CROI 2017 Abstract e-Book

Abstract eBook

Poster and Themed Discussion Abstracts

compared, in COVERTE and ENNS, separately for men and women, with directed standardization for overweight and educational level, and logistic regression to estimate odds ratios (ORs) and their 95% confidence intervals (CI) adjusted for educational level, overweight and alcohol consumption. Results: Most of the 269 HIV-infected patients included in COVERTE (47%male) had been on combined ARV therapy (40% including a protease inhibitor) for a mean of 19 years. Viral load remained detectable in 32%, but median CD4 count was 550 cells/mL. These subjects were compared with 245 young adults from ENNS (42%male). Tobacco use was similar in the two populations. After standardization, the prevalence of metabolic syndrome was 13.2% (CI: 7.1-19.2) in HIV-infected men versus 10.6% (CI: 1.5-19.7) in the general population. The corresponding prevalences in women were 9.8% (CI: 5.1-14.6) and 1.7% (CI: 0-4.1). HIV-infected men were more likely to have high triglyceride levels (OR 4.97; CI: 1.02-24.22) though less likely to be overweight (OR 0.23; 0.08-0.65) or have hyperglycemia (OR 0.25; CI: 0.07-0.86) than men from the general population. HIV-infected women were more likely to have a large waist circumference (OR: 3.47; CI: 1.14-10.55), low HDL-cholesterol levels (OR 2.68; CI: 1.30-5.50) and less likely to have high total cholesterol levels (OR 0.31; 0.14-0.66). Conclusion: Young adults infected with HIV since childhood had no excess of behavioral risk factors, but were more likely to have dyslipidemia and metabolic syndrome than the general population, justifying close monitoring for cardiometabolic diseases. 822 EFFECTS OF VITAMIN D SUPPLEMENTATION ON CAROTID INTIMA-MEDIA THICKNESS IN HIV+ YOUTH Allison R. Eckard 1 , Paolo Raggi 2 , Mary Ann O’Riordan 3 , Julia C. Rosebush 4 , Danielle Labbato 3 , Ann Chahroudi 4 , Joshua H. Ruff 4 , Chris T. Longenecker 3 , Vin Tangpricha 4 , Grace A. McComsey 3 1 Med Univ of South Carolina, Charleston, SC, USA, 2 Mazankowski Alberta Heart Inst, Edmonton, Canada, 3 Case Western Reserve Univ, Cleveland, OH, USA, 4 Emory Univ, Atlanta, GA, USA Background: HIV+ youth are at an increased risk of cardiovascular disease (CVD). Vitamin D deficiency is associated with CVD risk in HIV, but it is not known whether supplementation could affect this risk. Methods: This is a 24-month randomized, active-control, double-blind trial comparing 2 different monthly vitamin D3 doses [60,000 (medium) or 120,000 (high) IU/month] vs. control of 18,000 IU/month in 8-26 year old HIV+ youth on ART with baseline 25-hydroxyvitamin D (25(OH)D) ≤30 ng/mL & HIV-1 RNA <1000 copies/mL. Carotid IMT was measured at baseline & 24 months. Comparisons of changes in IMT were made between the HIV+ control arm vs. combined supplementation arms (medium+high) & within these groups using appropriate two-sample tests. A matched healthy uninfected group was enrolled in a similarly-designed parallel study for comparison. Results: We enrolled 102 HIV+ subjects: 64%men, 89% black, median age of 20 years. HIV & ART duration were 8 & 3 years, respectively with a CD4 count of 652 cells/mm³. Baseline 25(OH)D was similar between groups (controls: 17 (11, 21) vs. supplementation group: 18 (14, 22) ng/mL; P=0.49) and increased to 32 (25, 38) and 41 (31, 46) ng/mL in the control and supplementation (medium+high) dose groups at 24 months, respectively (within & between groups P<0.001). Baseline bulb (0.65 vs. 0.63 mm, P=0.13) and common carotid artery (CCA) IMT (0.69 vs. 0.56 mm, P=0.81) were similar between groups. Over 24 months, bulb & CCA IMT decreased only in the control arm (Figure 1), with changes in bulb IMT being significantly different than supplementation arm at 24 months (P=0.02). Overall, changes in bulb IMT were significantly correlated with changes in 25(OH)D (R=0.43, P=0.001). In multivariable regression models, larger increases in 25(OH)D were associated with greater IMT increases. In contrast to the findings in HIV+ subjects, among the healthy uninfected group (N=88), there were no differences in changes in IMT between the control vs. supplementation arms and no significant correlations between changes in 25(OH)D and changes in IMT. Conclusion: A modest vitamin D3 dose of 18,000 IU/month given over 24 months resulted in significant decreases in carotid IMT compared to high monthly doses of 60,000 or 120,000 IU. These results suggest a potential for a less optimal effect of high-dose vitamin D supplementation in this population, an effect that was not seen in the parallel HIV- uninfected study. On-going analyses are underway to better understand these surprising findings. 