on behalf of the ABSORB II Investigators The 4-year Clinical Outcomes of the ABSORB II Trial: First Randomized Comparison between the Absorb Everolimus Eluting Bioresorbable Vascular Scaffold and the XIENCE Everolimus Eluting Stent Bernard Chevalier Institut Jacques Cartier, Massy, France Patrick W. Serruys Imperial College, London, UK Erasmus University MC, Netherlands on behalf of the ABSORB II Investigators

Presentor Disclosures Bernard Chevalier was a Consultant for Abbott Vascular and is currently consultant for Biotronik, Colibri, Cordis, Medtronic, Terumo. Patrick Serruys is a Member of the International Advisory Board of Abbott Vascular

Randomized 2:1 Absorb BVS:XIENCE / 46 sites (Europe and New Zealand) ABSORB II Study Design 501 subjects Randomized 2:1 Absorb BVS:XIENCE / 46 sites (Europe and New Zealand) Clinical Follow-Up 30d 6m 12m 24m 36m 48m 60m QoL follow-up Angio, IVUS follow-up MSCT follow-up (Absorb arm only)* Study Objective Randomized against XIENCE control. First Patient In: 28-Nov-2011 Co-primary Endpoints 36 months Vasomotion assessed by change in Mean Lumen Diameter between pre- and post-nitrate at 3 years (superiority) Minimum Lumen Diameter (MLD) at 3 years post nitrate minus MLD post procedure post nitrate (non-inferiority, reflex to superiority) Treatment Up to 2 de novo lesions in different epicardial vessels Planned overlapping allowed in lesions ≤ 48 mm Device Sizes Device diameters: 2.5, 3.0, 3.5 mm Device lengths: 12 (3.5 mm diameter only), 18, 28 mm The ABSORB II study is sponsored by Abbott Vascular. [NCT01425281]

Characteristics of Patients at Baseline Absorb 335 pts Xience 166 pts 95% CI  Age (year) mean ±SD 61.5 ± 10.0 60.9 ± 10.0 N.S. Male % 75.5 79.5 Current Tobacco Use % 23.6 21.7 Hypertension % 69.0 71.7 Dyslipidemia % 75.2 80.1 All Diabetes Mellitus % 23.9 24.1 Diabetes Mellitus Insulin Dependent % 6.6 8.4 Family History of Premature CAD % 36.6 41.3 Prior Intervention in Target Vessel % 11.7 8.9 Prior MI % 28.0 28.9 Stable Angina % 63.9 64.5 Unstable Angina % 20.3 22.3 Silent Ischemia % 12.5 11.4 Recent MI with normalized cardiac enzyme % 3.3 1.8

Procedural Assessment Pre and Post Procedure Absorb 364 Lesions Xience 182 Lesions p value Procedural Details Per Lesion Balloon dilatation prior to device implantation % 100 98.9 0.11 Planned overlap with the same type of device % 15.4 11.0 0.16 Unplanned/bailout implantation “same” % 3.8 6.0 0.25 Nominal size of study device mm 3.01 3.05 0.10 Balloon dilatation after device implantation % 60.7 58.8 0.67 Nominal diameter of last balloon used mm 3.08 3.16 0.02 Maximum last balloon pressure used atm 14.23 15.03 0.01 Acute recoil post device implantation mm 0.19 0.85 Acute Clinical Device Success % 99.2 0.55 Acute Clinical Procedural Success % 96.1 98.8 < <

20 patients did not sign protocol amendment for 4 & 5 year follow-up 4-Year Patient Flowchart 311* and 154** patients still in the study but 5 missed 3 yr FUP Intent To Treat N=501 Absorb BVS N=335 XIENCE N=166 Baseline 7 withdrawal 1 death 2 withdrawal N=328 1-year N=163 4 death 3 withdrawal 1 LFU 3 withdrawal N=320 2-year N=160 4 death 3 withdrawal 2 LFU, 3 MV, 5 death, 1 LFU 2 MV N=308* 3-year N=152** 2 death, 8 MV, 3 withdrawal 9 no consent, -3 still in study but missed 3yr FUP 1 death, 2 MV, 1 withdrawal 11 no consent ,-2 still in study but missed 3yr FUP N=289 (86%) 4-year N=139 (84%) At 4 years patients with missing visits (MV) were confirmed as alive and well by site PI. 20 patients did not sign protocol amendment for 4 & 5 year follow-up

