Everolimus-eluting Bioresorbable Vascular Scaffolds in Patients with Coronary Artery Disease: The ABSORB III trial Dean J. Kereiakes, MD, Stephen G. Ellis, MD, D. Christopher Metzger, MD, Ronald P. Caputo, MD, David G. Rizik, MD, Paul S. Teirstein, MD, Marc R. Litt, MD, Annapoorna Kini, MD, Ameer Kabour, MD, Steven O. Marx, MD, Jeffrey J. Popma, MD, Robert McGreevy, PhD, Zhen Zhang, PhD, Charles Simonton, MD and Gregg W. Stone, MD for the ABSORB III Investigators

Disclosure Statement of Financial Interest Within the past 12 months, I or my spouse/partner have had a financial interest/arrangement or affiliation with the organization(s) listed below. Affiliation/Financial Relationship Company Modest Consulting Fees Significant Consulting Fees Major Stock Shareholder/Equity HCRI Boston Scientific Abbott Vascular Svelte Medical Systems, Inc. Janssen Research & Development LLC Sanofi-Aventis U.S. LLC Ablative Solutions, Inc.

Absorb BVS Fully Bioresorbable Everolimus/PDLLA (1:1) matrix coating Conformal coating Controlled drug release similar to Xience CoCr-EES PLLA Backbone Semi-crystalline Circumferential sinusoidal rings connected by linear links Strut thickness 150 µm Platinum markers in each end ring Fully Bioresorbable

Oberhauser JP et al. EuroInt 2009;5:F15-22 Phases of Absorb Functionality Revascularization Restoration Resorption Mechanical Support Mass loss Drug Elution 1 3 6 Months 24 Oberhauser JP et al. EuroInt 2009;5:F15-22 4

Metallic DES vs. Absorb BVS Representative Human images at 5 Years Metallic DES1 Absorb-Treated Artery2 1Atherosclerosis 2014;237:23e29 2 Images courtesy of S Windecker, ABSORB Cohort B 5 Yrs

Absorb Program Objectives A Series of Randomized Trials Designed to: Demonstrate similar (non-inferior) results with ABSORB BVS compared to Xience CoCr-EES at 1 year Demonstrate superior results with ABSORB BVS compared to Xience CoCr-EES between 1 and 5 years

ABSORB III Study Design Prospective, multicenter, single-blind, trial ~2,000 patients randomized 2:1 Absorb BVS vs. Xience CoCr-EES Clinical follow-up: 30 d 6 mo 12 mo 24 mo 36 mo 48 mo 60 mo No routine angiographic follow-up 7

193 Enrolling Centers U.S. Australia

Top Enrollers (N patients) 1. Dr. Metzger (76) Holston Valley Wellmont Medical Center 8. Dr. DeGregorio (38) Englewood Hospital and Medical Center 15. Dr. Waksman (26) MedSTAR Washington Hospital Center 2. Dr. Caputo (52) St. Joseph's Hospital Health Center 9. Dr. Cannon (36) Northern Michigan Hospital 16. Dr. Abbas (24) William Beaumont Hospital 3. Dr. Rizik (49) Scottsdale Healthcare 10. Drs. Cambier & Stein (35) Morton Plant Hospital, 17. Dr. Zidar (24) Rex Hospital, Inc 4. Dr. Teirstein (45) Scripps Green Hospital 11. Dr. Newman (34) WakeMed 18. Dr. Dearing (24) Thomas Hospital 5. Dr. Litt (42) Baptist Medical Center 12. Dr. Feldman (31) Munroe Regional Medical Center 19. Dr. Williams (23) Presbyterian Hospital 6. Dr. Kini (41) Mount Sinai Medical Center 13. Dr. Broderick (28) The Christ Hospital 20. Dr. Choi (23) Baylor Jack and Jane Hamilton Heart and Vascular Hospital 7. Dr. Kabour (41) Mercy St. Vincent Medical Center 14. Dr. Jain (28) Washington Hospital, Fremont, CA 21. Dr. Moses (23) Columbia University Medical Center

