Keynote Lecture The 5 Best Trials from 2012 that will Impact My Practice Gregg W. Stone, MD Columbia University Medical Center NewYork-Presbyterian Hospital Cardiovascular Research Foundation

Consulting: Boston Scientific, Volcano Gregg W. Stone, MD Consulting: Boston Scientific, Volcano

Top 6 Trials of 2012 SYNTAX 5-Year Outcomes FREEDOM FAME 2 WOEST PARTNER 2-Year RESPECT

Mohr FW et al. Lancet 2013

Mohr FW et al. Lancet 2013

Mohr FW et al. Lancet 2013

Mohr FW et al. Lancet 2013

1 Endpoint: Death, Stroke, or MI FREEDOM: 1900 pts with diabetes +MVD randomized to SES/PES vs. CABG 1 Endpoint: Death, Stroke, or MI 30 PCI/DES CABG 26.6% 20 18.7% Death, Stroke, MI, % 13.0% 10 11.9% P = 0.005 1 2 3 4 5 6 Years PCI/DES 953 848 788 625 416 219 40 CABG 943 814 758 613 422 221 44 Farkouh ME et al. NEJM 2012 8

Death, Stroke, MI by Syntax Score SYNTAX Score  22 (N=669) SYNTAX Score 23-32 (N=844) 40 40 PCI/DES (N=329) CABG (N=340) PCI/DES (N=438) CABG (N=406) 30 30 27.2% 23.2% Death, MI, Stroke (%) 20 Death, MI, Stroke (%) 20 17.2% 17.7% 10 10 0.0 1.0 2.0 3.0 4.0 5.0 0.0 1.0 2.0 3.0 4.0 5.0 Years Years SYNTAX Score 33 (N=374) Years 0.0 1.0 2.0 3.0 4.0 5.0 PCI/DES (N=182) CABG (N=192) 40 30 20 10 30.6% PCI/DES Pint=0.58 22.8% Death, MI, Stroke (%) CABG Farkouh ME et al. NEJM 2012 9

FAME 2: Flow Chart 73% 27% N = 1220 Registry Randomized Trial 11% Silent ischemia 21% Class I 45% Class II 16% Class III 7% Class IV, stabilized Stable CAD patients scheduled for PCI N = 1220 FFR in all target lesions Randomized Trial Registry At least 1 stenosis with FFR ≤ 0.80 (n=888) When all FFR > 0.80 (n=332) *Of 1,632 planned * 332/1220 pts (27.2%) in whom PCI was planned had no physiologically significant lesion and were treated with MT only in the registry Randomization 1:1 PCI + MT MT MT 73% 27% 50% randomly assigned to FU Follow-up: 1 month, 6 months, 1, 2, 3, 4, and 5 years (mean ~7 mos) De Bruyne B et al. NEJM 2012:on-line 10

De Bruyne B et al. NEJM 2012:on-line FAME 2: Primary Endpoint All-cause death, MI or urgent revascularization PCI+MT vs. MT: HR 0.32 (0.19-0.53); p<0.001 30 PCI+MT vs. Registry: HR 1.29 (0.49-3.39); p=0.61 25 MT vs. Registry: HR 4.32 (1.75-10.7); p<0.001 20 urgent revasc (%) Death, MI, or 15 10 5 1 2 3 4 5 6 7 8 9 10 11 12 Months No. at risk MT 441 414 370 322 283 253 220 192 162 127 100 70 37 PCI+MT 447 414 388 351 308 277 243 212 175 155 117 92 53 Registry 166 156 145 133 117 106 93 74 64 52 41 25 13 De Bruyne B et al. NEJM 2012:on-line 11

FAME 1: Primary Endpoint 1005 pts with MVD undergoing PCI with DES were randomized to FFR-guided vs. angio-guided intervention FFR-guided (n=509) 30 days 2.9% 90 days 3.8% 180 days 4.9% 360 days 5.1% Angio-guided (n=496) FFR assessment → PCI deferred in 37.0% of lsns Days Freedom from death, MI, revasc 60 120 180 240 300 360 0.70 0.75 0.80 0.85 0.90 0.95 1.00 MACE 13.2% vs. 18.3% P=0.02 At 30 days, there is already a absolute diff of 2.9 %, but this is steadily increasing over the year and reaches a value of 5.3% at 1 year Tonino PAL et al. NEJM 2009;360:213–24 12

FAME 1: Relationship Between Visual Angiographic %DS and FFR (n=1,329) % of lesions: 47% 39% 15% 1.0 0.8 96% 0.6 35% % lesions with FFR ≤0.80 FFR 80% 0.4 0.2 0.0 50-70% 71-90% 91-99% Stenosis by angiography (visual assessment) Tonino PAL et al. JACC 2010;55;2816-2821

Stone GW et al. TCT2012; Submitted. Correlation Between QCA DS% and FFR ≤0.80 (n=544) F1RST P<0.0001 1/12 16/81 47/237 64/149 28/47 12/15 Stone GW et al. TCT2012; Submitted.