Clinical Endpoints at 3 Years Hierarchical Absorb 325 patients Xience 161 patients Relative Risk p value Device-oriented composite endpoint [DOCE] 10.5% 5.0% 2·11 [1·00, 4·44] 0·04 Cardiac death 0.9% 1.9% 0·50 [0·10, 2·43] 0·4   Target vessel MI 6.5% (21) 1.2% (2) 5·20 [1·23, 21·91] 0·01 Periprocedural MI* 4.0%(13) 3.22 [0.74, 14.10] 0.16 Spontaneous MI* 2.5% (8) 0% (0) NC [NC] 0.06 Clinically indicated TLR 3.1% (10) 1.9% (3) 1·65 [0·46, 5·92] 0·56 Patient-oriented composite endpoint [POCE] 20.9% 24.2% 0·86 [0·61, 1·22] All-cause death 2.5% 3.7% 0·66 [0·23, 1·87] 0·57 Any MI 7.7% 3.1% 2.48 [0.97, 6.35] 0.048 Any revascularization 10.8% 17.4% 0·62 [0·39, 0·98] 0.04 * Per extend historical MI definition

Device oriented Composite Endpoint (DOCE)/ Target Lesion Failure (TLF) TLF per WHO (%) 5 10 15 20 25 Time Post Index Procedure (Days) 180 360 540 720 900 1080 1260 1440 1620 HR [95% CI]= 2.17 [1.01,4.69] p=0.043 (Log rank test) 10.4% 4.9% BVS XIENCE Source doc: Data package: Abs2_4Y_KM_Karine_20171003 DoCE/TLF : Cardiac death, target-vessel myocardial infarction, and clinically indicated target-lesion revascularisation (TLR)

Device oriented Composite Endpoint (DOCE)/ Target Lesion Failure (TLF) TLF per WHO (%) 5 10 15 20 25 Time Post Index Procedure (Days) 180 360 540 720 900 1080 1260 1440 1620 HR [95% CI]= 2.04 [0.98,4.24] p=0.050 (Log rank test) 11.1% 5.6% BVS XIENCE 5 10 15 20 25 1152 1260 1440 1620 HR [95% CI]= 1.44 [0.15,13.80] p=0.75 (Log rank test) 1.0% 0.7% Δ = 0.3% Source doc: Data package: Abs2_4Y_KM_Karine_20171003 DoCE/TLF : Cardiac death, target-vessel myocardial infarction, and clinically indicated target-lesion revascularisation (TLR)

Patient oriented Composite Endpoint (PoCE)/DMR POCE / DMR per WHO (%) 5 10 15 20 25 30 Time Post Index Procedure (Days) 180 360 540 720 900 1080 1260 1440 1620 HR [95% CI]= 0.85 [0.58,1.27] p=0.43 (Log rank test) 20.9% 24.2% BVS XIENCE Source doc: Data package: Abs2_4Y_KM_Karine_20171003 PoCE=DMR: All Death, all Myocardial infarction, and all Revascularisation

Patient oriented Composite Endpoint (PoCE)/DMR POCE / DMR per WHO (%) 5 10 15 20 25 30 Time Post Index Procedure (Days) 180 360 540 720 900 1080 1260 1440 1620 HR [95% CI]= 0.90 [0.61,1.33] p=0.60 (Log rank test) 22.9% 24.9% BVS XIENCE 5 10 15 20 25 30 1152 1260 1440 1620 HR [95% CI]= 3.14 [0.71,13.93] p=0.11 (Log rank test) 4.3% 1.4% Δ = 2.9% Source doc: Data package: Abs2_4Y_KM_Karine_20171003 PoCE=DMR: All Death, all Myocardial infarction, and all Revascularisation

