SKG Congress, 2015 EVOLVE II. Stephan Windecker

SKG Congress, 2015 EVOLVE II Stephan Windecker Department of Cardiology Swiss Cardiovascular Center and Clinical Trials Unit Bern Bern University Hospital, Switzerland

BIODEGRADABLE POLYMER DES Stefanini, Taniwaki, Windecker. Heart 2013 All New DES Are Based on Biodegradable Polymer Coatings!

The Stent PLGA Polymer Everolimus Drug 74μm SEM of coating (5000x) Everolimus Mass Remaining PLGA Mass Remaining PLATFORM Thin strut PtCr stent Strength Flexibility & Conformability COATING Bioabsorbable PLGA polymer Abluminal surface only (rollcoat) Short term polymer exposure Synchronized drug release & polymer absorption DELIVERY SYSTEM Extended Laser-Cut Hypotube Pushability and Flexibility The stent is an investigational device and not for sale in the US. CE Mark Approved 2012. Information for the Stent is for use in countries with applicable product registrations In selected higher risk patients where the physician determines that the risks outweigh the benefits of continued DAPT, it may be reasonable to interrupt or discontinue therapy after 1 month based on low stent thrombosis rates and no observed increased risk for stent thrombosis as shown in the current literature.

PLGA Mass Remaining (%) Everolimus Releaesd (%) The Stent: Synchronous Drug Release & Polymer Absorption Kinetics of Drug Release and Polymer Absorption in a Preclinical Porcine Model 100 100 75 75 50 50 25 25 0 0 30 60 90 120 Time (Days) 0 Bennett and Dubois. Biologics: Targets and Therapy. 2013; 7: 149-159.

The Stent: Comparison with PROMUS Bioabsorbable polymer (PLGA) Everolimus applied only to the abluminal surface (rollcoat) Thin strut (74µm) platinum chromium stent Stent Strut Cross Sections PROMUS Element 81μm 74μm Arterial Wall

EVOLVE - Trial Design and Methods Meredith et al. JACC 2012;59(15):1362-70 Patients with de novo native coronary lesions 28 mm in length, RVD 2.25 mm 3.5, %DS>50 (excluded LM disease, CTO, AMI or recent MI) Randomized 1:1:1 at 29 sites (Europe, Australia, New Zealand) PROMUS Element N=98 N=94 ½ Dose N=99 Single-blind, noninferiority design Primary Clinical Endpoint: TLF (TV-CD, TV-MI, or TLR) at 30 days Primary Angiographic Endpoint: In-stent late loss at 6 months Per protocol patients were treated with clopidogrel, ticlopidine or prasugrel for at least 6 months following the index procedure

Late loss, mm Target Lesion Failure, % EVOLVE Primary Endpoint Meredith et al. JACC 2012;59(15):1362-70 Late Loss at 6 Months TLF at 30 days 0,6 P=0.19* P=0.56* 10,0 P=0.49* P=0.25* 0,5 8,0 0,4 0,3 0,2 6,0 4,0 0,1 2,0 0,0 0,15 0,10 0,13 PROMUS Element ½ Dose 0,0 PROMUS Element 0 1,1 3,1 ½ Dose Noninferiority was proven because the upper 95.2% confidence bound of the difference in 6-month late loss is <0.20 for both stents Intent-to-treat; Mean + Standard Deviation; *P values for superiority comparison

TLR, % Target Lesion Revascularisation 4-year Follow-up Meredith EuroPCR 2015 20 Protocol-required angiogram PE vs HR 0.18 [0.02, 1.47] P=0.07 PE vs ½ HR 0.17 [0.02, 1.74] P=0.06 0 0 1 2 3 Numbers at risk PE 98 98 93 92 64 92 90 87 84 59 ½ 99 95 93 90 63 Dose 4 6.1 1.1 1.0 Years Safety Population; KM Event Rate; log-rank P values

