Platelet Glycoprotein Iib/Iiia Integrin Blockade with Eptifibatide in Coronary Stent Intervention the ESPRIT Trial: a Randomized Controlled Trial

ORIGINAL CONTRIBUTION

Platelet Glycoprotein IIb/IIIa Integrin Blockade With Eptifibatide in Coronary Stent Intervention The ESPRIT Trial: A Randomized Controlled Trial

J. Conor O’Shea, MD Context The Enhanced Suppression of the Platelet IIb/IIIa Receptor with Integrilin Gail E. Hafley, MS Therapy (ESPRIT) trial showed the efficacy of adjunctive, double-bolus eptifibatide therapy Sally Greenberg, PhD in reducing ischemic complications of nonurgent coronary stent implantation at 48 hours and at 30 days. Vic Hasselblad, PhD Objective To determine whether the beneficial effects of eptifibatide persist at 6 months Todd J. Lorenz, MD after treatment. Michael M. Kitt, MD Design Follow-up study of a randomized, double-blind, placebo-controlled, crossover- permitted trial conducted from June 1999 through February 2000. John Strony, MD Setting Ninety-two tertiary care centers in the United States and Canada. James E. Tcheng, MD Participants A total of 2064 patients scheduled to undergo nonurgent percutane- for the ESPRIT Investigators ous coronary intervention with stent implantation.

ORONARY THROMBOSIS IS CEN- Intervention Patients were randomly assigned to receive placebo or eptifibatide (two tral to the pathogenesis of ath- 180-µg/kg boluses 10 minutes apart and continuous infusion of 2.0 µg/kg per minute), started immediately before stent implantation and continued for 18 to 24 hours. Com- erosclerosis, acute myocar- plete follow-up data were available for 988 (95.0%) of 1040 patients given eptifi- dial infarction (AMI), acute batide and 977 (95.4%) of 1024 patients given placebo. coronaryC syndromes, and ischemic com- Main Outcome Measures Composite rates of death or myocardial infarction (MI); plications of percutaneous coronary in- 1 death, MI, or target vessel revascularization; and their individual components 6 months tervention (PCI). Aside from the obvi- after enrollment, compared between the 2 groups. ous and dramatic effects of abrupt thrombosis of an epicardial vessel, the ef- Results By 6 months, the composite end point of death or MI had occurred in 7.5% of eptifibatide-treated patients and in 11.5% of placebo-treated patients (hazard ratio [HR], fects of thrombosis on microcirculation 0.63; 95% confidence interval [CI], 0.47-0.84; P=.002). The composite of death, MI, or (such as microembolization and vaso- target vessel revascularization was 14.2% in eptifibatide-treated patients vs 18.3% in pla- spasm) also portend poor outcomes in cebo-treated patients (HR, 0.75; 95% CI, 0.60-0.93; P=.008). Most of this benefit accrued coronary artery disease.2 Several large, early (Ͻ48 hours after initiation of therapy) and was maintained through 6 months. Six- randomized clinical trials have shown month mortality in the eptifibatide group was 0.8% vs 1.4% in the placebo group (HR, that inhibition of platelet aggregation 0.56; 95% CI, 0.24-1.34; P=.19) and target vessel revascularization occurred in 8.6% of with platelet glycoprotein (Gp) IIb/IIIa the eptifibatide group vs 9.4% of the placebo group (HR, 0.91; 95% CI, 0.68-1.22; P=.51). receptor antagonists improves out- Conclusion Adjunctive eptifibatide therapy during coronary stent implantation pro- comes in patients presenting with acute vides benefit through 6-month follow-up. coronary syndromes and after PCI.3,4 JAMA. 2001;285:2468-2473 www.jama.com These agents mitigate the thrombotic cas- cade both by blocking Gp IIb/IIIa, the pri- adverse cardiac events in several coro- tion, and unstable angina.14 In the En- mary surface-membrane receptor re- nary indications, including PCI,12 AMI hanced Suppression of the Platelet IIb/ sponsible for platelet aggregation, and by with13 or without14 ST-segment eleva- IIIa Receptor with Integrilin Therapy indirectly inhibiting the generation of thrombin.5-11 Author Affiliations: Duke Clinical Research Institute, Therapeutics, the trial sponsor. Dr Tcheng receives re- Durham, NC (Drs O’Shea, Hasselblad, and Tcheng and search grant support and speakers bureau honoraria Eptifibatide is an intravenous, rap- Ms Hafley); COR Therapeutics Inc, South San Francisco, and serves as a consultant for COR Therapeutics and idly reversible, competitive inhibitor of Calif (Drs Greenberg, Lorenz, and Kitt); and Schering- Schering-Plough Research Institute. Plough Research Institute, Kenilworth, NJ (Dr Strony). Corresponding Author: James E. Tcheng, MD, Duke Gp IIb/IIIa. Therapy with eptifibatide Financial Disclosures: Drs Greenberg, Lorenz, and Kitt Clinical Research Institute, PO Box 17969, Durham, has been shown to reduce short-term are employed by and own stock and options in COR NC 27705 (e-mail: [email protected]).