4-year Clinical Outcomes Composite Endpoints Absorb BVS N=335 XIENCE N=166 p value PoCE (%) 23.6 26.7 0.4682 MACE (%) 12.4 8.0 0.1545 DoCE, TLF (%) 11.5 6.0 0.0628 TVF (%) 14.0 0.9970 PoCE (Patient oriented Composite Endpoint): All death, all myocardial infarction, and all revascularisation MACE (Major Adverse Cardiac Events): Cardiac death, all myocardial infarction, and clinically indicated target-lesion revascularisation (TLR) DoCE (Device oriented Composite Endpoint)/ TLF (Target Lesion Failure): Cardiac death, target-vessel myocardial infarction, and clinically indicated target-lesion revascularisation (TLR) TVF (Target Vessel Failure): Cardiac death, all myocardial infarction, clinically indicated target-vessel revascularisation (TVR)

Clinical Outcomes Non Hierarchical Events 3-4 years Absorb BVS N=335 XIENCE N=166 p value Death* n(%) 2(0.7) 1(0.7) 1.0000 Cardiac 1(0.3) 0.5408 Vascular 0(0.0) Non-cardiovascular Myocardial Infarction n(%) Q-wave Non Q-wave All Revascularization* n(%) 10(3.3) 0.1142 All TLR All NTL-TVR 4(1.3) 0.3102 All NTVR 6(2.0) 0.4373 *Per ARC. Cutlip et al., Circulation. 2007;115:2344-2351

Definite/Probable Scaffold/Stent Thrombosis* ARCST DPr (%) 5 10 15 20 25 Time Post Index Procedure (Days) 180 360 540 720 900 1080 1260 1440 1620 HR [95% CI]= NA [NA] p=0.033 (Log rank test) 2.8% 0.0% BVS XIENCE Source doc: Data package: Abs2_4Y_KM_Karine_20171003 *Per ARC. Cutlip et al., Circulation. 2007;115:2344-2351

Definite/Probable Scaffold/Stent Thrombosis* ARCST DPr (%) 5 10 15 20 25 Time Post Index Procedure (Days) 180 360 540 720 900 1080 1260 1440 1620 HR [95% CI]= NA [NA] p=0.033 (Log rank test) 2.8% 0.0% BVS XIENCE No stent/scaffold thrombosis between 3 and 4 years” 5 10 15 20 25 1152 1260 1440 1620 HR [95% CI]= NA [NA] p= (Log rank test) 0.0% Source doc: Data package: Abs2_4Y_KM_Karine_20171003 *Per ARC. Cutlip et al., Circulation. 2007;115:2344-2351

Scaffold/Stent Thrombosis 4 years Absorb BVS N=335 XIENCE N=166 p value Definite ST* 0-1488 days (%) 2.6 0.0 0.0583 Acute/sub-acute (0-30 days 0.6 1.0000 Late (31-365 days) Very late (365 – 14888 days) 1.8 0.1851 Very late between 3 – 4 year follow-up (N) NA Definite/Probable ST* 0-1488 days(%) 3.0 0.0347 Acute/sub-acute (0-30 days) 1.000 0.3 Very late (365 – 1488 days) *Per ARC. Cutlip et al., Circulation. 2007;115:2344-2351

Post-Procedure Usage of Antiplatelet Medication through 4 years Absorb BVS N=335 XIENCE N=166 p value On Aspirin (%) at 1 year 95.8 95.2 0.7473 at 4 years 84.4 81.3 0.3794 On DAPT (%) 81.0 80.7 0.9357 at 2 years 28.6 28.9 0.9442 at 3 years 29.8 27.7 0.6254 25.9 21.1 0.2372