Study Leadership Principal Investigators Dean Kereiakes, MD, The Christ Hospital, Cincinnati, OH Stephen G. Ellis, MD, Cleveland Clinic, Cleveland, OH Study Chairman Gregg W. Stone, MD, Columbia University Medical Center, NY, NY Clinical Events Committee Cardiovascular Research Foundation, New York, NY Steven Marx, MD, chair Angiographic Core Laboratory Beth Israel Deaconess Medical Center, Boston, MA Jeff Popma, MD, director Data Safety Monitoring Board Axio Research, Seattle, WA; Robert N. Piana, MD, chair Sponsor Abbott Vascular, Santa Clara, CA 10

Major Endpoints at 1 Year Primary Endpoint: Target Lesion Failure (non-inferiority) Cardiac death, or Myocardial infarction attributed to the target vessel (TV-MI), or Peri-procedural MI: CK-MB >5x ULN w/i 48 hours Ischemia-driven target lesion revascularization (ID-TLR) Powered Secondary Endpoints (superiority) Angina All revascularization Ischemia-driven target vessel revascularization (ID-TVR) 11

Maximum observed difference in order to pass non-inferiority = 2% Statistical Design Primary Endpoint Non-inferiority analysis for TLF at 1 year with the following assumptions: 1-year TLF rate of 7% Non-inferiority margin of 4.5% “Putative placebo”, preserving ≥ 50% of the treatment effect of Xience vs. BMS 1-sided alpha of 0.025 (equivalent to 2-sided 0.05) 95% 1-year follow-up 2000 subjects → 96% power Maximum observed difference in order to pass non-inferiority = 2% 12

Key Patient Eligibility Criteria >18 years old Evidence of myocardial ischemia (stable/unstable/post- infarction angina or silent ischemia) No elevation of CK-MB 1 or 2 de novo target lesions in up to 2 native coronary arteries (max 1 lesion per artery) Diameter stenosis ≥50% and <100% with TIMI flow ≥1 If <70%, abnormal functional test (including FFR ≤0.80), unstable angina or post-infarct angina RVD ≥2.50 mm and ≤3.75 mm (site-determined) Lesion length ≤24 mm (site-determined) 13

Study Flow and Follow-up Randomized 2:1 N=2008 (ITT) ABSORB N=1322 Xience N=686 N=4 lost to follow-up N=6 withdrew consent N=6 lost to follow-up N=3 withdrew consent ABSORB N=1312 Xience N=677 12-month Follow-up 99.2% Complete 98.7% Complete

Baseline Characteristics Absorb (N=1322) Xience (N=686) p-value Age (mean) 63.5 ±10.6 63.6±10.3 0.75 Male 70.7% 70.1% 0.80 Race (Caucasian) 87.1% 88.3% 0.44 Current tobacco use 21.3% 20.7% 0.77 Hypertension 84.9% 85.0% 0.95 Dyslipidemia 86.2% 86.3% 0.97 Diabetes 31.5% 32.7% 0.60 Insulin-treated 10.5% 11.2% Prior MI 21.5% 22.0% 0.79 Prior coronary intervention 38.7% 38.0% Stable angina 57.3% 60.8% 0.13 Unstable angina 26.9 % 24.5% 0.25 Silent ischemia 10.0% 10.2% 0.88 Single vessel disease 69.5% 67.2% 0.29

Baseline Lesion Characteristics (QCA) Absorb (N=1322) (L=1385) Xience (N=686) (L=713) p-value ACC/AHA lesion class B2/C 68.7% 72.5% 0.08 # of target lesions treated 1.0 ± 0.2 0.38 One 95.1% 96.1% 0.32 Two 4.8% 3.9% 0.36 Target lesion LAD 44.5% 42.2% 0.31 RCA 29.2% 27.2% 0.35 Circumflex 26.2% 30.6% 0.03 Lesion length, mm 12.60 ± 5.41 13.12 ± 5.82 0.05 RVD, mm 2.67 ± 0.45 2.65 ± 0.46 RVD <2.25 mm 18% 19% 0.39 MLD, mm 0.92 ± 0.37 0.90 ± 0.34 0.11 %DS 65.3 ± 12.5 65.9 ± 11.7 0.24 N= number of subjects L= number of lesions