Downloaded From: https://jamanetwork.com/ on 10/02/2021 EPTIFIBATIDE IN CORONARY STENTING

(ESPRIT) trial, high-dose, double- Ͻ100ϫ103/µL), or a serum creatinine 200 to 300 seconds. Treatment with bolus eptifibatide was evaluated in non- level of more than 4.0 mg/dL (350 ticlopidine or clopidogrel was permit- urgent coronary stenting. This trial was µmol/L). ted on the day of the procedure but not terminated early for efficacy in Febru- beforehand; the choice of loading dose ary 2000.15 Final analysis of the pri- Treatments was decided by the treating physician. mary composite end point—death, MI, Patients were randomized in a 1:1 ratio The PCI was performed according to lo- need for urgent target vessel revascu- to receive either placebo or eptifibatide cal standards, and any approved stent larization, or crossover to Gp IIb/IIIa (Integrilin, COR Therapeutics, South San could be implanted. inhibitor therapy for thrombosis within Francisco, Calif, and Schering-Plough To provide emergency, open-label Gp 48 hours—showed a 37% relative risk Research Institute, Kenilworth, NJ), IIb/IIIa inhibitor therapy (for direct treat- reduction with treatment (10.5% with started immediately prior to PCI. The ment of abrupt closure, no reflow, coro- placebo vs 6.6% with eptifibatide; randomization allocation code was gen- nary thrombosis, or similar complica- P=.002). At 30 days, the key second- erated using random permuted blocks tions of PCI), “bailout” kits were ary composite end point of death, MI, within each investigative site. Each site supplied to the sites. These kits con- or urgent target vessel revasculariza- kept a randomization schedule unique tained 2 bolus vials of either eptifi- tion was reduced by a relative 35% to that site. Study drug kits were pre- batide (for patients allocated to the pla- (10.4% vs 6.8%; P=.003).15 The pur- pared only after the guiding scout im- cebo group) or placebo (for patients pose of the present analysis was to de- ages had been obtained in the catheter- allocated to the eptifibatide group). Once termine whether these benefits contin- ization laboratory in preparation for the bolus bailout treatment had begun, ued at 6 months for patients enrolled intervention. Patients were considered the study drug infusion was discontin- in the ESPRIT trial. randomized at the time of any adminis- ued and open-label eptifibatide infu- tration of study drug. If the clinician de- sion was initiated. Bailout use of Gp IIb/ METHODS cided not to give the study drug for any IIIa inhibition did not result in Patients reason, the patient was not randomized unblinding of treatment assignment. The design and methods of the ESPRIT and was not considered part of the study. The protocol was approved by the in- trial have been described.15,16 Briefly, 92 If a kit was prepared but not used, un- stitutional review board at each clini- centers in the United States and Canada blinding did not occur and replace- cal site, and patients gave informed con- enrolled patients who were scheduled ment kits were used to