1. ADAPT-DES Assessment of Dual AntiPlatelet Therapy with Drug-Eluting Stents A Large-Scale, Prospective, Multicenter Registry Examining the Relationship Between Platelet Responsiveness and Stent Thrombosis After DES Implantation Gregg W. Stone, MD Columbia University Medical Center NewYork-Presbyterian Hospital Cardiovascular Research Foundation

2. Disclosure Statement of Financial Interest Consulting Fees/Honoraria Abbott Vascular, Boston Scientific, Medtronic, Volcano, The Medicines Company, Daiichi Sankyo, Eli Lilly 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 RelationshipCompany

3. Although prior studies have shown a correlation between platelet hyporesponsiveness to ADP antagonists and stent thrombosis, all have been small to moderate in size. As such, several important questions remain unanswered:  What proportion of the risk of ST at different times after stent implantation can be attributed to platelet ADP antagonist response, and how useful is this to reclassify the risk of ST?  What is the optimal cutoff for platelet reactivity to predict stent thrombosis?  Is ADP antagonist hyporesponsiveness important in all pts? (e.g. non-diabetics as well as diabetics; stable CAD vs. ACS) ADAPT-DES: Background I

4. ADAPT-DES: Background II Prior studies have emphasized the absolute level of platelet activation/aggregation to ADP antagonists  The role of the baseline level of platelet activation and % platelet inhibition to ADP antagonists have largely been unstudied The impact of 1) platelet hyporesponsiveness to aspirin, and 2) overall platelet aggregation on DAPT on the risk of ST has been incompletely studied

5. ADAPT-DES Assessment of Dual AntiPlatelet Therapy with Drug-Eluting Stents Up to 11,000 pts prospectively enrolled No clinical or anatomic exclusion criteria 11 sites in US and Germany Clinical FU at 30 days, 1 year and 2 years Angio core lab assessment all STs w/1:2 matching controls Assess platelet function after adequate DAPT loading and GPI washout: Accumetrics VerifyNow Aspirin, VerifyNow P2Y12, and VerifyNow IIb/IIIa assays (results blinded) PCI with ≥1 non-investigational DES Successful and uncomplicated (IVUS/VH substudy; Up to 3000 pts enrolled) clinicaltrials.gov NCT00638794

6. DAPT Loading and GPI Washout for VerifyNow Assessment Post-PCI ADAPT-DES: DAPT Loading and GPI Washout for VerifyNow Assessment Post-PCI Aspirin loading: Aspirin loading: Pre-PCI mandatory: ≥300 mg non EC oral aspirin ≥6 hours prior to PCI or 324 mg chewed or ≥250 mg IV aspirin at least 30 minutes prior to PCI. Clopidogrel loading: Pre-PCI recommended, but in all cases 600 mg ≥6 hours or 300 mg ≥12 hours prior to VerifyNow, or ≥75 mg for ≥5 days prior to VerifyNow. Clopidogrel loading: Pre-PCI recommended, but in all cases 600 mg ≥6 hours or 300 mg ≥12 hours prior to VerifyNow, or ≥75 mg for ≥5 days prior to VerifyNow. GP IIb/IIIa inhibitor washout: GP IIb/IIIa inhibitors may be used per standard of care. If used, eptifibatide or tirofiban must have been discontinued for ≥24 hrs prior to VerifyNow, and abciximab must have been discontinued for ≥10 days prior to VerifyNow. GP IIb/IIIa inhibitor washout: GP IIb/IIIa inhibitors may be used per standard of care. If used, eptifibatide or tirofiban must have been discontinued for ≥24 hrs prior to VerifyNow, and abciximab must have been discontinued for ≥10 days prior to VerifyNow.

