icon-folder.gif   Conference Reports for NATAP  
  XV International AIDS Conference in Bangkok
July 11-16, 2004
Back grey_arrow_rt.gif
Changes in Body Fat on Reyataz & Sustiva
  Reported by Jules Levin
"Pattern of Adipose Tissue Redistribution After Initiation of Atazanavir-based Therapy: Analysis of 48-week Metabolic and Body Composition Data in Treatment-Naïve Patients"
Poster#: B5874
MA. Noor, J. Maa, MF Giordano, SL. Hodder
Bristol-Myers Squibb Company
This is poster presented at XV Intl AIDS Conference in Bangkok (July 2004) by BMS researchers. This study analysis is of the original study presented at IAS in Paris in 2003, the study report included mean changes in body fat, lipids, and insulin/glucose. This was the metabolic substudy of the pivotal phase III BMS study 034, which compared Reyataz (atazanavir) to efavirenz (Sustiva).
Summary of metabolic substudy regarding body changes: After 48 weeks on ATV or EFV there was no fat loss. After 48 weeks on therapy the DEXA results showed no loss of fat in the body compartments examined for either EFV or ATV treatment arms: appendicular, truncal, total fat. There were Increases in fat from baseline (mean percent) at week 48 were observed for appendicular (3%, atazanavir; 3%, efavirenz), truncal (5%, atazanavir; 8%, efavirenz), and total body fat (5%, atazanavir; 5%, efavirenz) as well as for visceral (40%, atazanavir; 29%, efavirenz), subcutaneous (19%, atazanavir; 5%, efavirenz), and total adipose tissue (23%, atazanavir; 11%, efavirenz). Similar results were seen, slight increases in fat, using CT scans for VAT, SAT, and TAT and there were no significant differences between treatment arms. Mean weight gain from baseline at week 48 was ATV 2 kg and EFV 0 kg. You can read a full report at this link, which reports on changes in lipids, insulin and glucose:
Atazanavir: Evaluation of Body Changes and Lipids/Glucose After 48 weeks Treatment
Background: Treatment of HIV with atazanavir is associated with fewer metabolic complications than other PIs, however, long-term data on body composition are not yet available. Data at 48 weeks demonstrate modest proportionate increase in all fat compartments in ATV treated patients. We describe a model to explain the relationship between metabolic changes and body composition and identify predictors of fat redistribution on ATV.
Methods: This was a post-hoc analysis from a randomized, multinational, double-blind, activecontrolled study (034) comparing ATV to efavirenz (EFV), each with lamivudine and zidovudine, in 810 antiretroviral-naïve patients. A metabolic sub-study assessed body composition using CT and DEXA. We used multivariate regression modeling to determine predictive variables.
Results: At 48 weeks, 169 of 211 patients had evaluable data. We excluded 6 patients with biochemical evidence of hyperglycemia from this analysis (n=163). Mean age was 31 years, 75% were male, mean body mass index (BMI) 23.7 Kg/m2, trunk fat (TF) 7.3 kg, appendicular fat (APF) 6.1 Kg, VAT 53 cm2 HOMA insulin resistance index 2.2.
Stratified according to baseline (BL) BMI, gain in APF, TF and VAT at 48 week were significantly higher in underweight individuals with BMI <22 when compared to overweight individual (BMI > 27 at BL).
There are two formats of the same chart immediately below. If you can't read either, this report will be posted on the NATAP website (http://www.natap.org). The chart shows mean changes in body fat by DEXA and CT for patients in the 034 Metabolic Substudy. You can see increases in fat in appendicular (limbs), truncal (belly), and total fat by DEXA for patients with <22 kg/m2 BMI (underweight) and 22-27 (normal weight) before the study. By DEXA patients who were overweight by BMI (>27 kg/m2) did not show mean increase in appendicular, truncal, & total fat, but did show increase in VAT by CT. This study examines and provides an explanation by the authors for these changes.
Mean Change in Fat by DEXA(kg) Mean Change in Fat Area by CT (cm2)
BMI (Kg/m2) -Appendicular Truncal Total Fat VAT
< 22 (underweight) 0.75 0.71 1.46 16.5
22-27 (normal weight) 0.27 0.53 0.81 11.9
> 27 (overweight) -1.28 -0.01 -1.29 26.6
< 3 0.26 0.41 0.67 12.0
>= 3 -1.12 -0.31 -1.43 13.3
Age (year)
>= 40 0.91 1.45 2.37 44.5
< 40 0.11 0.37 0.48 11.3