reestablish the sent for participation in the trial and to undergo nonurgent coronary stent- correct sequence in the randomization follow-up for 1 year. Masking of study ing. Enrollment of 2064 patients oc- list for the next eligible patient. Once drug allocation was maintained through curred from June 1999 through Febru- study drug was given, a patient’s partici- 1 year of follow-up. ary 2000. The primary inclusion pation in the study was communicated criterion was intent to treat a native coro- to the coordinating center within 30 min- Definitions nary artery with stent implantation with- utes of randomization. This strategy Outcomes at 6 months, including death, out planned use of a Gp IIb/IIIa inhibi- served to minimize the number of pa- MI, and target vessel revasculariza- tor. Exclusion criteria included MI tients randomized but not actually tion, were prospectively defined sec- within 24 hours before randomization treated in ESPRIT. ondary end points of the trial. Target and ongoing chest pain precipitating ur- Study drug kits were packaged to be vessel revascularization was defined as gent referral for PCI. Other exclusion cri- indistinguishable, regardless of con- coronary artery bypass grafting or a sec- teria included PCI within the previous tent. Any drug kit that became unus- ond PCI deemed by the site investiga- 90 days, previous stent implantation at able was replaced with another uniquely tor to involve the original target ves- the target location, staged PCI antici- numbered kit of the same treatment. sel. All other revascularizations were pated less than 30 days after random- This kit then was given to the next pa- reviewed by a clinical events commit- ization, treatment with a Gp IIb/IIIa in- tient enrolled at that site. Eptifibatide tee (CEC) to determine whether the hibitor or a thienopyridine within 30 was given as two 180-µg/kg boluses 10 procedure met the criteria for target ves- days before randomization, stroke or minutes apart and as a continuous in- sel revascularization. transient ischemic attack within 30 days fusion of 2.0 µg/kg per minute (1.0 An end-point MI could occur by 1 of before randomization, history of hem- µg/kg per minute in patients with a se- 2 criteria. The first criterion was eleva- orrhagic stroke, history of bleeding rum creatinine level Ͼ2.0 mg/dL [177 tion of the creatine kinase–MB isoen- diathesis or evidence of abnormal bleed- µmol/L]) started with the first bolus and zyme (CK-MB), assessed by the enzy- ing within 30 days before randomiza- continued for 18 to 24 hours. All pa- matic core laboratory in at least 2 tion, major surgery within the previ- tients received concomitant aspirin, and samples, to at least 3 times the upper ous 6 weeks, uncontrolled hypertension a weight-adjusted heparin regimen was limit of normal within 24 hours after (Ͼ200/100 mm Hg), documented recommended (initial bolus of 60 U/kg), PCI. The other criterion included MIs thrombocytopenia (platelet count with a target activated clotting time of reported by an investigator and adjudi-