How SYNTAX, FREEDOM and FAME should change practice CABG is superior to first generation DES in complex coronary artery disease ( diabetes) The heart team should evaluate high-risk pts Most pts with high Syntax Score should undergo surgery In other patients PCI is reasonable, but we must do better! Best in class stents Optimal pharmacotherapy Ischemia-guided complete revascularization

Study Design WOEST Dewilde W et al. ESC 2012 Inclusion criteria - Indication for OAC for ≥1 year - PCI of a single coronary lesion Study Design 1:1 Randomisation: Double therapy group: OAC + 75mg Clopidogrel qd 1 month minimum after BMS 1 year after DES Triple therapy group OAC + 75mg Clopidogrel qd + 80mg Aspirin qd 1 month minimum after BMS 1 year after DES Follow up: 1 year Primary Endpoint: The occurence of all bleeding events (TIMI criteria) (powered for a reduction from 12% to 5%) Secondary Endpoints: Combination of stroke, death, myocardial infarction, stent thrombosis and target vessel revascularisation - All individual components of primary and secondary endpoints |

Primary Endpoint: TIMI major or minor or minimal bleeding WOEST: n=583 pts requiring chronic oral anticoagulant therapy Primary Endpoint: TIMI major or minor or minimal bleeding Days Cumulative incidence of bleeding 30 60 90 120 180 270 365 0 % 10 % 20 % 30 % 40 % 50 % 284 210 194 186 181 173 159 140 n at risk: 279 253 244 241 236 226 208 Triple therapy group Double therapy group 44.9% 19.5% p<0.001 HR=0.36 95%CI[0.26-0.50] |

Secondary Endpoint (death, MI, TVR, Stroke, ST) WOEST Secondary Endpoint (death, MI, TVR, Stroke, ST) Days Cumulative incidence 30 60 90 120 180 270 365 0 % 5 % 10 % 15 % 20 % 284 272 266 261 252 242 223 n at risk: 279 276 273 263 258 234 Triple therapy group Double therapy group 17.7% 11.3% p=0.025 HR=0.60 95%CI[0.38-0.94]

All-Cause Mortality WOEST Triple therapy group Double therapy group 7.5 % Triple therapy group Double therapy group 6.4% HR=0.39 95%CI[0.16-0.93] p=0.027 5 % Cumulative incidence of death 2.6% 2.5 % 0 % 30 60 90 120 180 270 365 Days n at risk: 284 281 280 280 279 277 270 252 279 278 276 276 276 275 274 256

How WOEST will change medicine Recent studies have emphasized “stacking” potent therapies, valuing efficacy over safety The deleterious impact of side-effects such as bleeding have been overlooked WOEST, by itself, is too small and underpowered to change practice. However… WOEST is the first of what will hopefully be many studies designed to “peel back” layered therapies, thereby reducing risks and cost

Kodali S et al. NEJM 2012:on-line PARTNER 2A: Study Flow Randomized = 699 patients Transfemoral n = 492 TF = 492 (70%) TA = 207 (30%) Transapical n = 207 TAVR (244) AVR (248) TAVR (104) AVR (103) 2 Years Alive = 157 Dead = 74 LTFU = 2 Withdrawal = 2 Censored* = 9 2 Years Alive = 143 Dead = 80 LTFU = 2 Withdrawal = 16 Censored* = 7 2 Years Alive = 59 Dead = 42 LTFU = 1 Withdrawal = 0 Censored* = 2 2 Years Alive = 57 Dead = 34 LTFU = 1 Withdrawal = 8 Censored* = 3 Of the 699 patients in the study, 207 were in the TA arm and 492 were in the TF arm. 2 year follow-up was over 95% in all groups. 95.4% follow-up at 2 years 96.1% follow-up at 2 years 97.1% follow-up at 2 years 95.8% follow-up at 2 years *Censored = Patient is alive at last contact but no information available within follow-up window Kodali S et al. NEJM 2012:on-line

All-Cause Mortality (ITT) TAVR AVR HR [95% CI] = 0.88 [0.70, 1.12] p (log rank) = 0.310 35.0% All-Cause Mortality 26.8% 33.9% 24.3% The primary analysis reveals no difference in mortality between TAVR and AVR with a hazard ratio of 0.88. The two year KM estimates of mortality are 35% for AVR and 33.9% for TAVR. Months Numbers at Risk TAVR 348 298 260 234 172 70 31 AVR 351 252 236 217 165 65 32 Kodali S et al. NEJM 2012:on-line

All-Cause Mortality or Rehospitalization (ITT) TAVR AVR HR [95% CI] = 0.98 [0.79, 1.21] p (log rank) = 0.836 46.6% All-Cause Mortality or Rehospitalization 37.7% 46.5% 34.9% The KM curve for all cause mortality or repeat hospitalization shows no difference between the two arms with a hazard ratio of 0.98. Months Post Procedure Numbers at Risk TAVR 348 257 223 194 139 48 18 AVR 351 222 200 183 130 53 26

