Thrombolysis in Myocardial Infarction (TIMI) Study Group Institute of Medicine Workshop: Transforming Clinical Research in the United States October 7, 2009 Marc S. Sabatine, MD, MPH Investigator, TIMI Study Group Associate Physician, Cardiovascular Division, BWH Assistant Professor of Medicine, Harvard Medical School
Disclosure Statement Astra Zeneca (research grant, scientific advisory boards) Bristol-Myers-Squibb (honoraria, scientific advisory boards) Eli Lilly & Daiichi-Sankyo (honoraria) Eisai (research grant) sanofi-aventis (research grant, honoraria, scientific advisory boards) Schering-Plough (research grant support) The TIMI Study Group received research grant support from: Accumetrics, Amgen, AstraZeneca, Bayer Healthcare, Beckman Coulter, Biosite, Bristol-Myers Squibb, CV Therapeutics, Eli Lilly and Company, GlaxoSmithKline, Inotek Pharmaceuticals, Integrated Therapeutics, Merck & Co., Millennium Pharmaceuticals, Novartis Pharmaceuticals, Nuvelo, Ortho-Clinical Diagnostics, Pfizer, Roche Diagnostics, sanofiaventis, Schering-Plough Research Institute.
Description The Thrombolysis in Myocardial Infarction (TIMI) Study Group is an Academic Research Organization (ARO) devoted to conducting clinical trials to improve outcomes in patients with cardiovascular disease. The group is headquartered at Brigham and Women s Hospital & Harvard Medical School Boston, MA 3
Trials to Date Phase I to Phase IV studies Initial trials NHLBI-funded, now trials industry-funded NIH grant support for ancillary studies 30 to 25,000 subjects Trial Status 45 completed trials 6 ongoing trials 7 trials in various stages of planning 4
Current Staff Physician Staff Eugene Braunwald, MD Study Chairman 12 Staff Cardiologists 3 Senior Cardiology Fellows (2-year research fellowships) Rotating BWH Research Residents Operational Staff Carolyn McCabe, BS Director 8 Project Directors & Managers Research Assistants, etc. Biostatistical Core Charles Contant, PhD Director Programmers 5
Services Academic Leadership Project Management Biostatistics Site Management Medical Hotline Regulatory Services Safety Reporting Clinical Events Committee 6
Academic Leadership (1) Review the Compound Pharmacokinetics & Pharmacodynamics Animal & Phase I Data Refine Scientific Question Unmet clinical need Potential utility of compound Current and evolving concomitant treatments Initiation of study Investigator-initiated Industry-initiated 7
Clopidogrel in STEMI Occluded Artery or Death/MI (%) 25 20 15 10 5 0 Angiographic 36% Odds Reduction P<0.0001 15.0 Clopidogrel 21.7 n=1752 n=1739 Placebo CV Death, MI, Urgent Revasc (%) 0 5 10 15 Placebo Sabatine MS et al. NEJM 2005;352:1179 Clinical 0 5 10 15 20 25 30 days Clopidogrel Odds Ratio 0.80 (95% CI 0.65-0.97) 0.97) P=0.026 20%
Clopidogrel Response Variability and Increased Risk of Ischemic Events Primary PCI for STEMI (N = 60) 5 µm ADP-induced Platelet Aggregation Death/ACS/CVA by 6 mo Baseline (%) 120 100 80 60 40 20 0 1 2 3 4 5 6 Days Clop resist 40 40 Q1 Q2 Q3 Q4 Quartiles of response Percent 30 20 10 0 6.7 P = 0.007 0 0 Q1 Q2 Q3 Q4 Matetzky S et al. Circulation. 2004;109:3171-3175. Wiviott SD, Antman EM. Circulation. 2004;109:3064-3067.
Prasugrel: more potent and predictable platelet inhibition 100.0 Inhibition of Platelet Aggregation (%) 80.0 60.0 40.0 20.0 0.0-20.0 Interpatient Variability Response to Clopidogrel Clopidogrel Responder Clopidogrel Non-responder *Responder = 25% IPA at 4 and 24 h Interpatient Variability Response to Prasugrel Brandt JT et al. Am Heart J. 2007;153:66.e9-e16.