Downloaded From: https://jamanetwork.com/ on 10/02/2021 EPTIFIBATIDE IN CORONARY STENTING

ized patients, as randomized). Survival Figure 1. Flow of Patients Through the ESPRIT Trial analysis methods were used for the 6-month analyses. Pairwise compari- 2064 Patients Randomized sons between the 2 treatment groups were made using the log-rank test, with event rates calculated by the Kaplan- 1040 Assigned to Receive Eptifibatide 1024 Assigned to Receive Placebo 1040 Received Eptifibatide as Assigned 1024 Received Placebo as Assigned Meier method. Two-sided P values are reported. Time to first occurrence of any component of the composite end point 1025 Received a Percutaneous Coronary Intervention 1015 Received a Percutaneous Coronary Intervention 986 Received Stent 997 Received Stent is shown by Kaplan-Meier survival 39 Did Not Receive Stent 18 Did Not Receive Stent 15 Did Not Receive a Percutaneous Coronary 9 Did Not Receive a Percutaneous Coronary curves. Treatment effects by sub- Intervention Intervention groups are reported as hazard ratios (HRs) with 95% confidence intervals

1 Lost to Follow-up 1 Lost to Follow-up (CIs), calculated using a Cox propor- tional hazards model.

1039 With 30-d Follow-up 1023 With 30-d Follow-up RESULTS The 6-month follow-up database was 16 Lost to Follow-up 13 Lost to Follow-up locked on January 5, 2001. Follow-up of all end-point clinical events, obtained by telephone contact or clinic 1023 With 6-mo Follow-up for Mortality 1010 With 6-mo Follow-up for Mortality 988 With Follow-up of All End-Point Clinical Events 977 With Follow-up of All End-Point Clinical Events visit at 6 months after randomization, was available for 988 of 1040 patients ESPRIT indicates Enhanced Suppression of the Platelet IIb/IIIa Receptor with Integrilin Therapy. assigned to receive eptifibatide (95.0%) and 977 of 1024 patients assigned to re- cated as end-point MIs by the CEC. This and the reduction in the rate of the pri- ceive placebo (95.4%); mortality sta- required corroboration in the form of a mary composite end point (death, MI, tus was available for 1023 (98.4%) and clinical syndrome consistent with MI urgent target vessel revascularization, 1010 (98.6%), respectively (FIGURE 1). (symptoms and signs of cardiac chest or bailout Gp IIb/IIIa inhibitor use at Among the 2064 patients enrolled in pain, other anginal equivalent, or acute 48 hours). The secondary end point was the study, baseline demographic and an- shortness of breath with features of new- used for power calculations because of giographic characteristics were bal- onset pulmonary edema or hypoten- its application in other trials of Gp IIb/ anced and did not differ significantly be- sion) and supportive electrocardio- IIIa inhibitors in this setting. We ex- tween treatment groups.15 The cohort graphic (ECG) or cardiac marker data. pected a rate of the secondary end point had a median age of 62 years, about 73% Supportive ECG findings included de- of 11% in the placebo group, reduced were male, 20% had diabetes, 23% were velopment of Q waves (Ն0.04 sec- by 33% with treatment. With 86% current smokers, about one third had onds) in at least 2 contiguous leads or power and a 2-tailed ␣ level of .05, the had a prior MI, and slightly more than new left bundle-branch block. Support- projected sample size for the study was half of the patients in both treatment ive cardiac marker findings included (in 2400 patients. groups had hypertension and hyperlip- order of precedence and to the exclu- An independent board monitored the idemia. Almost 20% of patients had an sion of the next value) an elevated level safety of the trial. There was no plan for acute coronary syndrome within 48 of CK-MB, troponin I, or troponin T; or interim analyses of efficacy because re- hours or acute ST-segment elevation MI a total CK level elevated to at least twice cruitment was anticipated to be com- within 7 days before intervention. More the upper limit of normal. In cases of re- plete within 6 months. When enroll- than 98% of patients randomized into peat PCI or bypass surgery, cardiac ment continued beyond 8 months, the trial underwent PCI, at least 1 stent marker elevations to at least 3 times or however, the board independently was placed in 97.2% of patients who un- 5 times the upper limit of normal were elected to evaluate both safety and ef- derwent PCI, and more than 97% of pa- required, respectively. ficacy, using a prespecified criterion of tients received a thienopyridine, pre- PϽ.005 for the reduction in death or dominantly clopidogrel.15 Statistical Analyses MI at 48 hours for early termination of The 6-month composite end point of The sample size for ESPRIT was calcu- the trial. death or MI occurred in 11.5% of pla- lated based on both the projected re- Study coordinators collected data on cebo-treated patients and in 7.5% of ep- duction in the rate of a key secondary case report forms at the sites. All analy- tifibatide-treated patients (HR, 0.63; 95% composite end point (death, MI, or tar- ses were performed according to the in- CI, 0.47-0.84; P=.002; FIGURE 2A; get vessel revascularization at 30 days) tention-to-treat principle (all random- TABLE 1). There was consistency in the

Downloaded From: https://jamanetwork.com/ on 10/02/2021 EPTIFIBATIDE IN CORONARY STENTING

Figure 2. Cumulative Incidence of Study End Points Among Patients Treated With Eptifibatide or Placebo