7. ADAPT-DES: Study organization Principal investigator:Gregg W. Stone ( & Chuck Simonton prior to joining AVD) Co-principal investigators:Thomas Stuckey, Bruce Brodie, Mike Rinaldi Pharmacology committee:Paul Gurbel and Steve Steinhubl Sponsor (IDE):Cardiovascular Research Foundation Site management & monitoring:R. Stuart Dickson Institute For Health Studies Michael Dulin, director, Sherry Laurent, consultant Data management:R. Stuart Dickson Institute For Health Studies Susan Christopher, project lead Event adjudication:Cardiovascular Research Foundation Roxana Mehran and Ecaterina Cristea, directors Angio and IVUS core labs:Cardiovascular Research Foundation Ecaterina Cristea and Akiko Maehara, directors Biostatistics:Cardiovascular Research Foundation Helen Parise, director Financial support:Boston Scientific, Abbott Vascular, Medtronic, Cordis, Biosensors, The Medicines Company, Daiichi-Sankyo, Eli Lilly, Volcano, Accumetrics

8. ADAPT-DES: Sites and enrollment 8,575 pts were enrolled at 11 sites between 1/7/2008 and 9/16/2010; 2,158 pts were enrolled in the IVUS substudy SitePrincipal investigator(s)N enrolled Charité Benjamin FranklinBernhard Witzenbichler1,426 Columbia University Medical CenterGiora Weisz1,369 Herz-Zentrum Bad KrozingenFranz-Josef Neumann1,035 Carolinas Medical CenterMike Rinaldi1,110 Wellmont Holstein ValleyChris Metzger790 Minneapolis Heart InstituteTim Henry and Ivan Chavez788 Lehigh Valley HospitalDavid Cox673 Firsthealth Moore RegionalPeter Duffy544 LeBauer CV ResearchBruce Brodie, Tom Stuckey534 Ohio State UniversityErnest Mazzaferri304 Indiana Heart InstituteJim Hermiller2

9. ADAPT-DES: Baseline features (n=8,575) Age (years)63.6 ± 10.9 Female25.9% Non-caucasian11.4% Diabetes mellitus32.4% - Insulin-treated11.6% Hypertension79.6% Hyperlipidemia74.4% Cigarette smoking, current22.6% Prior MI25.2% Prior PCI42.8% Prior CABG17.1% Prior CHF8.1% Prior PAD10.2% History of renal insufficiency7.7% - Dialysis1.6% BMI29.5 ± 5.7

10. Presentation during PCI - Stable CAD48.3% - ACS51.7% - UA, biomarker negative27.7% - NSTEMI14.5% - STEMI9.5% Extent of CAD - 1 vessel disease38.3% - 2 vessel disease33.0% - 3 vessel disease28.7% - Left main disease3.0% LVEF (%)55.0 ± 14.1 LVEDP (mmHg)16.7 ± 9.3 ADAPT-DES: Baseline features (n=8,575)

11. ADAPT-DES: Anti-platelet agents (n=8,575) Aspirin - Pre-admission82.0% - Loading dose pre-PCI88.7% - Discharge99.2% Thienopyridine - Pre-admission42.8% - Loading dose pre-PCI86.4% - Discharge99.7% Ticlopidinen=3 (0.04%) Clopidogreln=8,541 (99.7%) Prasugreln=26 (0.3%)

12. ADAPT-DES: PCI procedure (n=8,575) N vessels treated per pt1.2 ± 0.4 - LM3.7% - LAD46.0% - LCX30.9% - RCA37.0% - bypass graft3.3% N lesions treated per pt1.8 ± 1.1 N stents per pt1.7 ± 1.0 Total stent length (mm) 32.4 ± 22.3 DES type used per pt / lesion - Xience V / Promus64.4% / 58.3% - Taxus (Express, Liberté)16.5% / 14.4% - Cypher13.5% / 13.0% - Endeavor6.2% / 5.2% - Resolute2.2% / 2.1% - Other0.2% / 0.2% N = 10,091 vessels, 12,898 lesions

13. ADAPT-DES: Platelet function test results (n=8,575) Post-PCI to VerifyNow (hrs)19.0 [16.3, 21.8] VerifyNow Aspirin (ARU)419 ± 55 - ≥ 550 ARU*5.6% VerifyNow P2Y12 (BASE) 310 ± 58 VerifyNow P2Y12 (PRU)188 ± 97 - > 208 PRU*42.7% - ≥ 230 PRU*35.0% VerifyNow P2Y12 Inhibition (%) 40.0 ± 28.3 VerifyNow IIb/IIIa PAU193 ± 53 *Pre-specified cut-off values