Wiviott SD Circulation 2005 Prasugrel: Phase II Efficacy RR=0.72 [0.4,1.2] P = 0.23 P= NS MI at 30 days 10.0% 8.0% 6.0% 4.0% 7.9% 5.7% 7.0% 6.5% 4.0% 2.0% R/N 0.0% Clop Pras 40/7.5 60/10 60/15 Treatment Group Prasugrel LD/MD 20/254 37/650 14/199 13/200 10/251 11
13,608 Patients with ACS and Planned PCI Randomized to Prasugrel (60/10) vs. Clopidogrel (300/75) Endpoint (%) 15 10 5 7/2009 0 CV Death / MI / Stroke TIMI Major Non-CABG Bleeds Clopidogrel Prasugrel 0 30 60 90 180 270 360 450 Days Prasugrel Clopidogrel 12.1 9.9 2.4 1.8 HR 0.81 (0.73-0.90) P=0.0004 HR 1.32 (1.03-1.68) P=0.03 Wiviott SD et al. NEJM 2007;357:2001-15
PROVE IT TIMI 22 Death or Major CV Events (%) 30 25 20 15 10 5 0 Pravastatin 40 mg (avg achieved LDL = 95 mg/dl) Atorvastatin 80 mg (avg achieved LDL = 62 mg/dl) Circulation 2004;110:227-239 16% RR (P = 0.005) 0 3 6 9 12 15 18 21 24 27 30 Months of Follow-up Cannon et al. NEJM 2003; 350: 1495
Academic Leadership (2) Develop Study Design Study population Timing of intervention Control arm Background therapy Endpoint(s) and timing of ascertainment Statistical analysis plan Develop Key Trial Documents Protocol Case Report Form Clinical Events Committee (CEC) Charter Data Safety and Monitoring Board (DSMB) Charter 14
Academic Leadership (3) Study Startup Country & site selection Applicability to U.S. Acceptability to other countries Cost Steering Committee of National Lead Investigators Investigator Training 15
U.S. Enrollment Population STEMI w/ lytic STEMI w/ lytic NSTE-ACS Post ACS STEMI w/ lytic STEMI w/ lytic NSTE-ACS ACS NSTE-ACS Post ACS N 3002 15,078 2220 4162 20,506 3491 6560 13,608 3241 3491 Intervent. Enrollment from U.S. (%) 90 80 70 60 50 40 30 20 10 0 Hirudin vs. Heparin 59 npa vs. TPA 13 INV vs. CONS 83 Atorva vs. Prava 71 ENOX vs. UFH 1 Clopi vs. pbo 9B 17 18 22 25 28 36 38 42 46 4 Ranolaz. vs. pbo 11 Prasugrel vs. clopi 30 Phase II: OTAM vs. UFH 5 Phase II: Rivarox. vs. pbo 12 TIMI Study Group Trial
K-M estimate of time to first primary efficacy event (composite of CV death, MI or stroke) Cumulative incidence (%) 13 12 11 10 9 8 7 6 5 4 3 2 1 0 Clopidogrel Ticagrelor HR 0.84 (95% CI 0.77 0.92), p=0.0003 11.7 9.8 0 60 120 180 240 300 360 No. at risk Days after randomisation Ticagrelor 9,333 8,628 8,460 8,219 6,743 5,161 4,147 Clopidogrel 9,291 8,521 8,362 8,124 6,743 5,096 4,047 K-M = Kaplan-Meier; HR = hazard ratio; CI = confidence interval Wallentin L et al. NEJM 2009;361:1045-57.
Academic Leadership (4) Monitor Study Progress Enrollment Changes in medical landscape Aggregate event rates (efficacy & safety) Retention Lead Study Analysis Data analysis Separate copy of the database Rapidly move to presentation of data at scientific meeting Drafting of primary manuscript and subsequent analyses 18
PPI Use & Thienopyridines PPI use at randomization (n= 4529) Clopidogrel CV death, MI or stroke Prasugrel CLOPIDOGREL PPI vs no PPI: Adj HR 0.94, 95% CI 0.80-1.11 PRASUGREL PPI vs no PPI: Adj HR 1.00, 95% CI 0.84-1.20 Days O Donoghue M et al. Lancet 2009.
