometabolic and inflammatory parameters.

in HIV-infected adults, with lower density reflecting larger, poorer quality adipocytes. We assessed relationships between SAT and visceral AT (VAT) density and cardiometabolic and inflammatory parameters among HIV-1- infected adults enrolled in a completed ART initiation trial. Methods: AIDS Clinical Trials Group study A5224s participants were included in baseline analyses (n=54) if they had both abdominal SAT biopsy and CT data, and 96-week analyses (n=30) if they remained on their original randomized antiretroviral therapy (ART) regimen, had HIV-1 RNA <50 copies/mL and had biopsy or CT data. Partial Spearman’s correlations adjusting for AT area assessed relationships between AT density (in Hounsfield Units, HU) and metabolic and inflammatory parameters. The Jonckheere-Terpstra test assessed significance in metabolic syndrome analyses. Results: At baseline, median age was 40 years, CD4+ T lymphocyte count 219 cells/mm 3 , body mass index (BMI) 26.0 kg/m2, SAT area 199 cm2 and density -100 HU, VAT area 83 cm2 and density -83 HU; 89%were male and 67%white. Greater baseline SAT density correlated with lower triglyceride levels; greater VAT density correlated with higher HDL cholesterol (Table). Baseline AT density did not correlate with HOMA-IR nor inflammatory biomarker levels. The number of metabolic syndrome components (0, 1-2, 3+) at baseline increased as VAT density decreased (-75, -83 and -90 HU, p=0.002), irrespective of total adiposity (BMI <25 p=0.02, BMI 25+ p=0.04). Over 96 weeks of ART, SAT (+18%) and VAT (+35%) area increased, and SAT (-3%) and VAT (-6%) density decreased. Decreases in SAT HU correlated with increases in hs-CRP, IL-6, soluble TNFRI/II and sICAM-1 independent of SAT area (Table). Decreases in VAT HU correlated with increases in IL-6 and soluble TNFRII independent of VAT area. Conclusion: SAT and VAT quality, as measured by CT density, are associated with metabolic disturbances in ART-naïve HIV-infected adults, and inflammatory biomarker concentrations on ART, independent of AT quantity. Increased AT area and decreased AT density on ART (suggestive of hypertrophic AT expansion) is associated with increased systemic inflammation and has important implications for comorbidity development following ART initiation.

Poster Abstracts

738 HIV IMMUNE DYSREGULATION PREDICTS LEAN TISSUE AND FAT CHANGES DURING CHRONIC DISEASE

Louie M. Gangcuangco , Dominic Chow, Scott A. Souza, Glen M. Chew, Brooks I. Mitchell, Michelle L. D’Antoni, Jason T. Huynh, Lishomwa C. Ndhlovu, Cecilia Shikuma University of Hawaii at Manoa, Honolulu, HI, USA Background: Changes in body composition such as fat accumulation and lean tissue loss may impact physical function and mortality. We investigated the association between various HIV-associated immune parameters and 2-year body composition changes assessed by dual-energy x-ray absorptiometry (DXA) among HIV patients on stable ART. Methods: Longitudinal analysis of HIV+ individuals ≥40 years old on stable ART ≥3 months in the Hawaii Aging with HIV cohort. Multiparametric flow cytometry was performed on cryopreserved year 1 peripheral blood mononuclear cells to quantitate the percentages of monocyte (MO) phenotypes, and T cells expressing activation (HLA-DR/CD38) and exhaustion markers (PD-1/ TIM-3/TIGIT). Select plasma soluble biomarkers were measured by Luminex technology. Among subjects not on zidovudine and stavudine, multivariate linear regression analyses were performed to assess the association between T cell subsets with DXA changes. Results: Of 97 subjects, median age at enrollment was 52 (48, 57) years, BMI 26.5 kg/m2, CD4 count 502 cells/uL, and nadir CD4 count 150 cells/uL. Majority were males (87.6%), Caucasian (56.7%), and 82.5% had undetectable HIV RNA < 50 copies/mL. Median (Q1, Q3) body composition changes: total fat 10.0

737 FAT QUALITY IS INDEPENDENTLY ASSOCIATED WITH CARDIOMETABOLIC RISK IN HIV INFECTION Jordan E. Lake 1 , Carlee Moser 2 , Liz Johnston 2 , Clara Magyar 3 , Scott D. Nelson 3 , Kristine M. Erlandson 4 , Todd T. Brown 5 , Grace A. McComsey 6 1 University of Texas at Houston, Houston, TX, USA, 2 Harvard University, Cambridge, MA, USA, 3 University of California Los Angeles, Los Angeles, CA, USA, 4 University of Colorado Anschutz Medical Campus, Aurora, CO, USA, 5 Johns Hopkins Hospital, Baltimore, MD, USA, 6 Case Western Reserve University, Cleveland, OH, USA Background: Adipose tissue (AT) quality and quantity may have independent metabolic effects. Computed tomography (CT)-quantified abdominal subcutaneous AT (SAT) density reflects biopsy-quantified SAT adipocyte size

CROI 2018 275