14. ADAPT-DES: Stent Thrombosis Within 30 Days Definite or probable 0.46% (39) - Definite 0.32% (27) - Probable 0.14% (12) Days to definite or probable stent thrombosis Stent thrombosis (ARC def/prob) occurred in 39 (0.46%) pts Probable Frequency 0 1 2 3 4 5 6 0123456789101112131415161718192021222324252627282930 Definite

15. Rates are KM estimates (n). VerifyNow testDef/prob ST No def/prob ST P (n=39)(n=8536) Aspirin ARU 425.6 ± 60.1 419.2 ± 55.30.46 - ARU ≥5507.7%5.6%0.57 P2Y12 Base 301.7 ± 63.9309.6 ± 58.10.41 P2Y12 PRU249.4 ± 88.5187.6 ± 96.70.0001 - PRU >208 74.4%42.6%0.0002 - PRU ≥230 64.1%34.9%0.0003 P2Y12 % Inhibition19.8 ± 23.740.1 ± 28.2<0.0001 - Inhibition ≤11% 51.3%19.9% <0.0001 IIb/IIIa PAU 188.2 ± 54.9192.7 ± 53.40.60 ADAPT-DES: Relationship between VerifyNow platelet response to DAPT and subsequent definite or probable stent thrombosis

16. VerifyNow test Def/prob ST Definite ST AUCCut-offAUCCut-off Aspirin ARU0.5634030.626403 P2Y12 Base 0.5362890.604303 P2Y12 PRU0.6792060.716230 P2Y12 % Inhibition0.72025%0.78711% IIb/IIIa PAU0.5421950.587181 ADAPT-DES: ROC curve analysis of the relationship between VerifyNow assessed platelet response to DAPT and subsequent stent thrombosis

17. ADAPT-DES: Relationship Between VerifyNow P2Y12 PRU and Stent Thrombosis within 30 Days Definite or probable stent thrombosis Definite/Probable ST (%) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 Days 051015202530 1691166516631640 1688165716571630 1705167716761656 1666163316311607 1691166216591635 Number at Risk Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 0.18% 0.24% 0.36% 0.79% 0.78% P2Y12 PRU Q1 (≤94) (n=1691) P2Y12 PRU Q2 (95-160) (n=1701) P2Y12 PRU Q3 (161-216) (n=1705) P2Y12 PRU Q4 (217-275) (n=1666) P2Y12 PRU Q5 (≥276) (n=1691)

18. ADAPT-DES: Relationship Between VerifyNow P2Y12 PRU and Stent Thrombosis within 30 Days Definite or probable stent thrombosis Definite/Probable ST (%) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 Days 051015202530 3607354035343482 4834475447524686 Number at risk >208 PRU ≤208 PRU 0.81% 0.81% 0.21% 0.21% P2Y12 PRU > 208 (n=3607) P2Y12 PRU ≤ 208 (n=4834) P <0.001 HR [95% CI]= 3.89 [1.90, 7.98]

19. ADAPT-DES: Relationship Between VerifyNow P2Y12 % Inhibition and Stent Thrombosis within 30 Days Definite or probable stent thrombosis Definite/Probable ST (%) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 Days 051015202530 1694166716631637 16721647 1627 1676164116391613 17441712 1692 1653162616241598 Number at Risk Quintile 1 Quintile 2 Quintile 3 Quintile 4 Quintile 5 1.19% 0.42% 0.17% 0.12% P2Y12 % Q1 (≤11) (n=1694) P2Y12 % Q2 (12-28) (n=1672) P2Y12 % Q3 (29-46) (n=1676) P2Y12 % Q4 (47-68) (n=1744) P2Y12 % Q5 (≥69) (n=1626)

20. ADAPT-DES: Relationship Between VerifyNow P2Y12 % Inhibition and Stent Thrombosis within 30 Days Definite or probable stent thrombosis Definite/Probable ST (%) 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0 Days 051015202530 1694166716631637 6745662666226530 Number at risk ≤11% >11% 1.19% 0.29% VerifyNow P2Y12 % Lowest Quintile (≤11) (n=1694) VerifyNow P2Y12 % Highest 4 Quintiles (>11) (n=6745) P<0.001 HR [95% CI]= 4.18 [2.23, 7.82]