823 DOES PREMATURE VASCULAR STIFFNESS SLOWLY IMPROVE FOLLOWING EARLY ART?DATA FROM CHER Steve Innes 1 , Mark Cotton 1 , Kennedy N. Otwombe 2 , Barbara Laughton 1 , Sara H Browne 3 1 Stellenbosch Univ, Cape Town, South Africa, 2 Univ of the Witwatersrand, Johannesburg, South Africa, 3 Univ of California San Diego, San Diego, CA, USA Background: ➢ Cross-sectional data suggests increased prevalence of vascular disease in HIV+ children on antiretroviral therapy (ART) after adjusting for traditional atherosclerosis risk factors, typically associated with advanced HIV disease and ART, particularly lopinavir/ritonavir (LPVr) ➢ Thus far, pediatric studies have focused on children initiating ART much later than 3 months of age ➢ Whether very early ART will prevent HIV-related premature vascular disease is unknown ➢ Aorto-femoral pulse wave velocity (PWV) is a sophisticated and sensitive measure of elevated arterial wall stiffness, typically due to atherosclerosis or subclinical arteritis ➢ Reduced arterial wall elasticity leads to faster propagation of the arterial pulse wave ➢ PWV elevations strongly predict subsequent cardiovascular events in asymptomatic adults Methods: ➢ Baseline and 1-year follow-up PWV measurements were performed in perinatally-HIV-infected primary-school-age children who initiated LPVr + zidovudine + lamivudine very early in infancy with minimal HIV disease and normal CD4 counts in a well-resourced trial setting (CHER); and in HIV-uninfected controls (HIV-exposed uninfected, HEU, and HIV-unexposed, HU) from the same communities and socio-economic background ➢ Changes in raw PWV, height-based PWV Z-score (PWVZ-ht) and age-based PWV Z-score (PWVZ-age) were compared by ANOVA followed by pairwise T-test, and adjusted, using multivariable regression, for body mass index, fasted glucose, total and low density lipoprotein cholesterol, triglycerides and serum cotinine Results: ➢ 84 HIV+ (median age 7.7 [IQR: 7.6-8.5] years) who initiated ART at median 9 (7–12) weeks of age, with cumulative time on ART of median 7.1 (6.7–7.5) years and normal CD4 counts ➢ 51 uninfected (31 HEU; 20 HU) of median age 8.5 (IQR: 7.8-8.7) years, with similar anthropometric Z-scores between groups(p>0.10) ➢ Baseline PWV metrics in both HIV+ and HEU were higher than HU and this difference persisted in HIV+ after adjustment (p≤0.04 for all). (HEU could not be adjusted due to limited n) ➢ At follow-up, both PWVZ-ht and PWVZ-age had improved in HIV+ and HEU (p≤0.03) (see figure 1), whereas HU remained unchanged. This difference did not persist after adjustment, however the power of the adjustment was restricted by limited n Conclusion: ➢ Suggests that, in children who initiated ART very early in life, PWV abnormalities gradually improve with accumulating time on ART ➢ Interestingly, early-onset PWV elevations in HEU also appear to wane with time 824 SVCAM AND MCP-1 INVOLVED IN PREMATURE ARTERIAL-WALL STIFFNESS IN PREPUBERTAL CHILDREN Background: ➢ Cross-sectional evidence strongly suggests increased prevalence of vascular disease in HIV+ children on antiretroviral therapy (ART) after adjusting for traditional atherosclerosis risk factors. ➢ Soluble vascular cell adhesion molecule-1 (sVCAM-1), soluble intercellular adhesion molecule-1 (sICAM-1) and monocyte chemoattractant protein-1 (MCP-1) predict incident vascular events in adults. ➢ Whether these chemokines indicate clinically-detectable vascular disease in children is unknown. ➢ Aorto-femoral pulse wave velocity (PWV) is a sophisticated and sensitive measure of elevated arterial wall stiffness, typically due to atherosclerosis or subclinical arteritis. ➢ Reduced arterial wall elasticity leads to faster propagation of the arterial pulse wave. ➢ Early PWV elevations strongly predict subsequent cardiovascular events in asymptomatic adults. Methods: ➢ Baseline and 1-year follow-up measurements of serum sVCAM-1, sICAM-1 and MCP-1 (by Luminex Multiplex® assays), and PWV (using Vicorder® device) in HIV+ primary-school-age children who initiated lopinavir/ritonavir + zidovudine + lamivudine very early in infancy with minimal HIV disease and normal CD4 counts in a well-resourced trial setting (CHER); and in HIV-uninfected controls (HIV-exposed uninfected, HEU, and HIV-unexposed, HU) from the same communities and socio-economic background. ➢ To avoid post-hoc analysis bias and the multiple comparisons problem, the choice of biomarkers and adjustment factors was set a priori. ➢ Changes in sVCAM-1, sICAM-1 and MCP-1 were compared with changes in raw PWV, height-based PWV Z-scores (PWVZ-ht) and age-based PWV Z-scores (PWVZ-age) by multivariable linear regression adjusting for body mass index, fasted glucose, total and low density lipoprotein cholesterol, triglycerides and serum cotinine. Steve Innes 1 , Richard Glashoff 1 , Shalena Naidoo 1 , Kennedy N. Otwombe 2 , Barbara Laughton 1 , Maile Karris 3 , Mark Cotton 1 , Sara H Browne 3 1 Stellenbosch Univ, Cape Town, South Africa, 2 Univ of the Witwatersrand, Johannesburg, South Africa, 3 Univ of California San Diego, San Diego, CA

Poster and Themed Discussion Abstracts

CROI 2017 356