Patients, % 4-Year Clinical Outcomes Meredith EuroPCR 2015 10 8,4 PROMUS Element ½ Dose Components of TLF 5 5,5 5,2 6,1 3,3 3,3 3,0 0 1,1 1,0 1,1 1,1 0,0 0,0 (7) (5) (5) (6) (1) (1) (2) (3) (3) (1) (1) 0,0 0,0 TLF TLR MI* Cardiac Death* Def/Prob ST Number of Events (N) Safety Population; KM Event Rates; All P values are >0.05; *Target vessel-related

EVOLVE II Pivotal Trial Design Kereiakes DJ et al. Circ Cardiovasc Interv. 2015;8(4):e002372 Patients with 3 native coronary artery lesions in 2 major epicardial vessels; lesion length 34 mm, RVD 2.25 mm 4.0, %DS 50<100 (excluded LM disease, CTO, SVG, ISR or recent STEMI) Randomized Cohort (RCT) PROMUS Element Plus N=842 Up to 160 global sites N=842 RCT Design Multicenter noninferiority trial Pivotal, single-blind, 1:1 randomization Primary Endpoint: TLF (CD, TV-MI, or TLR) at 12 mo Follow-up through 5 years PK Substudy N=21 Diabetes Substudy N=203 DAPT (ASA + clopidogrel, ticlopidine, prasugrel, ticagrelor) 6 months or longer as tolerated

EVOLVE II Major Endpoints Kereiakes DJ et al. Circ Cardiovasc Interv. 2015;8(4):e002372 Primary endpoint Target lesion failure (TLF) at 12 months Cardiac death, or MI * related to the target vessel, or Ischemia-driven target lesion revascularization ITT and Per Protocol patient populations Additional endpoints Components of TLF Stent thrombosis (ARC definite/probable) Technical success Clinical procedural success Longitudinal stent deformation * Spontaneous MI : rise and/or fall of cardiac biomarkers with 1 value >99th percentile of the URL + evidence of myocardial ischemia. Peri-PCI MI: 1 of the following: i) CK-MB >3X URL within 48 hrs, ii) new pathological Q waves, iii) autopsy evidence.

Per Patient Baseline Clinical Characteristics Kereiakes DJ et al. Circ Cardiovasc Interv. 2015;8(4):e002372 PROMUS Element Plus n=838 patients n=846 patients P value Male 72.7% 70.6% 0.34 Age (yr) ± SD 63.9 ± 10.5 63.5 ± 10.4 0.40 Caucasian 79.2% 77.4% 0.37 Smoking, Ever 62.8% 61.7% 0.63 Current Smoker 22.4% 21.8% 0.76 Diabetes * 30.8% 31.1% 0.89 Treated with Insulin 10.9% 12.3% 0.36 Hyperlipidemia * 74.5% 74.0% 0.82 Hypertension * 75.1% 77.3% 0.29 Previous PCI 37.3% 35.8% 0.52 Previous CABG 6.1% 4.6% 0.18 History of CHF 9.0% 8.3% 0.63 Unstable Angina 34.8% 33.9% 0.69 MI 29.2% 25.9% 0.12 Intent-to-treat; * medically-treated; P values from Student's t test or Chi-square test; SD=standard deviation

Per Patient * Per Lesion Baseline Lesion Characteristics (QCA) PROMUS Element Plus n=1043 lesions n=838 patients n=1059 lesions n=846 patients P value Target lesions * 1.24 ± 0.49 1.25 ± 0.50 0.77-2 lesions treated 19.3% 18.6% 0.69-3 lesions treated 2.4% 3.3% 0.26 - >3 lesions treated 0.1% 0.0% 0.50 Target lesion location : Kereiakes DJ et al. Circ Cardiovasc Interv. 2015;8(4):e002372 LAD 41.5% 41.3% 0.91 LCx 26.4% 25.0% 0.48 RCA 32.0% 33.7% 0.41 LM 0.1% 0.0% 0.50 RVD, mm 2.63 ± 0.50 2.62 ± 0.49 0.63 - RVD <2.25 mm 23.3% 23.9% 0.76 MLD, mm 0.89 ± 0.36 0.89 ± 0.35 0.99 Diameter Stenosis, % 66.26 ± 11.75 66.02 ± 12.03 0.65 Lesion length, mm 13.67 ± 7.00 14.09 ± 7.50 0.18 - Length >20 mm 16.7% 19.2% 0.14 Modified AHA/ACC B2/C 74.3% 76.8% 0.19 Intent-to-treat; P values from Student's t test or Chi-square (Fisher s Exact test denoted by ); MLD=minimum lumen diameter; RVD=reference vessel diameter