21. ADAPT-DES: Multivariable (Cox PHR) models of 30-day stent thrombosis stratified by propensity quintiles VerifyNow testN N Adj. HR*PAttributableAttributable P2Y12 at risk events[95%CI]valueeventspercent P2Y12 PRU >208 † 8439393.000.00519.349.6% [1.39, 6.49][8.1, 24.5][20.7%, 62.9%] P2Y12 PRU ≥230 † 8439392.750.00515.940.8% [1.35, 5.60][6.4, 20.5][16.5%, 52.7%] Inhibition ≤11% † 8437392.780.00312.832.8% [1.43, 5.40][6.0, 16.3][15.4%, 41.8%] Definite or probable stent thrombosis † Pre-specified measures *Adjusted for non-ACS vs NSTEMI vs STEMI, diabetes vs no diabetes, and stent length Model c-statistics = 0.609, 0.591, 0.623

22. ADAPT-DES: Multivariable (Cox PHR) models of 30-day stent thrombosis stratified by propensity quintiles VerifyNow testN N HRPAttributableAttributable at risk events[95%CI]valueeventspercent P2Y12 PRU >2088439393.000.00519.349.6% [1.39, 6.49][8.1, 24.5][20.7%, 62.9%] P2Y12 PRU ≥2308439392.750.00515.940.8% [1.35, 5.60][6.4, 20.5][16.5%, 52.7%] Inhibition ≤11%8437392.780.00312.832.8% [1.43, 5.40][6.0, 16.3][15.4%, 41.8%] ARU ≥5508517391.690.391.23.1% [0.51, 5.61][-2.9, 2.5][-7.4%, 6.3%] P2Y12 Base ≥3608436391.180.701.43.6% [0.50, 2.80][-9.0, 5.8][-23.0%, 14.8%] GPIIb/IIIa ≥238 8265380.660.37-3.1-8.1% [0.27, 1.64][-16.5, 2.3][-43.5%, 6.1%] Definite or probable stent thrombosis

23. ADAPT-DES: Multivariable (Cox PHR) models of 30-day stent thrombosis stratified by propensity quintiles VerifyNow testN N Adj. HR*PAttributableAttributable P2Y12 at risk events[95%CI]valueeventspercent † P2Y12 PRU >208 † 8439275.360.00217.966.3% [1.89, 15.21][10.4, 20.6][38.3%, 76.1%] † P2Y12 PRU ≥230 † 8439274.460.00114.754.6% [1.80, 11.03][8.5, 17.3][31.4%, 64.0%] † Inhibition ≤11% † 8437274.600.000313.349.3% [2.01, 10.55][8.5, 15.4][31.6%, 57.0%] Definite stent thrombosis † Pre-specified measures *Adjusted for non-ACS vs NSTEMI vs STEMI, diabetes vs no diabetes, and stent length Model c-statistics = 0.753, 0.721, 0.722

24. ADAPT-DES: Multivariable (Cox PHR) models of 30-day stent thrombosis stratified by propensity quintiles VerifyNow testN N HRPAttributableAttributable at risk events[95%CI]valueeventspercent P2Y12 PRU >2088439275.360.00217.966.3% [1.89, 15.21][10.4, 20.6][38.3%, 76.1%] P2Y12 PRU ≥230 8439274.460.00114.754.6% [1.80, 11.03][8.5, 17.3][31.4%, 64.0%] Inhibition ≤11%8437274.600.000313.349.3% [2.01, 10.55][8.5, 15.4][31.6%, 57.0%] ARU ≥5508517272.870.092.07.2% [0.84, 9.82][-0.6, 2.7][-2.1%, 10.0%] P2Y12 Base ≥3608436270.990.990.0-0.1% [0.33, 2.99][-10.2, 3.3][-37.7%, 12.3%] GPIIb/IIIa ≥238 8265270.430.18-3.9-14.5% [0.13, 1.49][-20.8, 1.0][-77.0%, 3.7%] Definite stent thrombosis