A Death or MI B Death, MI, or Target Vessel Revascularization C Death

20 Group 20 10 18.3 18 Placebo 18 9 Eptifibatide 16 16 8 14.2 14 14 7 11.5 12 12 6 10 10 5 8 7.5 8 4

Incidence, % 6 6 3 4 4 2 1.4 2 2 1 0.8 03060 90 120 150 180 03060 90 120 150 180 03060 90 120 150 180 Time Since Randomization, d Time Since Randomization, d Time Since Randomization, d No. at Risk Placebo 919 878 872 869 867 865 910 860 841 826 809 799 1017 1006 1001 999 998 996 Eptifibatide 973 924 921 918 918 915 961 906 891 874 858 849 1035 1019 1018 1017 1016 1014

MI indicates myocardial infarction. For the composite end point of death or MI, hazard ratio [HR], 0.63; 95% confidence interval [CI], 0.47-0.84; P=.002. For the composite end point of death, MI, or target vessel revascularization, HR, 0.75; 95% CI, 0.60-0.93; P=.008. For the end point of death, HR, 0.56; 95% CI, 0.24-1.34; P=.19.

direction of treatment effect of eptifi- Table 1. Kaplan-Meier Event Rates for Study End Points at 6 Months, by Treatment batide with regard to this end point Event Rates, % across patient subgroups defined by age, Hazard Ratio weight, sex, presence or absence of dia- Eptifibatide Placebo Log-Rank (95% Confidence End Points (n = 1040) (n = 1024) P Value Interval) betes, and clinical condition (FIGURE 3). Composite end points However, as can be seen from the wide Death or myocardial infarction 7.5 11.5 .002 0.63 (0.47-0.84) CIs in some of the analyses in Figure 3, Death, myocardial infarction, or 14.2 18.3 .008 0.75 (0.60-0.93) the study did not have power to test the target vessel revascularization treatment effect in these predefined sub- Death 0.8 1.4 .19 0.56 (0.24-1.34) groups. In additional analyses that tested Myocardial infarction 7.0 10.4 .005 0.65 (0.48-0.88) for treatment-by-subgroup interac- Target vessel revascularization tions, an increased benefit of eptifi- All 8.6 9.4 .51 0.91 (0.68-1.22) batide was seen with increasing age Coronary artery bypass grafting 2.9 2.7 .82 1.06 (0.63-1.79) Percutaneous transluminal 6.1 7.1 .36 0.85 (0.61-1.20) (P=.03). The composite end point of coronary angioplasty death, MI, or target vessel revascularization at 6 months also differed significantly between treatment groups (Fig- ence between groups at 6 months was solute treatment benefit for the com- ure 2B). The incidence of the individual manifested after the first 48 hours. posite end point of MI and death also components varied. The only signifi- For mortality, the rate of death be- showed that most of the benefit ac- cant effect was observed in reduction of tween 30 days and 6 months with pla- crued early and was at least main- MI. Mortality was reduced, but not to a cebo was nearly twice that with eptifi- tained over the ensuing 6 months. significant extent. Finally, target vessel batide treatment (1.4% vs 0.8%; P=.19). revascularization was not substantially Furthermore, the Kaplan-Meier curves COMMENT affected (Table 1). continued to separate over time; about The acute benefits of eptifibatide have Event rates for the composite end two thirds of the absolute difference ac- been established in several settings, in- point of death, MI, or target vessel re- crued between 30 days and 6 months cluding treatment of patients with acute vascularization and selected indi- (Figure 2C). However, only 14 and 8 coronary syndromes and as adjunctive vidual end points at 48 hours, 30 days, deaths had occurred by 6 months in the therapy for patients undergoing PCI. The and 6 months are shown in TABLE 2. placebo and eptifibatide groups, re- 6-month follow-up of patients in the About 83% of the 4.1% benefit of ep- spectively, and this difference was not ESPRIT trial shows the efficacy of epti- tifibatide in reducing this end point at statistically significant. The rate of tar- fibatide in coronary stenting to be clini- 6 months was achieved in the first 48 get vessel revascularization also was cally relevant, statistically significant, and hours, during which an absolute re- lower in the eptifibatide group, but the durable over time. Adjunctive treat- duction of 3.4% occurred. However, an difference was small and not signifi- ment with eptifibatide had the greatest additional 17% of the absolute differ- cant at any of the time points. The ab- effect on MI and reduced the compos-

