Reducing Stroke in Patients with AF: Paradigm Shift with LAA Closure – What does the future hold? Sameer Gafoor, MD Swedish Medical Center, Seattle WA Cardio Vascular Center, Frankfurt Germany

Unmet Need – the Paradigm Shift

OAT Discontinuation and Risk of Major Bleeds Treatment Study Drug Discontinuation Rate Major Bleeding (rate/year) Rivaroxaban1 24% 3.6% Apixaban2 25% 2.1% Dabigatran3 (150 mg) 21% 3.3% Edoxaban4 (60 mg / 30 mg) 33 % / 34% 2.8% / 1.6% Warfarin1-4 17 – 28% 3.1 – 3.6% 1 Connolly, S. NEJM 2009; 361:1139-1151 – 2 years follow-up (Corrected); 2 Patel, M. NEJM 2011; 365:883-891 – 1.9 years follow-up, ITT. 3 Granger, C NEJM 2011; 365:981-992 – 1.8 years follow-up. 4 Giugliano, R. NEJM 2013; 369(22): 2093-2104 – 2.8 years follow-up.

Estimate of persistence to OAC therapy following AF diagnosis is <80% at 1 year Real world discontinuation rates consistent with trials, even after accounting for switching between OACs Martinez, et al. Thromb Haemost 2015; 114: In press. 4 N

Trials and Patient Population: the Paradigm Shift

WATCHMAN is the most studied LAAC Device- Most Patients and Only One with Long-term Clinical Data Key Trials N Highlights PROTECT AF1 (2005-2008) 707 Prospective, randomized 2:1, non-inferiority trial of LAA closure vs. warfarin. CAP2 (2008-2010) 566 Prospective registry allowing continued access to the WATCHMAN Device and gain further information prior to PMA approval. PREVAIL3 (2010-2012) 407 Prospective, randomized 2:1, non-inferiority trial to collect additional information on the WATCHMAN Device. (2012-2014) 579 Prospective registry allowing continued access to the WATCHMAN Device prior to PMA approval. Total patients >2,000 ~6,000 Patient-Years of Follow-up 1 Reddy, et al. JAMA. 2014 ;312(19): 1988-1998. 2 Reddy VY et al. Circulation. 2011; 123:417-424. 3 Holmes et al., JACC 2014,;4(1): 1-11.

Patient Risk Factors Across Trials Characteristic PROTECT AF N=707 PREVAIL N=407 CAP N=566 CAP2 N=579 p-value CHADS2 Score 2.2 ± 1.2 2.6 ± 1.0 2.4 ± 1.2 2.7 ± 1.1 <.0001 CHADS2 Risk Factors (% of Patients) CHF 26.9 19.1 23.3 27.1 0.004 Hypertension 89.8 88.8 91.4 92.5 0.15 Age ≥ 75 43.1 51.8 53.6 59.7 <0.001 Diabetes 26.2 24.9 32.4 33.7 0.001 Stroke/TIA 18.5 30.4 27.8 29.0 <0.0001 CHA2DS2-VASc 3.5 ± 1.6 4.0 ± 1.2 3.9 ± 1.5 4.5 ± 1.3 Holmes, DR et al. J Am Coll Cardiol. 2015;65(24):2614-2623.

Majority of patients were at a high stroke risk, and all were eligible for anti-coagulation Anticoagulation Eligible1 High Risk1 CHA2DS2-VASc Score ≥22 93% PROTECT AF CAP PREVAIL CAP2 96% 100% Patients (%) CHA2DS2-VASc Score AHA/ACC/HRS Guidelines (2014); Holmes, DR et al. J Am Coll Cardiol. 2015;65(24):2614-2623.

Paradigm Shift: Increase in Safety and Efficacy Evidence

Procedure Safety: 7-Day Event Rates Patients with Safety Event (%) PROTECT AF N=232 N=231 N=566 N=269 N=579 N=1021 1st Half 2nd Half All Device and/or procedure-related serious adverse events within 7 Days. WATCHMAN FDA Panel Sponsor Presentation. Oct 2014. Boersma LVA et al., EurHeartJ 2016; In Press.

PROTECT AF/PREVAIL Meta-Analysis: WATCHMAN Comparable to Warfarin HR p-value Efficacy 0.79 0.22 All stroke or SE 1.02 0.94 Ischemic stroke or SE 1.95 0.05 Hemorrhagic stroke 0.004 Ischemic stroke or SE >7 days 1.56 0.21 CV/unexplained death 0.48 0.006 All-cause death 0.73 0.07 Major bleed, all 1.00 0.98 Major bleeding, non procedure-related 0.51 0.002 Favors WATCHMAN   Favors warfarin Hazard Ratio (95% CI) Holmes, DR et al. J Am Coll Cardiol. 2015;65(24):2614-2623.

Warfarin Ischemic Stroke Rate in PREVAIL Differs from Other Trials Trial (Warfarin Arm) Ischemic Stroke Rate per 100 pt-yrs Mean CHADS2 PREVAIL1 2.6 PROTECT AF1 2.2 RE-LY2 2.1 ROCKET AF2 3.5 ARISTOTLE2 ENGAGE3 2.8 Rate per Patient-years 1 WATCHMAN FDA Panel Sponsor Presentation. Oct 2014. 2 Miller. AJC (2012) 3 Giugliano. NEJM (2013).