25. ADAPT-DES: ADAPT-DES: Predictive accuracy of VerifyNow testing – all pts (n=8,575) VerifyNow test SensitivitySpecificityPPVNPVAccuracy ASA ARU > 5507.7%94.4%0.6%99.6%94.0% P2Y12 PRU > 20874.4%57.4%0.8%99.8%57.5% P2Y12 PRU ≥ 23064.1%65.1%0.8%99.7%65.1% P2Y12 % Inhibition ≤ 11%51.3%80.1%1.2%99.7%79.9% IIb/IIIa PAU ≥ 23815.8%80.2%0.4%99.5%79.9% Definite or probable stent thrombosis by 30 days (n=39) VerifyNow test SensitivitySpecificityPPVNPVAccuracy ASA ARU > 55011.1%94.4%0.6%99.7%94.1% P2Y12 PRU > 20881.5%57.4%0.6%99.9%57.5% P2Y12 PRU ≥ 23070.4%65.1%0.6%99.9%65.1% P2Y12 % Inhibition ≤ 11%63.0%80.1%1.0%99.9%80.0% IIb/IIIa PAU ≥ 23811.1%80.1%0.2%99.6%79.9% Definite stent thrombosis by 30 days (n=27)

26. ADAPT-DES: ADP Platelet Responsiveness in Pts with and without Definite/Probable Stent Thrombosis within 30 Days Median [IQR] 188 [112, 260] Median [IQR] 252 [206, 311] P=0.0001 N=8402N=39 No Stent Thrombosis Stent Thrombosis 0 100 200 300 400 500 600 VerifyNow P2Y12 (PRU) N=8400N=39 No Stent Thrombosis Stent Thrombosis 0 Median [IQR] 38 [16, 62] P<0.0001 Median [IQR] 11 [0, 36] 25 50 75 100 VerifyNow P2Y12 PERCENT (%)

27. ADAPT-DES: Relationship between ACS and stent thrombosis P=0.002 9/414030/443510/23777/124613/812

28. ADAPT-DES: Relationship between ACS and VerifyNow response to DAPT VerifyNow testACSNo ACSP (n=4435)(n=4140) Aspirin ARU 419.5 ± 54.4 419.0 ± 56.4 0.66 - ARU ≥5505.4%5.8%0.43 P2Y12 Base 304.6 ± 57.6314.8 ± 58.1<0.0001 P2Y12 PRU193.8 ± 96.2181.7 ± 97.0<0.0001 - PRU >208 45.6%39.7% <0.0001 - PRU ≥230 37.6%32.3%<0.0001 P2Y12 % Inhibition37.3 ± 28.242.9 ± 28.1<0.0001 - Inhibition ≤11% 23.3%16.6% <0.0001 IIb/IIIa PAU 187.9 ± 52.3197.8 ± 54.1<0.0001

29. ADAPT-DES: Multivariable (Cox PHR) models of 30-day stent thrombosis stratified by propensity quintiles VerifyNow testN N Adj. HR*PAttributableAttributable P2Y12 at risk events[95%CI]valueeventspercent † PRU >208 † 4347303.910.00517.959.5% [1.51, 10.11][8.1, 21.6][27.0%, 72.1%] † PRU ≥230 † 4347 302.950.0113.244.1% [1.29, 6.77][4.5, 17.0][14.9%, 56.8%] † Inhibition ≤11% † 4346303.530.00112.240.6% [1.66, 7.52][6.8, 14.7][22.6%, 49.1%] ACS: Definite or probable stent thrombosis † Pre-specified measures *Adjusted for diabetes vs no diabetes and stent length Model c-statistics = 0.541, 0.464, 0.524