Downloaded From: https://jamanetwork.com/ on 10/02/2021 EPTIFIBATIDE IN CORONARY STENTING

Figure 3. Hazard Ratios and 95% CIs for Risk of Death or MI at 6 Months With Eptifibatide Table 2. Kaplan-Meier Event Rates at vs Placebo Treatment, by Subgroup 48 Hours, 30 Days, and 6 Months, by Treatment Rate of Death or MI (%) Event Rates, % Subgroup No. of Patients (%) Eptifibatide Placebo Favors Eptifibatide Favors Placebo Age, y Eptifibatide Placebo <65 1172 (57) 7.5 8.7 End Points (n = 1040) (n = 1024) ≥65 892 (43) 7.5 15.2 Death, Myocardial Infarction, or Target Vessel Revascularization Weight, kg <76 571 (28) 8.1 15.1 48 h 6.0 9.4 76-89 723(35) 7.7 12.1 30 d 7.5 11.0 ≥ 6 mo 14.2 18.3 90 770(37) 6.7 8.4 Absolute increase Diabetes 48 h-30 d 1.5 1.7 Present 419(20) 6.3 10.2 30 d-6 mo 6.7 7.2 Absent 645(80) 7.7 11.8 48 h-6 mo 8.2 8.9

Sex Death Male 1502(73) 7.4 10.3 48 h 0.1 0.2 Female 562(27) 7.5 14.6 30 d 0.4 0.6 6 mo 0.8 1.4 Cardiac Events Absolute increase Stable Angina 794(38) 6.9 7.3 48 h-30 d 0.3 0.4 ACS ≥2 d 664(32) 6.1 11.7 30 d-6 mo 0.4 0.8 ACS <2 d 279(14) 9.5 18.6 48 h-6 mo 0.7 1.2 ST-elevation MI <7 d 93(5) 11.5 20.5 Myocardial Infarction 0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 48 h 5.4 9.0 Hazard Ratio (95% CI) 30 d 6.2 9.7 6 mo 7.0 10.4 CI indicates confidence interval; MI, myocardial infarction; and ACS, acute coronary syndrome. Absolute increase 48 h-30 d 0.8 0.7 30 d-6 mo 0.8 0.7 ite incidence of death or MI, as well as Mac-1, were forwarded as hypotheses to 48 h-6 mo 1.6 1.4 its components, and that of death, MI, explain the differences observed in clini- Target Vessel Revascularization and target vessel revascularization. Of in- cal trials.17-20 The ESPRIT results sug- 48 h 0.6 1.1 30 d 1.9 2.3 terest, benefit continued to accrue be- gest that long-term benefits can be ex- 6 mo 8.6 9.4 tween 48 hours and 6 months, well be- plained more simply by appropriate use Absolute increase 48 h-30 d 1.3 1.3 yond the 18 to 24 hours of initial of high-level inhibitors of platelet Gp IIb/ 30 d-6 mo 6.7 7.1 treatment. The suggestion that treat- IIIa integrin during PCI. In the Integri- 48 h-6 mo 8.0 8.4 ment with eptifibatide may be associ- lin to Minimize Platelet Aggregation and ated with lower mortality at 6 months Coronary Thrombosis (IMPACT-II) get vessel revascularization within the (a nonsignificant 44% reduction in this study, a 135-µg/kg bolus and either a 0.5- first 48 hours, this phenomenon ac- study) supports the concept of the or 0.75-µg/kg-per-minute infusion were counted for only a small proportion of complementary effects of stenting and studied in coronary intervention. There the long-term benefit. In particular, the Gp IIb/IIIa inhibition in improving this was only a borderline significant 16% to treatment effect on mortality appears to fundamental adverse outcome in pa- 22% reduction in the composite end be a delayed phenomenon—the differ- tients with ischemic heart disease.17 point of death, MI, or urgent target ves- ence in mortality between the placebo Until this report, abciximab had been sel revascularization at 30 days with ep- and eptifibatide groups tripled, from the only Gp IIb/IIIa inhibitor shown to tifibatide treatment, with maintenance 0.2% to 0.6%, between 30 days and 6 reduce ischemic complications of PCI (but not augmentation) of the results at months. Clearly, these numbers are small over the long term.17 Accordingly, much 6 months.12 The results of the ESPRIT (and the differences are not statistically of the discussion of putative mecha- study, which used a dose of eptifibatide significant); there were only 22 deaths nisms by which abciximab improves some 3 to 4 times higher than that stud- across both treatment arms at 6 months, long-term outcomes compared with ied in IMPACT-II, lend credence to the and the outcomes of patients lost to fol- smaller, competitive Gp IIb/IIIa inhibi- need for greater inhibition of platelet ag- low-up in the 2 treatment arms are un- tors (such as eptifibatide) had focused on gregation during PCI to maximize both known. Nevertheless, these data are con- the unique pharmacodynamic and phar- short- and long-term outcomes. sistent with the observations for mortality macokinetic properties of abciximab. The That benefits continued to accrue over at the same follow-up point in the Evalu- avid binding of abciximab to platelets for time is a provocative and unexplained ation of c7E3 for Prevention of Ische- up to 2 weeks after administration and observation. Although eptifibatide treat- mic Complications (EPIC),18 Evalua- its cross-reactivity with other recep- ment reduced epicardial thrombosis re- tion in PTCA to Improve Long-term tors, including ␣␯␤3 (vitronectin) and sulting in abrupt closure and urgent tar- Outcomes with abciximab Gp IIb/IIIa

