Framingham Risk and Lipoprotein Changes after 24 Weeks of Treatment with Abacavir/Lamivudine (ABC/3TC) and Raltegravir (RAL) in Antiretroviral-Naive HIV-1 Infected Subjects
Reported by Jules Levin
11th Adverse Events/Lipodystrophy Workshop
Oct 25-28 2009
B Young1, T Vanig2, E DeJesus3, T Hawkins4, L Yau5, B Ha5 and the SHIELD Study Team
1Rose Medical Center/Division of General Internal Medicine, University of Colorado, Denver, CO and Health Connections International, Amsterdam; 2Spectrum Medical Group, Phoenix, AZ;
3Orlando Immunology Center, Orlando, FL; 4Southwest CARE Center, Santa Fe, NM; 5GlaxoSmithKline, RTP, NC
The changes in fasting lipid parameters observed in this study were consistent with the results of previous trials. RAL has been shown to have a minimal impact on serum levels of total cholesterol, LDL cholesterol, or triglycerides, with modest
increases in HDL cholesterol (P004, P021).2,3 ABC/3TC has been associated with moderate increases in lipids (SHARE, HEAT),9,10 although the total cholesterol to HDL ratio remains largely unchanged.
This study is one of a few prospective trials to include extensive NMR lipoprotein analyses. The SMART Study11 included some of these analyses and showed a relationship between low HDL particle numbers and CVD risk in HIV-infected
patients. Based on the limited sample size in this study, we show an increase in HDL particle number and in LDL particle number through W24.
Inflammatory markers measured in this study decreased by 15% (hs-CRP) and 19% (IL-6) between baseline and W24 based on the ratio of geometric means. Neither decrease reached statistical significance, but the percentage of subjects
with hs-CRP <3 mg/L (indicating a low or average risk of CVD8) increased from 63% at baseline to 74% at W24.
There are limited prospective data on changes in inflammatory biomarkers soon after the initiation of antiretroviral therapy. A post-hoc analysis of hs-CRP and IL-6 after 48 weeks in the HEAT Study10 (ABC/3TC + lopinavir/r) showed decreases of similar magnitude.
Subjects in this study had no history of myocardial infarction but, at baseline, exhibited a number of known CVD risk factors including current/former smoker (43%), overweight/obese (37%), family history of heart attack/stroke (20%), and metabolic syndrome (17%). Of the 5 subjects with a "not low risk" Framingham score (10-year CVD risk ≥6%) at W24, 5 (100%) were male, 3 (60%) had blood pressure ≥140/90 mmHg, 2 (40%) were current smokers, and 1 (20%) had
metabolic syndrome, indicating the need to address modifiable CVD risk factors in people with HIV.
Limitations of this pilot study include its small sample size and single-arm design.
· The combination of ABC/3TC + RAL exhibited potent antiretroviral activity through W24.
· Moderate elevations in fasting lipid parameters were observed, although the total:HDL cholesterol ratio did not change.
· Tested W24 cardiovascular inflammatory biomarker levels were not significantly different from baseline.
Introduction - Cardiovascular disease (CVD) is an emerging concern in HIV-infected patients on antiretroviral therapy (ART). In this pilot study, we evaluated changes in metabolic profile after 24 weeks of treatment with ABC/3TC+RAL in ART-naïve HIV-1 infected subjects.
Methods - COL111429 is an ongoing 96-week, open-label, prospective, multicenter study evaluating ABC/3TC (600 mg/300 mg once daily) + RAL (400 mg twice daily) in HLA-B*5701-negative subjects with entry viral load >1,000 c/mL. We measured nuclear magnetic resonance (NMR)-derived lipoproteins and categorized subjects based on lipoprotein particle number (P) or size: "high risk" profile (LDL size
18-20.5 nm, large VLDL-P >5 nmol/L, LDL particle ≥1600 nmol/L, and small LDL-P >1200 nmol/L) and "not-at-high-risk" profile (LDL size >20.5 nm, large VLDL-P ≤5 nmol/L, LDL-P <1600 nmol/L, or small LDL-P ≤1200 nmol/L). Week 24 (W24) changes from baseline (BL) were assessed by median changes from BL and the associated 95% confidence interval derived from Hodges-Lehmann estimate.
Results - 35 subjects were enrolled. At BL, 37% were overweight or obese (BMI ≥25), 43% were smokers, and 20% had family history of CVD. Most subjects had low (<6%) Framingham risk scores (94%), "not-at-high-risk" profile (100%), and did not have metabolic syndrome (89%). The median BL HIV-1 RNA was 4.80 log10 c/mL and median BL CD4 cell count was 301 cells/mm3 (20% <200 cells/mm3). At W24, 94% had VL <50 c/mL and 97% had VL <400 c/mL by missing-equals-failure analysis.
*Increase from BL; 1n=34; 2n=25; data not collected for n=5 at BL and n=5 at W24; 3LDL-P <1300 nmol/L and LDL size 20.6-23 nm are low-moderate risk; 4Small LDL-P 600-1200 nmol/L and large VLDL-P 0.5-5 nmol/L are moderate risk
Most subjects maintained the BL risk profile but 3 shifted to "high risk" profile at W24 due to increased LDL and LDL sub-fraction particle numbers. Of these, 2 subjects with borderline/high lipids at BL also met the criteria for metabolic syndrome at W24.
Conclusion - In this pilot study, ABC/3TC+RAL was associated with rapid virologic suppression, with most subjects maintaining BL CV risk status through W24. While TC/HDL ratios were largely unchanged, NMR lipoprofile analysis based on the limited sample size suggested that lipid changes may be mediated through increases in LDL particle numbers. These data support the generally favorable lipid profile of ABC/3TC+RAL, but reinforce the need to address modifiable CV risk factors along with virologic suppression.
As novel agents become available, it is important to evaluate these agents with existing antiretrovirals for potency, tolerability, and treatment-emergent resistance patterns. The combination of abacavir sulfate 600 mg + lamivudine 300 mg (ABC/3TC [EPZICOM®], GlaxoSmithKline, RTP, NC) and raltegravir (RAL [Isentress®] Merck & Co., Inc., Whitehouse Station, NJ) has not been studied previously.
Recent pharmacogenetic data strongly suggest that the safety profile of ABC-based treatments can be improved with the use of HLA-B*5701 screening.1 Studies to date appear to show that RAL has a minimal impact on fasting lipids.2,3 This study provides an opportunity to evaluate the impact of this
new combination regimen on lipid profiles and cardiovascular risk profiles in a cohort over 96 weeks. In addition to standard lipid panel testing, nuclear magnetic resonance (NMR) lipoprotein subclass analyses will be performed to provide data on LDL particle number and size of LDL particles, as well as
direct measurement of HDL and VLDL subclasses. A number of studies have shown that NMR-derived LDL particle number is a strong independent predictor of cardiovascular risk.4,5
The SHIELD study (COL111429) is an ongoing 96-week, open-label, prospective, multicenter trial evaluating ABC/3TC (600 mg/300 mg once daily) + RAL (400 mg twice daily) in antiretroviral-naïve HLA-B*5701 -negative adults with screening HIV-1 RNA >1,000 c/mL. Subjects were excluded if they had an active or acute CDC Clinical Category C event at screening, were HBSAg+, or if their screening HIV-1 genotype indicated the presence of any of the following mutations: K65R, L74V, and Y115F or ≥2 thymidine analog mutations (M41L, D67N, K70R, K219Q or E) that included changes at either L210 or T215, associated with ABC and 3TC resistance, and mutations Q148H/R/K and N155H associated with RAL resistance. There were no restrictions on baseline CD4 cell count.
Virologic failure is defined as having either confirmed virologic non-response (HIV-1 RNA >400 c/mL at week 24 [W24]) or confirmed virologic rebound (HIV-1 RNA ≥400 c/mL after an initial response of HIV-1 RNA <400 c/mL or >1 log10 c/mL increase above nadir). In the case of clinically suspected hypersensitivity to ABC, subjects are permitted to substitute ABC/3TC with lamivudine 150 mg and
zidovudine 300 mg (ZDV/3TC [COMBIVIR®], GlaxoSmithKline, RTP, NC) twice daily and will not be discontinued from the study. This poster presents the results of a planned 24-week interim analysis.
Thirty-five subjects enrolled in the study. At baseline, 37% were overweight or obese (BMI ≥25), 43% were current or former tobacco smokers, and 20% had a family history of cardiovascular disease (CVD). Most subjects had a low risk of CVD based on Framingham score (94%), a "not high risk" of CVD based on NMR lipoprotein analysis (100%), and did not have metabolic syndrome (89%).
* Subjects were classified as having metabolic syndrome6 when 3 or more of
the following were present:
(1) waist circumference >102 cm in male or >88 cm in female
(2) triglycerides >150 mg/dL
(3) HDL-C <40 mg/dL in male or <50 mg/dL in female
(4) systolic BP ≥130 mmHg, diastolic BP ≥85 mmHg, or use of
(5) fasting glucose ≥100 mg/dL
High Risk7 = LDL size 18 to 20.5 nm, Large VLDL-P >5 nmol/L, total LDL-P ≥1600 nmol/L, small LDL-P >1200 nmol/L
Subject Accountability: Thirty four of 35 subjects completed W24 of the study. The status of 1 subject was unknown at the time of the data cut, but the subject was later determined to be lost-to-follow-up.
Efficacy Results: The primary efficacy endpoint of this study is the percentage of subjects with HIV-1 RNA <50 c/mL at W48. For this interim analysis at W24, 94% (33/35) of subjects had HIV-1 RNA <50 c/mL and 97% (34/35) had HIV-1 RNA <400 c/mL using an ITT, missing-equals-failure analysis. No subject met the criteria for virologic failure. At W24, the median increase from baseline in CD4 cells count was 193 cells/mm3.
Safety Results: The primary safety endpoint is the percentage of subjects with treatment-related grade 3 or 4 adverse events (AEs) and laboratory abnormalities. Through W24, 11% (4/35) of subjects experienced treatment-related grade 2 to 4 AEs; no subjects experienced treatment-related grade 3/4 AEs. 17% (6/35) of subjects experienced grade 3/4 laboratory abnormalities, including increased cholesterol, triglycerides, creatine kinase, glucose, AST, lipase, and organic phosphorus. One subject had uncontrolled diabetes at baseline. There were no suspected hypersensitivity reactions to abacavir.
The secondary safety endpoints include changes from baseline in fasting lipids, NMR lipoproteins, and inflammatory biomarkers. In each box of Figures 1 and 2, the solid orange bar spans Q1 to Q3, the lighter circle within the solid bar is the median, the lines indicate the minimum and maximum, and the circles are outliers in the data. The boxes in Figure 3 show the geometric mean and 95% confidence interval.
Of the 2 subjects with Framingham scores ≥6% at baseline (not low risk), 1 was in the same category at W24. The other person had an increase in HDL from 35 to 66 mg/dL, lowering his risk. The four additional people with Framingham
scores ≥6% at W24 had increases in lipids and/or blood pressure during the study.
The 3 subjects with high CVD risk at W24 based on NMR lipoproteins all had significant increases in total cholesterol and triglycerides during the study.
Of the 6 subjects with metabolic syndrome at baseline, one subject did not have data at W24 (LTFU), 4 subjects still had metabolic syndrome at W24, and 1 subject no longer had metabolic syndrome at W24 because his systolic blood pressure decreased from 142 mmHg at baseline to 123 mmHg at W24. There was 1 subject with borderline high lipids at baseline who subsequently developed metabolic syndrome at W24 due to increases in lipids, glucose, and blood pressure.
1. Mallal S, Phillips E, Carosi G, et al. NEJM. 2008;358:568-579.
2. Markowitz M, Nguyen BY, Gotuzzo E, et al. J Acquir Immune Defic
3. Isentress® (Raltegravir) [package insert]. Merck & Co., Inc., NJ.
4. Otvos JD, Jeyarajah EJ, Cromwell WC. Am J Cardiol . 2002;90:22i-29i.
5. Cromwell WC, Otvos JD. Curr Atheroscler Rep. 2004;6:381-387.
6. Grundy SM, Cleeman JI, Bairey Merz CN, et al. Circulation.
7. Jeyarajah EJ, Cromwell WC, Otvos JD. Clin Lab Med. 2006;26:847-870.
8. Pearson TA, Mensah GA, Alexander RW, et al. Circulation.
9. Elion R, DeJesus E, Sension M, et al. HIV Clin Trials . 2008;9:152-163.
10. Smith KY, Patel P, Fine D, et al. AIDS. 2009;23:1547-1556.
11. Duprez DA, Kuller LH, Tracy R, et al. Atherosclerosis . May 13, 2009
[epub ahead of print].