Clinical Events Committee Physician Staff Stephen D. Wiviott, MD Director Board-certified cardiologists, neurologists, oncologists, endocrinologists, hepatologists Seamless integration with ecrf Adjudicate Key Endpoints Efficacy & Safety, including non-cardiovascular events 2 Independent reviewers with 3 rd as needed Consistent application of definitions High-level of granularity 20
Universal MI Classification Efficacy Analysis by Universal MI Classification Cumulative Incidence (%) 8 7 6 5 4 3.4 3 2 1 0 CLOPIDOGREL PRASUGREL 29% 2.5 18% 0.4 0.3 0.1 24% 6.4 Type 1 Type 2 Type 3 Type 4 Type 5 Spont. Secondary SCD PCI-related Peri-CABG p 0.0015 0.53 -- <0.0001 -- 4.8 4a 4a 1.7 0.8 0 4b 4b 0 0.1
Core Labs Biomarker Laboratory David A. Morrow, MD, MPH - Director 2000 sq ft CLIA-accredited lab with state-of-the-art equipment Samples Baseline Follow-up Analytes Existing protein biomarkers Discovery using proteomics & metabolomics Analysis Independent risk prediction Guiding therapy 22
TIMI IIIB 8 Troponin levels in NSTE-ACS 7.5 Mortality at 42 days (%) 7 6 5 4 3 2 1 0 N=1404 1 1.7 P<0.001 χ 2 for trend 3.4 3.7 831 174 148 134 50 67 0 to <0.4 0.4 to <1.0 1.0 to <2.0 2.0 to <5.0 5.0 to <9.0 >9.0 Cardiac Troponin I 6 Antman et al. NEJM 1996; 335: 1342
D/MI/ACS at 6 mos (%) 30 25 20 15 10 5 0 Troponin T: Treatment Interaction CONS 14.5 p=ns 16.9 INV Interaction P<0.001 OR=0.52 p<0.001 24.2 14.3 N= 414 396 463 495 TnT - TnT + TnT cut point = 0.01 ng/ml (54% of Pts TnT +) Cannon NEJM 2001; 344: 1879
Lp-PLA 2 Levels and Risk of Major Adverse CV Events Adj HR = 1.33 (1.01-1.74) 1.74) P trend = 0.002 Event Rate (%) 30 25 20 15 10 17.6 21.8 22.6 22.6 26.4 5 0 654 649 657 654 651 Q1 Q2 Q3 Q4 Q5 MV Model: age, index dx,, prior MI, DM, renal, Rx arm, 30d LDL, 30d CRP O Donoghue M et al., Circulation 2006;113:1745
Hypothesis Targeting Lp-PLA 2 a Key Player in Atherosclerosis native LDL carrier of Lp-PLA 2 Lp-PLA 2 Leukocyte Lumen Atheroma Intima Oxidized LDL substrate for Lp-PLA 2 Lp-PLA 2 Darapladib (Lp-PLA 2 inhibitor) Sustained Inflammation Necrotic Core Expansion Macphee, Biochem J 1999; Zalewski and Macphee. ATVB. 2005; Shi Atherosclerosis 2007; Kolodgie, ATVB 2006 26
Core Labs Pharmacogenetics Laboratory Marc S. Sabatine, MD, MPH - Director Multiple core genotyping laboratories at Harvard & MIT Genotyping Candidate genes Arrays Analysis Risk prediction Response to therapy 27
CYP2C19 & Clinical Outcomes 1477 Patients w/ ACS and planned PCI Rx d w/ clopidogrel CV Death, MI, or Stroke (%) 14 13 12 11 10 9 8 7 6 5 4 3 2 1 CYP2C19 Reduced-Function Carriers Allele Carriers Non-carriers Non-carriers Hazard HR 1.53 Ratio 1.53 (95% (95% CI CI 1.07-2.19) P=0.014 12.1 8.0 Definite or Probable Stent Thrombosis (%) 4 3 2 1 Hazard Ratio 3.09 HR 3.09 (95% CI 1.19-8.00) P=0.015 (95% CI 1.19-8.00) Carriers CYP2C19 Reduced-Function Allele Carriers Non-carriers Non-carriers 2.6 0.8 0 0 30 90 180 270 360 450 Days After Randomization Number at Risk: Non-Carrier 1064 1009 999 980 870 755 542 Carrier 395 364 360 348 306 270 181 0 0 30 90 180 270 360 450 Days After Randomization Number at Risk: Non-Carrier 1014 1004 1001 989 885 765 547 Carrier 375 368 366 359 316 279 186 * Carriers ~30% of the population Mega JL et al. & Sabatine MS. N Engl J Med 2009;360:354-62.
Plavix Prescribing Information, Revised May 2009
Value Added by ARO Internal validity of a clinical trial Compliance with the protocol Data collection & adjudication Statistical analyses External validity or generalizability of a clinical trial Hypothesis to be tested Study population (including location) Endpoints Answering additional questions Well-characterized dataset Embedded mechanistic studies 30