We identified baseline HOMA Insulin resistance index as significant predictor for the net increase in APF (p=0.002), TF(p=0.018), and total fat (p=0.001). BL BMI was a significant predictor for net increase in APF (p=0.030) and Age (>=40 years) was a significant predictor for net increase in VAT (p=0.004). Gender, BL fasting glucose, triglyceride, insulin, and drug (ATV vs. EFV) were not significant predictors.
Conclusions: The amount of fat gained on therapy with ATV was significantly higher in underweight individuals who tended to be relatively insulin sensitive at baseline. We conclude that the pattern of fat increase on ATV containing regimen in this population is consistent with return to health rather than worsening central adiposity.
Highly active antiretroviral treatment (HAART) of HIV has been associated with increases in cholesterol, triglycerides, glucose as well as with changes in body shape that include visceral fat accumulation and subcutaneous fat loss. Non-modifiable factors (age, sex, family history of dyslipidemia and diabetes) and modifiable factors (body weight and composition at baseline prior to initiation of a protease inhibitor (PI)-containing HAART) may modulate changes in body fat composition and distribution on therapy. Atazanavir is a potent once daily PI that has demonstrated comparable clinical efficacy to standard of care in naïve patients and when dosed as ATV/r, to LPV/r in treatment experienced patients. ATV is a PI that is associated with fewer metabolic complications, however, long term data on body composition are not yet available. We analyzed the relationship between patterns of changes in fat compartments and insulin sensitivity and body composition both at baseline and after 48 weeks of HAART treatment with either ATV or efavirenz (EFV)-containing regimens.
This post-hoc data analysis from a randomized, multinational, double-dummy, active-controlled study (BMS 424-034) comparing ATV to EFV, in combination with lamivudine (3TC) and zidovudine (ZDV), was performed in antiretroviral naïve patients. Dosing for each arm was: Atazanavir 400 mg QD or EFV 600 mg QD and 3TC + ZDV fixed combination. Body composition was measured objectively by dual-energy x-ray absorptiometry (DEXA) and computed tomography (CT) and included total, truncal, appendicular, visceral and subcutaneous compartments . Only individuals with available body composition data at baseline and 48 weeks were included in this post hoc analysis (n=163). Factors predictive of body composition changes were identified using multivariate logistic regression modeling on data from both arms and are depicted in all figures 1 to 3. Body mass index was defined as weight in Kg divided by height in meters squared. Homeostatis Model of Assessment-Insulin resistance (HOMA-IR) was calculated as fasting [glucose (mg/dL) x insulin (µU/ml)]/405.

034 n-805 subgroup n=211
Median age 33 30
Female gender 35% 26%
-white 33% 45%
-hispanic/lation 37% 49%
HIV RNA median 4.88 4.84
CD4 median 282 328

Mean Metabolic Parameters and Body Composition Data at Baseline and After 48 Weeks
This chart ispresented here in 2 formats & will be posted in aryicle on NATAP website. Mean fasting glucose, insulin, and HOMA-IR index did not increase for patients on ATV or EFV. LDL chol and TG did not increase for patients on ATV but did increase for patients on EFV. Body fat by DEXA did not showincreases in appendicular fat for EFV or ATV, but there was a trend (not significant) for increase in truncal fat. However, by CT there were increases in abdominal fat in visceral adipose tissue & subcutaneous fat for both EFV & ATV, and that is the subject of this poster.
ATV n=88 EFV n=75
Base wk 48 base wk 48 p-value
fasting glucose (mg/dL) 90 92 88 93 0.07
fasting insulin (uU/mL) 9.9 12.7 8.8 11 0.58
HOMA-IR index 2.4 3.1 1.9 2.6 0.69
Weight (kg) 67 89 69 69 0.02
BMI (kg/m2) 23 24 23 24 0.02
Total CHOL 157 160 151 181 0.01
-HDL 40 44 37 45 0.05
-LDL 89 90 89 104 <0.01
TG 154 135 125 149 0.09
Body fat ny DEXA (kg)
-appendicular fat 5.9 6.2 6.2 6.4 0.81
-truncal fat 7.2 7.7 7.4 8.1 0.34
Abdominal fat by CT (cm2)
-Visceral adipose tissue 52.8 73.5 53.5 68.7 0.86
-Subcutaneous adipose tissue 146 167 143 156 0.69

--A multivariate analysis identified body mass index (BMI), sensitivity to insulin (HOMA-IR), and age at time of initiation of ARV to be associated with changes in body fat compartments.
--Underweight individuals (BMI <22 kg/m2) gained significantly more weight than overweight individuals (BMI > 27 kg/m2) a pattern associated with improved survival in advanced HIV1.
--In contrast to insulin resistance and its associated central obesity, adipose tissue gain was higher in individuals with relative insulin sensitivity (HOMA-IR <=3) than in those with relative insulin resistance (HOMA-IR >3).
--Older age (> 40 years) was associated with increased visceral adipose tissue (VAT). Age related failure of lipo-regulation contributes to abnormalities in lipid metabolism including an increase in total body adipose content relative to total body weight (i.e. percent body fat).
--Consistent with physiological changes of senescence fat deposition in older individuals favored ectopic (visceral/perihepatic) compartments as opposed to the peripheral / subcutaneous compartments.
--Confounding factors such as smoking habits, treatment with lipid lower or insulin sensitizing agents, and dietary/lifestyle factors could not be completely controlled.
--The amount of adipose tissue gained on therapy with ATV or EFV-containing HAART was higher in individuals who were underweight and insulin-sensitive at baseline.
--Visceral adipose tissue deposition was favored in older individuals who gained weight, a finding consistent with normal physiological changes of senescence.
--Overall pattern of change in body composition in this population is consistent with a return to health rather than worsening central adiposity.
--Longer term data are needed to confirm these findings.