CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

Poster and Themed Discussion Abstracts RAL (OR: 0.72 [95% CI: 0.53 to 0.99]; p=0.043), while odds of severe BMI gain were significantly lower for DRV/r vs RAL (OR: 0.73 [95% CI: 0.53 to 0.99]; p=0.041). Conclusion: Predictors of severe weight/BMI gain in this population included black race, higher baseline disease severity, and use of RAL. Understanding factors predisposing individuals to unhealthy weight gain may help better manage metabolic complications of HIV. 696 RALTEGRAVIR SWITCH AND BIOMARKERS OF LIVER STEATOSIS AND METABOLIC SYNDROME IN WOMEN David Reynoso 1 , Anoma Somasunderam 1 , Maitreyee Nigalye 1 , Judith S. Currier 2 , Netanya S. Utay 1 , Jordan E. Lake 3 1 Univ of Texas Med Branch at Galveston, Galveston, TX, USA, 2 Univ of California Los Angeles, Los Angeles, CA, USA, 3 Univ Texas Houston, Houston, TX, USA Background: Persons with well-controlled HIV infection on antiretroviral therapy (ART) are at risk for metabolic syndrome (MetS) and fatty liver disease. Hepatic steatosis and fibrosis and MetS have been associated with changes in circulating levels of adiponectin, soluble ST2 (sST2, or IL-33R), chitinase 3-like 1 (Chi3L1, or YKL40), hyaluronic acid (HA), TIMP-1, lysyl oxidase-like-2 and TGF-β in non-HIV-infected populations and animal models. Protease (PI) and non-nucleotide reverse transcriptase inhibitors (NNRTI) may contribute to MetS and other comorbidities. The effect of switching from PI- or NNRTI-based regimens to raltegravir-based regimens on these biomarkers is unknown. Methods: Plasma was obtained from a completed, prospective trial of 37 women with lipohypertrophy and well-controlled HIV infection on NNRTI- or PI-based regimens who were randomized to immediate vs delayed (24 weeks) switch to raltegravir. We quantified the above biomarkers by ELISA and Multiplex assays at baseline and 24 weeks after randomization. Wilcoxon signed-rank test evaluated within group changes. We investigated correlations among biomarkers and clinical covariates with nonparametric (Spearman) and parametric (linear regression) analyses. Associations were also evaluated by regression modeling. Results: Participants had median age of 43 years, CD4 558 cells/mm3 and BMI 32 kg/m2; 35%met criteria for MetS. At baseline, higher adiponectin levels correlated with higher Chi3L1 levels (r=0.42, P=0.02), as did changes after 24 weeks (r=0.40, P=0.03). Baseline sST2 levels correlated with HA (r=0.52, P=0.003) and TIMP-1 levels (r=0.48, P=0.006); changes in sST2 correlated with changes in Chi3L1 (r=0.43, P=0.02) and adiponectin (r=0.40, P=0.03). Adiponectin and Chi3L1 levels decreased more in women switched to raltegravir immediately compared to those continuing NNRTI- or PI-based ART (Table). Other biomarkers did not change significantly. Adiponectin levels increased 10% per 1 mg/ dL HDL increase. Adiponectin (1453 vs 3346 ng/ml, P=0.01) and sST2 (8473 vs 13206, P=0.02) were lower in participants with MetS vs without MetS. Adiponectin levels were also lower among women with higher subcutaneous adipose tissue volumes. Conclusion: In women with HIV and lipohypertrophy, the hepatic steatosis/fibrosis marker Chi3L1 and the adipokine adiponectin decreased with switching to raltegravir. Whether switching from NNRTI/PI-based regimens to raltegravir would improve hepatic steatosis and dysmetabolism requires further study. Methods: 40 overweight or obese HIV-infected patients (49.9± 8.8 years of age; BMI of 34.2±34.2), with an undetectable viral load and CD4 count >200 were randomly assigned to a fully-automated Internet-delivered behavioral Weight Loss program (WT LOSS) or Internet Education Control. The behavioral weight loss program includes 12 weekly video lessons, a platform to submit self-monitoring data, and automated feedback tailored to the individual. The primary outcome was weight loss over the 12-week program; secondary outcomes were health-related quality of