30. ADAPT-DES: Multivariable (Cox PHR) models of 30-day stent thrombosis stratified by propensity quintiles VerifyNow testN N Adj. HR* PAttributableAttributable P2Y12 at risk events[95%CI]valueeventspercent † PRU >208 † 409291.490.591.618.3% [0.35, 6.36][-9.3, 4.2][-103.0%, 46.8%] † PRU ≥230 † 4092 92.020.352.528.0% [0.46, 8.74][-5.8, 4.4][-64.0%, 49.2%] † Inhibition ≤11% † 409192.220.281.618.3% [0.53, 9.28][-2.7, 2.7][-29.6%, 29.7%] No ACS: Definite or probable stent thrombosis † Pre-specified measures *Adjusted for diabetes vs no diabetes and stent length Model c-statistics = 0.704, 0.705, 0.760

31. Conclusions and Implications ADAPT-DES: Conclusions and Implications The absolute and relative levels of platelet inhibition to ADP antagonists as assessed by the VerifyNow P2Y12 test are powerful independent predictors of stent thrombosis within 30 days, with a significant proportion of events independently attributable to clopidogrel hyporesponsiveness. The absolute and relative levels of platelet inhibition to ADP antagonists as assessed by the VerifyNow P2Y12 test are powerful independent predictors of stent thrombosis within 30 days, with a significant proportion of events independently attributable to clopidogrel hyporesponsiveness. In contrast, the Base level of platelet P2Y12 response, as well as aspirin and overall platelet responsiveness after DAPT loading as assessed by VerifyNow were not shown to be related to the 30-day rate of stent thrombosis. In contrast, the Base level of platelet P2Y12 response, as well as aspirin and overall platelet responsiveness after DAPT loading as assessed by VerifyNow were not shown to be related to the 30-day rate of stent thrombosis.

32. Conclusions and Implications ADAPT-DES: Conclusions and Implications These data suggest that agents which more effectively inhibit ADP-induced platelet activation should reduce 30-day stent thrombosis when applied to large patient populations (underlying the positive findings of TRITON-TIMI 38 and PLATO). These data suggest that agents which more effectively inhibit ADP-induced platelet activation should reduce 30-day stent thrombosis when applied to large patient populations (underlying the positive findings of TRITON-TIMI 38 and PLATO). However, the modest sensitivity and specificity of platelet function testing, coupled with the low prevalence of events, implies that testing of platelet ADP antagonist responsiveness is unlikely to provide useful information to guide clinical decision-making in most individual patients for the prevention of stent thrombosis at 30 days. However, the modest sensitivity and specificity of platelet function testing, coupled with the low prevalence of events, implies that testing of platelet ADP antagonist responsiveness is unlikely to provide useful information to guide clinical decision-making in most individual patients for the prevention of stent thrombosis at 30 days.

33. The degree of platelet responsiveness to ADP antagonist loading is useful to predict 30-day stent thrombosis in diabetic and non-diabetic patients, as well as those with ACS, but may have less clinical utility in patients with stable CAD. The degree of platelet responsiveness to ADP antagonist loading is useful to predict 30-day stent thrombosis in diabetic and non-diabetic patients, as well as those with ACS, but may have less clinical utility in patients with stable CAD. The very low stent thrombosis rate in pts with stable CAD, coupled with the poor prognostic utility of platelet function testing in this setting suggests that assessing DAPT response in pts without ACS undergoing PCI is unlikely to provide incremental clinical utility, and may explain the negative results of trials such as GRAVITAS and TRIGGER-PCI. The very low stent thrombosis rate in pts with stable CAD, coupled with the poor prognostic utility of platelet function testing in this setting suggests that assessing DAPT response in pts without ACS undergoing PCI is unlikely to provide incremental clinical utility, and may explain the negative results of trials such as GRAVITAS and TRIGGER-PCI. Conclusions and Implications ADAPT-DES: Conclusions and Implications

34. The relationship between platelet responsiveness testing and the occurrence of late and very late stent thrombosis (in patients who have maintained and discontinued DAPT) will be assessed during the 2-year clinical follow-up phase of the ADAPT- DES study. The relationship between platelet responsiveness testing and the occurrence of late and very late stent thrombosis (in patients who have maintained and discontinued DAPT) will be assessed during the 2-year clinical follow-up phase of the ADAPT- DES study. Conclusions and Implications ADAPT-DES: Conclusions and Implications