Procedural Characteristics Absorb (N=1322) (L=1385) Xience (N=686) (L=713) p-value Per Subject Bivalirudin use 60.7% 58.7% 0.39 GP IIb/IIIa inhibitor use 10.1% 12.4% 0.11 Only unassigned devices implanted 4.4% 0.6% <0.001 Unplanned overlapping devices 6.2% 8.5% 0.06 Post-dilatation performed 65.5% 51.2% Intravascular imaging guidance 11.2% 10.8% 0.81 Procedure duration (min) 42.2 ± 23.1 38.3 ± 20.9 Per Lesion Total study device length (mm) 20.5 ± 7.2 20.7 ± 9.0 0.56 Max device/balloon diameter (mm) 3.18 ± 0.43 3.12 ± 0.45 0.007 Max device/balloon to vessel diameter ratio 1.21  0.15 1.19  0.14 0.05 Maximum device/balloon pressure (atm.) 15.4 ± 3.0 15.4 ± 3.2 0.83 N= number of subjects L= number of lesions

Post-procedural QCA Measurement Absorb (N=1322) (L=1385) Xience p-value RVD 2.70 ± 0.45 2.68 ± 0.47 0.33 In-Device MLD 2.37 ± 0.40 2.49 ± 0.40 <0.0001 Acute gain 1.45 ± 0.45 1.59 ± 0.44 %DS 11.6 ± 8.77 6.4 ± 8.91 In-Segment 2.15 ± 0.41 2.14 ± 0.43 0.58 1.23 ± 0.46 1.24 ± 0.44 0.50 20.0 ± 7.94 19.8 ± 8.20 0.55 . N= number of subjects L= number of lesions

Acute Success Absorb (N=1322) (L=1385) Xience (N=686) (L=713) p-value Device Success 94.3% 99.3% <0.0001 Procedural Success 94.6% 96.2% 0.12 Device Success (lesion basis) Successful delivery and deployment of study scaffold/stent at intended target lesion Successful withdrawal of delivery system and final in-scaffold/stent DS <30% (QCA) Procedure Success (patient basis) Successful delivery and deployment of at least one study scaffold/stent at intended target lesion No in-hospital (maximum 7 days) TLF

Antiplatelet Agent Usage Absorb (N=1322) Xience (N=686) p-value At index procedure P2Y12 inhibitor 99.0% 98.8% 0.70 Clopidogrel 62.6% 64.7% 0.34 Prasugrel 21.8% 19.5% 0.24 Ticagrelor 14.8% 14.9% 0.94 Aspirin 99.3% 1.00 At 30 days 99.1% 0.81 68.3% 72.0% 0.09 20.7% 17.5% 0.08 11.8% 10.6% 0.44 Aspirin usage 98.6% 0.43 At 1 year 94.4% 95.0% 0.55 67.5% 72.2% 0.03 17.9% 14.0% 9.0% 8.9% 95.8%

Months Post Index Procedure Target Lesion Failure 100% Absorb BVS (n=1322) Xience CoCr-EES (n=686) Diff [95% CI] = 1.7% [-0.5% to 3.9%] Psuperiority=0.16 20% 15% 10% 5% 0% 1 2 3 4 5 6 7 8 9 10 11 13 7.7% 6.0% 12 80% 60% TLF (%) 40% 20% 0% 1 2 3 4 5 6 7 8 9 10 11 12 13 Months Post Index Procedure No. at Risk: Absorb 1322 1254 1230 1218 1196 Xience 686 661 651 643 634

Peri-Procedural MI by Definition CK-MB threshold Absorb (N=1322) Xience (N=686) Difference p-value >3x ULN 6.8% 6.6% 0.2 0.89 >5x ULN (protocol) 3.0% 2.8% 0.75 >8x ULN 1.3% 0.0 0.96 >10x ULN 0.9% 1.2% -0.3 0.58 SCAI definition* *>10x ULN or >5x ULN with new Q waves or new persistent LBBB J Am Coll Cardiol 2013;62:1563-70