Per Patient * Per Lesion Per Stent Procedural Characteristics PROMUS Element Plus n=1043 lesions n=838 patients n=1079 stents n=1059 lesions n=846 patients N=1011 P value Technical success 96.9% 98.3% 0.04 Clinical procedural success * 94.3% 94.9% 0.56 Stents per patient * 1.29 ± 0.56 1.31 ± 0.60 0.46 Stents per target lesion 1.04 ± 0.25 1.05 ± 0.25 0.32 Total Stent Length Implanted (mm) Kereiakes DJ et al. Circ Cardiovasc Interv. 2015;8(4):e002372 20.81 ± 9.16 21.45 ± 9.04 0.11 Pre-dilatation, % 98.0% 97.1% 0.18 Post-dilatation, % 61.0% 60.7% 0.90 Max pressure overall (atm) 16.09 ± 3.13 15.98 ± 3.06 0.41 Longitudinal Stent Deformation 0.1% 0.1% >0.99 LSD occurred in a PROMUS Element Plus stent used in a patient

Post-procedural Angiographic Characteristics Kereiakes DJ et al. Circ Cardiovasc Interv. 2015;8(4):e002372 Per Lesion PROMUS Element Plus n=1043 lesions n=1059 lesions P value MLD, in-stent, mm 2.46 ± 0.44 2.44 ± 0.44 0.23 MLD, in-segment, mm 2.10 ± 0.47 2.10 ± 0.47 0.78 %DS, in-stent, % 6.55 ± 9.71 7.19 ± 9.16 0.12 %DS, in-segment, % 20.93 ± 9.13 20.60 ± 8.41 0.39 Acute gain, in-stent, mm 1.57 ± 0.45 1.55 ± 0.45 0.33 Acute gain, in-segment, mm 1.21 ± 0.47 1.22 ± 0.48 0.72

100 90 Antiplatelet Medication Usage * Kereiakes DJ et al. Circ Cardiovasc Interv. 2015;8(4):e002372 98,7% 98,4% 97,7% 96,9% P=ns for all between group comparisons 98,3% 98,0% 89,7% 87,3% 80 0 0 6 12 Time (Months) (Discharge) ASA: PROMUS Element Plus ASA: DAPT: PROMUS Element Plus DAPT: * Per protocol, patients were treated with one of the following P2Y 12 inhibitors (clopidogrel, ticlopidine, prasugrel, or ticagrelor) for at least 6 months following the index procedure. Intent-to-treat. ASA=acetylsalicylic acid; DAPT=dual antiplatelet therapy

Target Lesion Failure, % *One-sided 97.5% Farrington-Manning Upper Confidence Bound (UCB) EVOLVE II Primary Endpoint: 12-month TLF 10 5 Kereiakes DJ et al. Circ Cardiovasc Interv. 2015;8(4):e002372 ITT Population 6,5 6,7 Difference [97.5% UCB] P=0.0005 2.68% 1-sided UCB * Non-inferiority Margin 4.4% 0 PROMUS Element Plus -5 0 5 PROMUS Element Plus Better Better Noninferiority is proven because the one-sided upper 97.5% confidence bound for the difference in 12-month TLF is <4.4%