Downloaded From: https://jamanetwork.com/ on 10/02/2021 EPTIFIBATIDE IN CORONARY STENTING

blockade (EPILOG),19 and the Evalua- as platelet-derived growth factor and dence that outcomes of PCI can be im- tion of Platelet IIb/IIIa Inhibitor for Stent- transforming growth factor ␤,byGp proved through inhibition of the plate- ing (EPISTENT) trials of abciximab.20 IIb/IIIa inhibition is another possible let Gp IIb/IIIa integrin.

Recent studies have suggested addi- mechanism that might improve long- A complete list of the ESPRIT Investigators was pub- tional mechanisms by which platelet Gp term vascular responses. lished previously (Lancet. 2000;356:2037-2044). IIb/IIIa blockade might indirectly im- In summary, these data, particu- Author Contributions: Study concept and design: O’Shea, Hafley, Greenberg, Lorenz, Kitt, Strony, prove long-term outcomes. Periproce- larly when coupled with the anteced- Tcheng. dural MI, as evidenced by increases in ent clinical trials of Gp IIb/IIIa inhibi- Acquisition of data: O’Shea, Hafley, Lorenz, Kitt, Strony, Tcheng. CK-MB levels, long has been associ- tion in PCI, argue strongly that all Analysis and interpretation of data: O’Shea, Hafley, ated with increased mortality and other patients undergoing PCI should be con- Greenberg, Hasselblad, Lorenz, Kitt, Strony, Tcheng. 21,22 Drafting of the manuscript: O’Shea, Hafley, Lorenz, adverse cardiac events. In an angio- sidered for treatment with an inhibi- Kitt, Strony, Tcheng. graphic substudy of the ESPRIT trial, tor of the platelet Gp IIb/IIIa integrin. Critical revision of the manuscript for important in- 23 tellectual content: O’Shea, Hafley, Greenberg, Has- Gibson et al found that eptifibatide ad- Across a series of randomized trials in- selblad, Lorenz, Kitt, Strony, Tcheng. ministration was associated with sig- volving different intravenous Gp IIb/ Statistical expertise: O’Shea, Hafley, Greenberg, Has- selblad, Tcheng. nificantly increased coronary flow re- IIIa antagonists, benefits have been ob- Obtained funding: Lorenz, Kitt, Tcheng. serve and a trend toward increased served among patients of all risk Administrative, technical, or material support: Lorenz, Strony, Tcheng. microvascular perfusion after stent categories and with all approved inter- Study supervision: O’Shea, Lorenz, Kitt, Strony, placement. In that study, poor myo- ventional devices. The strategy of maxi- Tcheng. Funding/Support: The ESPRIT trial was funded by cardial perfusion was the strongest pre- mal inhibition beginning just before the grants from COR Therapeutics, South San Francisco, dictor of release of CK-MB. Reduc- procedure and maintained through- Calif, and Schering-Plough Research Institute, Kenil- worth, NJ. Both companies provided study drug. tions in release of inflammatory factors, out the infusion, especially in the early Previous Presentation: This work was presented in part such as CD-40 ligand (David Phillips, hours immediately after PCI, appears at the American Heart Association Scientific Sessions 2000, New Orleans, La, November 14, 2000. PhD, written communication, March to be critical to imparting long-term Acknowledgment: We thank Pat French for edito- 2001), and of vascular mediators, such benefits. This study adds to the evi- rial assistance.