Strokes (ITT Population) 5 10 15 20 25 30 35 Total 24 20 TAVR AVR Number of Events ≤ 30 Days 16 8 TAVR AVR > 30 days 8 12 TAVR AVR Another way to look at this is to numerically count events. Within 30 days, there are twice as many strokes in the TAVR group reflecting the increased periprocedural risk. However, after 30 days there are numerically more strokes in the AVR group such that the total events over the course of follow-up are 24 in the TAVR group and 30 in the AVR group. Many of these are late events with 5 occurring in the AVR group between 2 and 3 years. Kodali S et al. NEJM 2012:on-line

Mild Paravalvular AR and Mortality TAVR Patients (AT) None - Trace Mild Moderate - Severe p (log rank) < 0.001 41.7% Mortality 39.2% 29.5% 29.2% 24.8% 14.5% When you separate the groups further, it is clear that mortality is being driven by patients with mild AR who have higher mortalities. These curves start to separate in the first year. Months Post Procedure Numbers at Risk None-Tr 167 149 140 126 87 41 16 Mild 136 115 95 86 51 21 10 Mod-Sev 24 19 17 15 13 5 2

Implications of PARTNER A at 2 yrs At many centers TAVR has become the preferred treatment in high-risk patients with aortic stenosis (FDA approved, CMS reimbursed) - Integrated heart team approach is mandatory Stroke, vascular complications, and now paravalvular regurgitation remain major issues for TAVR which are being addressed by improved technique, novel device designs and adjunctive devices The role off TAVR is being explored in moderate-risk pts in ongoing trials

RESPECT: Subject Distribution With cryptogenic stroke within 9 months TEE with bubble study at 6 months CARROLL JD et al. NEJM 2013 13 1. Aspirin + clopidogrel was removed from the protocol in 2006 based on changes to the AHA/ASA treatment guidelines

980 pts w/cryptogenic stroke within 9 mos Device Performance (in 499 pts) Maximum Residual Shunting at Rest and Valsalva at 6 Months Grade 0: 72.7% Grade 1: 20.8% Grade 2-3: 6.5% CARROLL JD et al. NEJM 2013 Defined as successful delivery and release of the device for subjects in whom the delivery system was introduced into the body Defined as successful implantation with no reported in-hospital serious adverse events Defined as complete obliteration or trivial residual shunting (Grade 0 or I at rest and Valsalva) at 6 months, adjudicated by echo core lab 17

980 pts with cryptogenic stroke within 9 mos Procedure/device-related SAEs For all AE’s, atrial fibrillation occurred in 3.0% versus 1.5% in the device and medical groups respectively, p=0.13 Pericardial tamponade is a subset of major bleeds, and thus counted in the major bleed category as well For all SAEs, pulmonary embolism occurred in 1.2% and 0.2% in device and medical groups, respectively, p=0.124 1 case of right atrial thrombus resulted in abandonment of device implant procedure (no device received); 1 case of right atrial thrombus (located inferiorly) not attached to device detected in patient with DVT and PE 4 months after procedure 1 ischemic stroke one week post implant; 1 five months post implant with finding of severe shunting related to previously undiagnosed sinus venosus defect, requiring surgical closure For all SAEs, there were 3 device group deaths (0.6%) and 6 medical group deaths (1.2%) all of which were not study related, p= 0.334 P-values are calculated using Fisher’s Exact test 16

Primary Endpoint Analysis – ITT Cohort 50 Primary Endpoint Analysis – ITT Cohort 50.8% risk reduction of stroke in favor of device N=980 3/9 device group patients did not have a device at time of endpoint stroke 20 Cox model used for analysis

Primary Endpoint Analysis – As Treated Cohort 72 Primary Endpoint Analysis – As Treated Cohort 72.7% risk reduction of stroke in favor of device N=980 The As Treated (AT) cohort demonstrates the treatment effect by classifying subjects into treatment groups according to the treatment actually received, regardless of the randomization assignment 22 Cox model used for analysis

Subpopulation Differential Treatment Effect 24

How will RESPECT change practice? The rates of follow-up stroke were low; PFO closure is not required in most patients with cryptogenic stroke, especially if the shunt is small A comprehensive work-up for other causes of stroke should be performed, such as concealed paroxysmal atrial fibrillation PFO closure may be appropriate for selected pts, such as large shunt, large or recurrent stroke, or anticoagulant ineligible New therapies (drugs and devices) should not be widely adopted without positive demonstration of clinical benefit in RCT(s)

Top ACC Late Breaking Trials PREVAIL: LAA closure with the Watchman as an alternative to OAC PARTNER 2, Cohort B: SAPIEN XT vs. standard of care in inoperable pts with aortic stenosis CHAMPION PHOENIX: Cangrelor in patients undergoing PCI STREAM: Pharmacoinvasive strategy in STEMI pts with delay to primary PCI