Event Rate (%) Components of TLF Kereiakes DJ et al. Circ Cardiovasc Interv. 2015;8(4):e002372 12 PROMUS Element Plus 8 P=0.34 P=0.71 P=0.21 4 0 ITT Population 0,9 0,5 Cardiac Death 4,7 4,3 Target Vessel-Related MI 1,7 2,6 Clinically-indicated TLR * Per protocol spontaneous MI is defined as rise and/or fall of cardiac biomarkers with 1 value >99th percentile of the URL + evidence of myocardial ischemia. Peri-PCI MI is defined as 1 of the following: i) biomarker elevations within 48 hours of PCI (based on CK-MB >3X URL), ii) new pathological Q waves, or iii) autopsy evidence of acute MI

Stent Thrombosis through 12-months Definite/Probable : ITT Population Kereiakes DJ et al. Circ Cardiovasc Interv. 2015;8(4):e002372 Acute ( 1 day) Subacute (2-30 days) Late (30 days 1 year) PROMUS Element Plus N=2 (Definite) N=5 (2 Definite/3 Probable) N=1 (Probable) 0.6% (N=5) P=0.50 0.4% (N=3) No definite/probable stent thrombosis in the arm after Day 6

EVOLVE II Clinical Program Patients with 3 native coronary artery lesions in 2 major epicardial vessels; lesion length 34 mm, RVD 2.25 mm 4.0, %DS 50<100 (excluded LM disease, CTO, SVG, ISR or recent STEMI) EVOLVE II Randomized Cohort EVOLVE II Diabetes Substudy Diabetic Patients from: PROMUS Element Plus N=838 N=846 cohort EVOLVE II RCT N=263 + Single-arm Diabetes Study N=203 RCT Design Multicentre, single-blind, 1:1 randomisation 1 Endpoint: TLF at 12 mo Presented at AHA 2014 1 endpoint met EVOLVE II Diabetes Substudy Design Consecutive, multicentre, single-arm, non-randomised 1 Endpoint: TLF at 12 mo

EVOLVE II Diabetes Substudy Design Design Prospective, single-arm, multicentre, observational study Diabetic subjects randomized to the cohort of the EVOLVE II RCT pooled with subjects enrolled in the consecutive Diabetes single-arm study following completion of EVOLVE II RCT enrollment Primary endpoint Target lesion failure (TLF) at 12 months Cardiac death, or MI related to the target vessel (based on CK-MB >3x URL), or Ischemia-driven target lesion revascularisation Compared to a performance goal based on historical results in diabetic patients 1-4 1. Stone GW et al. J Am Coll Cardiol. 2011;57(16):1700-8; 2. Stone GW et al. N Engl J Med. 2010;362(18):1663-74.; 3. Kedhi E et al. Lancet. 2010;375(9710):201-9.; 4. Meredith IT et al. J Am Coll Cardiol. 2012;59:1362-1370.

Clinical Characteristics Baseline Clinical and Lesion Characteristics n=466 patients Lesion Characteristics (Core laboratory) n=466 patients Male 70.2% Target Lesions 1.26 ± 0.51 Age (yr) ± SD 64.8 ± 9.7-2 Lesions Treated 20.0% Caucasian 75.3% - 3 Lesions Treated 3.2% Smoking, Ever 59.6% LAD 39.7% Current Smoker 16.6% Target Lesion LCx 26.0% Diabetes 100% Location : RCA 34.3% Treated with Insulin 37.3% LM 0.0% Hyperlipidemia 84.5% RVD, mm 2.56 ± 0.50 Hypertension 88.8% - RVD <2.25 mm 27.2% Previous PCI 41.8% MLD, mm 0.89 ± 0.36 Previous CABG 7.5% Diameter Stenosis, % 65.47 ± 11.70 History of CHF 10.3% Lesion Length, mm 14.10 ± 7.49 Unstable Angina 35.6% - Length >20 mm 19.7% NSTEMI 26.2% Modified AHA/ACC B2/C 74.9% Intent-to-treat; Per patient unless per lesion indicated by (N=589 lesions); SD=standard deviation