Device Thrombosis to 1 Year Absorb (N=1322) Xience (N=686) p-value Device Thrombosis (def*/prob) 1.54% 0.74% 0.13 - Early (0 to 30 days) 1.06% 0.73% 0.46 - Late (> 30 to 1 year) 0.46% 0.00% 0.10 - Definite* (1 year) 1.38% 0.21 - Probable (1 year ) 0.15% 0.55 *One “definite ST” in the Absorb arm by ITT was in a pt that was treated with Xience

1-Year Device Thrombosis Subgroup Absorb (N=1322) Xience (N=686) RR (95% CI) Relative Risk p-value (interaction) Age ≥64 years 1.8% 0.6% 3.22 (0.73-14.32 0.38 Age <64 years 1.2% 0.9% 1.33 (0.36-4.99) Female 1.6% 2.0% 0.79 (0.23-2.78) 0.07 Male 1.5% 0.2% 7.21 (0.95-54.63) Diabetes 3.2% 1.4% 2.34 (0.67-8,13) 0.78 No diabetes 0.8% 0.4% 1.79 (0.37-8.56) Unstable angina/recent MI 1.0% 1.88 (0.21-16.74) 0.91 Stable CAD 1.7% 2.16 (0.73-6.42) Single TL/TV treated 2.09 (0.79-5.55) n/a Dual TL/TV treated 0.0% - Clopidogrel 0.7% 2.69 (0.78-9.24) 0.33 Prasugrel or ticagrelor 0.96 (0.18-5.20) ACC/AHA class A or B1 1.36 (0.14-12.98) 0.67 ACC/AHA class B2 or C 1.9% 2.32 (0.79-6.87) Lesion length <11.75 mm 1.58 (0.43-5.78) 0.56 Lesion length ≥11.75 mm 2.82 (0.63-12.67) RVD <2.63 mm 2.3% 2.65 (0.77-9.07) 0.48 RVD ≥2.63 mm 1.28 (0.25-6.54) Favors Absorb Favors Xience

Powered Secondary Endpoints Absorb (N=1322) Xience (N=686) p-value Angina 18.3% 18.4% 0.93 All Revascularization 9.1% 8.1% 0.50 ID-TVR 5.0% 3.7% 0.21

ABSORB III Very Small Vessel Analysis ABSORB III eligibility criteria included vessels with RVD 2.5 mm – 3.75 mm (visual estimation) The thicker struts of ABSORB may be of particular concern in very small vessels Subgroup analysis was therefore performed in vessels with QCA RVD <2.25 mm vs. ≥2.25 mm (correlates with visual estimate ~2.5 mm) ~19% of patients had a target lesion with RVD <2.25 mm by QCA 30

Median based on pooled Absorb and Xience Outcomes by QCA RVD 2.25 mm RVD <2.25 mm (median 2.09 mm) RVD ≥2.25 mm (median 2.74 mm) TLF: Pint diff = 0.31 ST: Pint diff = 0.12 1-Year Events (%) # Events: 31 11 2 71 30 9 3 # Risk: 241 133 238 1067 542 1058 540 Median based on pooled Absorb and Xience

Limitations ABSORB III enrolled patients with stable ischemic heart disease and stabilized ACS, and excluded specific complex lesions (e.g. CTO, LM, large bif); results may therefore not be generalizable to higher- risk patients and more complex disease Underpowered for low frequency events Results should be viewed in context that Xience was the control device which has been associated with the lowest rates of ST and other events BVS is a first generation device and was used for the first time by most operators within this trial 32

Summary and Conclusions (1) ABSORB BVS was non-inferior to Xience CoCr- EES for TLF at 1 year (primary endpoint met) TLF components (cardiac death, TV-MI, ID-TLR) were not significantly different between devices Angina, all revascularization and ID-TVR were similar between devices No statistically significant differences in device thrombosis were present 33

Summary and Conclusions (2) The ABSORB III trial has demonstrated safety and efficacy of Absorb BVS at 1 year in patients with stable CAD and stabilized ACS Longer term evaluation is ongoing to determine if ABSORB improves late outcomes compared to Xience 34