life (HRQOL), use of weight control strategies, and CVD risk factors. Results: 92% of participants completed the study. Average weight losses in intent-to-treat analyses were significantly greater for WT LOSS than Control (4.4 ± 5.4 kg vs 1.0±3.3 kg, p=.02). On average, participants viewed 7 lessons and submitted their data on 8 of the 12 weeks; both measures of adherence were strongly related to weight loss (r=.61 and .63, p<.01). Participants in WT LOSS reported greater increases in the use of weight control strategies than Controls; moreover, 59% of WT LOSS versus 21% of Controls reported improvements in HRQOL (p<.05). There were no significant differences between WT LOSS and Control on changes in CVD risk factors. Conclusion: HIV-infected patients adhered to the behavioral weight loss program and, on average, lost 4.4 kg, which was similar to the outcomes previously reported using the same Internet program in non-HIV participants. HRQOL and use of healthy weight control strategies also improved. Thus, this population responded well to the program despite their low socioeconomic status (60% had income <$20,000), mental health comorbidities (67% had history of depression), and complex medical regimens (average 4.3 medications in addition to cART). This weight loss program is completely automated and can be easily disseminated. Further research on the efficacy of weight loss interventions for improving the health of HIV-infected patients is needed. 695 PREDICTORS OF SEVERE WEIGHT/BODY MASS INDEX GAIN FOLLOWING ANTIRETROVIRAL INITIATION Priya Bhagwat 1 , Igho Ofotokun 2 , Grace A. McComsey 3 , Todd Brown 4 , Carlee Moser 5 , Catherine A. Sugar 1 , Judith S. Currier 1 1 Univ of California Los Angeles, Los Angeles, CA, USA, 2 Emory Univ, Atlanta, GA, USA, 3 Case Western Reserve Univ, Cleveland, OH, USA, 4 The Johns Hopkins Univ, Baltimore, MD, USA, 5 Harvard Univ, Boston, MA, USA Background: Excessive weight gain following antiretroviral therapy (ART) is common and may predispose individuals to HIV-associated metabolic syndrome, sometimes leading to ART discontinuation and/or poor adherence. The objective of this study is to understand predictors of severe weight/body mass index (BMI) gain in individuals initiating ART. Methods: This was a retrospective analysis of the ACTG A5257 study, where ART-naïve HIV-infected individuals were randomized to one of 3 regimens: atazanavir/ritonavir (ATV/r), darunavir/ritonavir (DRV/r), or raltegravir (RAL) each in combination with tenofovir disoproxil fumarate/emtricitabine. Severe weight/BMI gain outcomes over 96 weeks were defined two ways: (1) percent weight increase ≥ 10%; (2) an upward change in BMI category. Among those underweight at baseline, only those who were overweight or higher at follow-up were included in both outcomes. Logistic regression was used to examine the association between participant characteristics and severe weight/BMI gain. Results: The study population (N=1,809) was 76%male, largely black non-Hispanic (41.9%) and white non-Hispanic (34.1%), with a mean baseline weight of 79 kg and BMI of 26 kg/m^2. Over 96 weeks, the average weight increased by 3.8 kg and BMI by 1.3 kg/m^2. Those with severe weight gain had a mean increase of 14.9 kg (N=373), and those with severe BMI gain had a mean increase of 4.4 kg/m^2 (N=361). The odds of severe weight gain were 1.55 times higher for black non-Hispanic compared to white non-Hispanic individuals (95% CI: 1.10 to 2.20; p=0.013). The odds of severe weight gain were 2.52 times higher for every 1 log (10-fold) higher in baseline HIV-1 RNA (95% CI: 2.00 to 3.16; p<0.0001), and 1.28 times higher for every 100 cell/mm^3 lower in baseline CD4+ count (95% CI: 1.18 to 1.39; p<0.0001). Results were similar for severe BMI. Results also suggested that treatment with protease inhibitors vs RAL may be protective against severe weight/BMI gain. The odds of severe weight gain were significantly lower for ATV/r vs

CROI 2017 304