Antiplatelet Medication Usage * 100% 80% 99,8% 98,2% 97,8% 97,6% 96,9% 89,5% ASA DAPT 60% 40% 20% 0% 0 6 12 Time (Months) (Discharge) * Per protocol, patients were treated with one of the following P2Y 12 inhibitors (clopidogrel, ticlopidine, prasugrel, or ticagrelor) for at least 6 months following the index procedure. Intent-to-treat. ASA=acetylsalicylic acid; DAPT=dual antiplatelet therapy

EVOLVE II Diabetes 1 Endpoint: 12-month TLF Target Lesion Failure, % 20 P<0.0001 15 Performance Goal = 14.5% 10 1-sided 97.5% UCB* 5 0 7,5 ITT P-value from the one-sided Clopper-Pearson test is <0.025, the 12-month TLF rate from is concluded to be less than the performance goal (14.5%) *One-sided 97.5% Clopper Pearon Upper Confidence Bound (UCB)

TLF (%) EVOLVE II Diabetes 1 Endpoint: 12-month TLF : ITT 15 10 5 7.6% 0 No. at risk 0 6 12 DM 466 442 315 Months ITT; KM Event Rate; log-rank P value

TLF (%) EVOLVE II Diabetes 1 Endpoint: 12-month TLF : ITT 15 10 5 7.6% 6.7% 0 No. at risk 0 6 12 DM 466 442 315 Months RCT 846 802 600 ITT; KM Event Rate; log-rank P value

Percent Clinical Outcomes at 12 months 25 20 15 EVOLVE II Diabetes Substudy Definite/Probable ST Timing DM Acute ( 1 day) 0.9% Subacute (2-30 days) 0.2% Late (31-365 days) 0% 10 5 0 1,3 All Death 0,4 QWMI 5,5 Non- QWMI 8,6 7,5 4,4 1,1 TVF TLF TLR ST ARC Def/Prob * * Cutlip et al, Circulation. 2007; 115(17):2344; Spontaneous MI was defined as the rise and/or fall of cardiac biomarkers with 1 value >99th percentile of the upper reference limit (URL) with 1 of the following: symptoms of ischemia, ECG changes, and or evidence of loss of myocardium. Peri-PCI MI was defined by any of the following: i) CK-MB >3X URL within 48 hours, ii) new pathological Q waves, iii) autopsy evidence. All ST were definite.

Percent Clinical Outcomes at 12 months EVOLVE II Diabetes Substudy EVOLVE II RCT: arm 25 20 15 Definite/Probable ST Timing DM RCT Acute ( 1 day) 0.9% 0.2% Subacute (2-30 days) 0.2% 0.1% Late (31-365 days) 0% 0% 10 5 0 1,3 1,1 All Death 0,4 0,2 QWMI 5,5 5,2 Non- QWMI 8,6 8,2 7,5 6,6 4,4 2,6 1,1 0,4 TVF TLF TLR ST ARC Def/Prob * * Cutlip et al, Circulation. 2007; 115(17):2344; Spontaneous MI was defined as the rise and/or fall of cardiac biomarkers with 1 value >99th percentile of the upper reference limit (URL) with 1 of the following: symptoms of ischemia, ECG changes, and or evidence of loss of myocardium. Peri-PCI MI was defined by any of the following: i) CK-MB >3X URL within 48 hours, ii) new pathological Q waves, iii) autopsy evidence.

Conclusions and Significance In this pivotal non-inferiority trial designed to support approval of the first bioresorbable polymer DES in the U.S., the stent proved non-inferior to the Promus Element Plus stent for TLF at 1 year. Procedural, angiographic and clinical outcomes were comparable between stents in a more comers population (>60% ACS, >25% MI, 31% diabetes, smaller vessels, longer lesions, 75% AHA/ACC B2/C lesion morphology). Despite the clinical and angiographic complexity of the study population, definite/probable stent thrombosis rates were low. Definite ST not observed beyond 24 hrs following. The longer term relative efficacy and safety of the stent is currently under evaluation.