Ischemic Stroke Rate Aligns with Expected Rate Based on Risk Score Untreated AF Treated with Anticoagulants WATCHMAN Arm Ischemic Stroke Risk (events per 100 pt-yrs) PREVAIL PROTECT AF CAP2 CAP Baseline CHA2DS2-VASc Score Friberg. Eur Heart J (2012); NICE UK (2014). WATCHMAN FDA Panel Sponsor Presentation. Oct 2014

Warfarin Cessation: Majority of patients are able to stop long-term warfarin therapy Study 45-day 12-month PROTECT AF 87% >93% CAP 96% >96% PREVAIL 92% >99% Reddy, VY et al. Cardiostim 2014 (Abstract). CAP2 Follow-up too limited to report

72% reduction in bleeding events > 6 months post procedure Landmark Analysis: Freedom of Major Bleeding Over 3 Adjunctive Pharmacotherapy Intervals 72% reduction in bleeding events > 6 months post procedure Price MJ et al., J Am Coll Cardiol Interv 2015, In Press

What did we learn? Device procedure is safe across the spectrum of implant experience In clinical trials, WATCHMAN provided comparable results to warfarin for the prevention of all-stroke events Superiority for hemorrhagic stroke and CV mortality Consistent efficacy over entire device experience Additional reductions in bleeding events after warfarin and clopidogrel therapy

Paradigm Shift: Reimbursement and Coverage

CMS NCD: Highlights Eligible patients must have a CHADS2 score ≥2 or a CHA2DS2-VASc score ≥3 Documented evidence of a formal shared decision interaction between the patient and an independent, non-interventional physician Patients must be suitable for short-term warfarin, but deemed unable to take long-term oral anticoagulation   Procedure must be performed: in a hospital with an established structural heart disease or electrophysiology program. by an IC or EP meeting the following criteria: Trained by the manufacturer ≥25 interventional cardiac procedures involving transseptal punctures through an intact septum Continues to perform  ≥25 interventional cardiac procedures involving transseptal punctures through an intact septum, with at least 12 being LAAC over a two year period Patients must be enrolled in a prospective national registry

Cost Effectiveness of LAAC versus warfarin and NOACs Reddy, V.Y. et al. J Am Coll Cardiol. 2015; 66(24):2728–39.

Paradigm Shift: Future Device and Clinical Trials

WATCHMAN FLX™ LAA Closure Device Next Gen Design Goals

WATCHMAN FLX™ LAA Closure Device Next Gen Design Goals

ASAP-TOO Study Design Prospective, randomized, multi-center, global Patients with non-valvular atrial fibrillation deemed not suitable for oral anti-coagulation therapy to reduce the risk of stroke. Randomized 2:1 (WATCHMAN vs Control) Considering Group Sequential Design Allows early looks; potential to stop early for benefit 900 subjects at up to 100 global sites Follow-Up* 45 Day with TEE 6,18 month phone visit 12 month with TEE Years 2-5 annually * Brain imaging required at baseline if prior stroke or TIA * Brain imaging required at baseline if prior stroke or TIA

ASAP-TOO Reasons for OAC Unsuitability History of overt bleeding related or unrelated to oral anticoagulants: Prior history of intracranial or subdural hemorrhage. Other major organ bleeding including: gastrointestinal, genitourinary, ocular, spinal, pulmonary retroperitoneal, pericardial, or ENT Epistaxis requiring ER visit, hospitalization or physician intervention Increased risk of bleeding or bleeding tendencies: Gastrointestinal lesions Inflammatory bowel disease, peptic ulcer disease History of falls/seizures and with a likelihood of recurrent falls Cerebral amyloid angiopathy Significant thrombocytopenia (level TBD, consider < 50 x 109/L) Need for long term dual antiplatelet therapy HAS-BLED ≥3 Contraindications to warfarin and/or DOAC Severe renal failure (GFR <30) Allergy to the above agents Other reasons: Lifestyle or occupational bleeding risk, or Poor control on warfarin (time in therapeutic range <50%) and intolerance to the NOACs, or Patient refusal or inability to take oral anticoagulants, or Other medical or social reasons that make OATs unsuitable

ASAP -TOO Endpoints Primary Effectiveness Endpoint The occurrence of ischemic stroke and systemic embolism at 24-months. Primary Safety Endpoint Same as historical WATCHMAN studies

ASAP -TOO Inclusion Criteria The patient has documented paroxysmal, persistent, permanent or long-term/longstanding persistent non-valvular atrial fibrillation (i.e., the patient has not been diagnosed with rheumatic mitral valvular heart disease) The patient has a calculated CHA2DS2-VASc score of 2 or greater The patient is deemed by their physician to be unsuitable for oral anticoagulation  The patient is deemed by their physician to be suitable for the defined protocol pharmacologic regimen of aspirin and clopidogrel* therapy following WATCHMAN FLX Closure Device implant. *Ticagrelor may be used in place of clopidogrel if patient has other indication  Initial

Thank You