REFERENCES 1. Fuster V, Adams PC, Badimon JJ, Chesebro JH. The tice. Philadelphia, Pa: JB Lippincott; 1994:489-507. EPISTENT Investigators. Outcomes at 1 year and eco- pathogenesis of coronary artery disease and the acute 11. Lefkovits J, Plow EF, Topol EJ. Platelet glycopro- nomic implications of platelet glycoprotein IIb/IIIa coronary syndromes. N Engl J Med. 1992;326:242-250. tein IIb/IIIa receptors in cardiovascular medicine. N Engl blockade in patients undergoing coronary stenting: re- 2. Roe MT, Ohman EM, Maas AC, et al. Shifting the J Med. 1995;332:1553-1559. sults from a multicentre randomised trial. Lancet. 1999; open-artery hypothesis downstream: the quest for op- 12. The IMPACT-II Investigators. Randomised pla- 354:2019-2024. timal reperfusion. J Am Coll Cardiol. 2001;37:9-18. cebo-controlled trial of effect of eptifibatide on com- 18. The EPIC Investigators. Use of a monoclonal an- 3. Madan M, Berkowitz SD, Tcheng JE. Glycopro- plications of percutaneous coronary intervention: tibody directed against the platelet glycoprotein IIb/ tein IIb/IIIa integrin blockade. Circulation. 1998;98: IMPACT-II. Lancet. 1997;349:1422-1428. IIIa receptor in high-risk coronary angioplasty. N Engl 2629-2635. 13. Ohman EM, Kleiman NS, Gacioch G, et al, for the J Med. 1994;330:956-961. 4. Kong DF, Califf RM, Miller DP, et al. Clinical out- IMPACT-AMI Investigators. Combined accelerated tis- 19. The EPILOG Investigators. Platelet glycoprotein comes of therapeutic agents that block the platelet gly- sue-plasminogen activator and glycoprotein IIb/IIIa in- IIb/IIIa receptor blockade and low-dose heparin dur- coprotein IIb/IIIa integrin in ischemic heart disease. Cir- tegrin receptor blockade with Integrilin in acute myo- ing percutaneous coronary revascularization. N Engl culation. 1998;98:2829-2835. cardial infarction: results of a randomized, placebo- J Med. 1997;336:1689-1696. 5. Hynes RO. Integrins: a family of cell surface re- controlled dose ranging trial. Circulation. 1997;95: 20. The EPISTENT Investigators. Randomised placebo- ceptors. Cell. 1987;48:549-554. 846-854. controlled and balloon-angioplasty-controlled trial to as- 6. Hynes RO. Integrins: versatility, modulation, and 14. The PURSUIT Trial Investigators. