Is abacavir (ABC)-containing combination antiretroviral therapy

Is abacavir (ABC)-containing combination antiretroviral therapy (CART) associated with myocardial infarction (MI)? No association identified in pooled summary of 54 clinical trials A Cutrell, J Hernandez, C Brothers, J Yeo, W Burkle, W Spreen. XVII International AIDS Conference, 3-8 August 2008, Mexico City Cutrell A, Hernandez J, Brothers C et al. 2008 Background and Objective • The D:A:D study group recently reported an unexpected increase in risk of MI with ABC and ddI in a large, prospective observational cohort (Lancet 2008; 371:1417-26). • To determine whether there is an increase in cardiovascular events associated with ABC containing CART as compared to non-ABC containing CART using GSK-sponsored clinical trials contained in the GSK HIV Data Repository. Cutrell A, Hernandez J, Brothers C et al. 2008 Methods (1) • GSK HIV Data Repository – history/purpose • GSK-sponsored clinical trials with > 24 weeks of follow up with data from 1995-2006 were analyzed. – Data from 54 clinical trials analyzed: • 13/54 adult trials were randomized for ABC vs control • 33/54 trials included ABC in background ART • 8/54 trials did not include ABC • Includes data from 14683 subjects (14174 adults and 509 children) • Descriptive statistics were summarized for naïve and experienced subjects treated with and without ABC. Cutrell A, Hernandez J, Brothers C et al. 2008 Methods (2) • To capture events of interest (myocardial ischemia or infarction) all data were queried for events coded as “Coronary Artery Disorders” and “Ischemic Coronary Artery Disorders” using MeDRA – Specific preferred terms included: coronary artery atherosclerosis, coronary artery disease, coronary artery occlusion, acute MI, angina pectoris, angina unstable, MI and myocardial ischemia. Cutrell A, Hernandez J, Brothers C et al. 2008 Methods (3) • All fatal cases (any cause) in the HIV Data Repository were reviewed by an external cardiologist. • Framingham risk not calculated at baseline – not all factors routinely collected – Smoking, hypertension, HDL levels • Rates per 1000 person/years were calculated, and Poisson regression models were used to calculate unadjusted relative rates and 95% confidence intervals using SAS. Cutrell A, Hernandez J, Brothers C et al. 2008 Demographic and HIV-disease Characteristics at Baseline ADULT NAIVE ABC N=5859 Age at Screening, n Median, year (min-max) Race, n White Black Sex, n Male Viral Load (log10), n Median (Min-Max) CD4 Count, n Median (Min-Max) 5859 36 (17-78) 5728 49% 29% 5858 78% 5793 4.81 (1.287.64) 5819 273 (0-1883) No ABC N=2406 2406 35 (18-74) 2314 54% 22% 2406 79% 2397 4.67 (1.28-6.6) 2393 337 (0-1729) ADULT EXPERIENCED ABC N=3643 3639 40 (18-78) 3591 64% 21% 3643 86% 3626 4.25 (0.84-6.74) 3629 265 (0-2089) No ABC N=2266 2263 39 (18-74) 2265 73% 16% 2266 85% 2247 4.18 (1.38-7.05) 2250 295 (0-1799) Cutrell A, Hernandez J, Brothers C et al. 2008 Baseline Lipids and Glucose (mmol/L) n Median (Min-Max) Cholesterol ADULT NAIVE ABC N=5859 5484 4.1 (1.0-10.1) 2143 2.5 (0.1-6.3) 2215 0.9 (0.05-3.1) 5482 1.4 (0.3-28.9) No ABC N=2406 2272 4.2 (1.6-10.1) 166 2.7 (0.8-6.4) 172 0.9 (0.2-2.5) 2272 1.4 (0.2-15.4) ADULT EXPERIENCED ABC N=3643 2249 4.8 (0.05-15.0) 39 3.2 (1.0-5.5) 48 1.0 (0.3-2.6) 2234 2.1 (0.04-38.1) No ABC N=2266 1808 4.6 (1.8-15.0) 23 3.7 (1.3-5.2) 24 1.1 (0.5-1.5) 1803 1.9 (0.4-35.3) LDL HDL Triglycerides Glucose 4999 5.0 (1.3-27.2) 2353 5.0 (0.9-24.1) 2201 5.2 (2.2-26.0) 1662 5.2 (2.1-26.2) Cutrell A, Hernandez J, Brothers C et al. 2008 CV Outcomes - Exposure to ABC Compared with No Exposure to ABC Exposure to ABC Events /Patients Frequency Events /PersonYears Rate /1000 PersonYears Relative rate (95% CI) Any Myocardial Infarction or Acute Myocardial Infarction: None ABC CART 7/5044 11/9639 0.139% 0.114% 11/4653 16/7845 2.36 2.04 0.863 (0.40,1.86) Any ischemic Coronary Artery Disease or Disorder: None ABC CART 21/5044 24/9639 0.416% 0.249% 27/4642 27/7832 5.82 3.45 0.593 (0.35 ,1.01) Cutrell A, Hernandez J, Brothers C et al. 2008 CV Outcomes - Exposure to ABC or No ABC (12 Adult Randomized Trials) Exposure to ABC Events /Patients Frequency Events /PersonYears Rate /1000 PersonYears Relative rate (95% CI) Any Myocardial Infarction or Acute Myocardial Infarction: None ABC CART 6/1692 2/1570 0.355% 0.127% 7/1706 4/1863 4.10 2.15 0.523 (0.15 – 1.79) Any ischemic Coronary Artery Disease or Disorder: None ABC CART 13/1692 5/1570 0.768% 0.318% 13/1702 8/1860 7.64 4.30 0.563 (0.23-1.36) Cutrell A, Hernandez J, Brothers C et al. 2008 Strengths and Limitations • Prospective clinical trial data with large number of patients exposed to ABC, consistent data collection • Standard and real-time monitoring and follow up of all reported AEs • External expert review supportive • Post-hoc analysis • Duration of exposure on average 24-48 weeks • Studies not designed to detect CV risk specifically • Low event rate therefore less power Cutrell A, Hernandez J, Brothers C et al. 2008 Conclusion • Overall, among pts in the GSK HIV Data Repository: – the incidence rate of MI and Coronary Artery Disorders was low – there was no difference in incidence of ischemic coronary artery events or myocardial infarction in subjects who received abacavir-containing vs. nonABC containing CART • Further study is needed to fully evaluate the association of ABC and cardiovascular disease. Cutrell A, Hernandez J, Brothers C et al. 2008 What’s Next to Further Address the Question?  Analysis of Internal Clinical Trial Data  Continued vigilance of spontaneous event data and published literature  In all future studies, incorporate cardiovascular risk factor and biomarker collection  Mechanism research  Additional cohort collaborations  Analyze inflammatory markers in current and future studies Cutrell A, Hernandez J, Brothers C et al. 2008 HEAT Biomarker Analysis: Geometric Mean (95% CI) by Study Week (ITT-E) 3 2.5 Interleukin 6 (IL-6) 3 2.5 High Sensitivity C-Reactive Protein (hsCRP) IL-6 (pg/mL) 2 1.5 1 0.5 0 0 48 96 hsCRP (mg/L) 2 1.5 1 0.5 0 0 256 252 48 96 232 216 n (obs) Epzicom Truvada Study Week = 243 = 249 248 235 230 216 Study Week 251 236 Poster LBPE1138 Thursday August 7, 2008 Cutrell A, Hernandez J, Brothers C et al. 2008 Acknowledgments The authors wish to thank… • Harry Staines, Suki Pabla, Dupe Bassey, Lee Tombs and Pinal Patel at SRG Interesource Satellite Operations for statistical support • Drs. Peter Kowey & Gary Koch for expert input • Investigators who supported the clinical trials • All of the patient volunteers. Cutrell A, Hernandez J, Brothers C et al. 2008