4-year Summary Excluding Subjects with ST Absorb BVS N= 326 XIENCE N= 166 p value Hierarchical PoCE*, % 21.3 26.7 0.2025 All death (Non-hierarchical) 3.0 4.7 0.3503 All MI (Non-hierarchical) 5.9 3.3 0.2398 All revascularization (Non-hierarchical) 15.1 22.7 0.0457 Hierarchical DoCE* or TLF, % 8.9 6.0 0.2893 Cardiac death (Non-hierarchical) 1.0 2.7 0.2254 TV-MI (Non-hierarchical) 4.6 1.3 0.0762 CI-TLR (Non-hierarchical) 3.9 2.0 0.4042 Per Protocol Myocardial Infarction (MI): Q wave MI Development of new, pathological Q wave on the ECG. Non-Q wave MI Elevation of CK levels to ≥ two times the upper limit of normal (ULN) with elevated CK-MB in the absence of new pathological Q waves. PoCE (Patient oriented Composite Endpoint)*: All death, all myocardial infarction, and all revascularisation. DoCE (Device oriented Composite Endpoint)*/ TLF (Target Lesion Failure): Cardiac death, target-vessel myocardial infarction, and clinically indicated target-lesion revascularisation (TLR). Source docs: Data package: Additional analysis -> Analysis excluding ST Abs2_Adhoc_T14-1_HiAE_WHO_EXCST_4Y Abs2_Adhoc15_NhiAE_EXCST_4Y *Per ARC. Cutlip et al., Circulation. 2007;115:2344-2351

4-year Summary of Subjects on DAPT at 4 Years Absorb BVS N= 86 XIENCE N= 35 p value Hierarchical PoCE*, % 37.2 42.9 0.5633 All death (Non-hierarchical) 0.0 1.0000 All MI (Non-hierarchical) 10.5 5.7 0.5076 All revascularization (Non-hierarchical) 33.7 0.3435 Hierarchical DoCE* or TLF, % 17.4 0.1472 Cardiac death (Non-hierarchical) TV-MI (Non-hierarchical) 0.0578 CI-TLR (Non-hierarchical) 12.8 0.3429 Per Protocol Myocardial Infarction (MI): Q wave MI Development of new, pathological Q wave on the ECG. Non-Q wave MI Elevation of CK levels to ≥ two times the upper limit of normal (ULN) with elevated CK-MB in the absence of new pathological Q waves. PoCE (Patient oriented Composite Endpoint)*: All death, all myocardial infarction, and all revascularisation. DoCE (Device oriented Composite Endpoint)*/ TLF (Target Lesion Failure): Cardiac death, target-vessel myocardial infarction, and clinically indicated target-lesion revascularisation (TLR). Source doc: Data package: Additional analysis -> Analysis for patients still on DAPT at 4Y Tables: Abs2_Adhoc_T14-1_HiAE_WHO_DAPT_4Y; Abs2_Adhoc15_NhiAE_DAPT_4Y *Per ARC. Cutlip et al., Circulation. 2007;115:2344-2351

Limitations The ABSORB II study was not powered for clinical endpoints Patients in the ABSORB II Study were enrolled before the current recommendations for scaffold implantation

Conclusions The Absorb scaffold polymer has been reported to be completely bio-resorbed by 3 years. Between 3 and 4 years follow up there were no ST events in the Absorb arm DOCE/TLF events were similar between Absorb and Xience In a trial which was not powered for clinical events, at 4 years there were no statistically significant differences in the clinical outcomes between the two arms: PoCE (all death, all MI and all revascularization) Absorb BVS: 23.6% vs XIENCE: 26.7%, p=0.47 DoCE/TLF (cardiac death, TV-MI and TLR) Absorb BVS: 11.5% vs XIENCE: 6.0%, p=0.06 The exploratory observations presented in this report are hypothesis generating and need to be confirmed in larger randomized trials such as ABSORB III and ABSORB IV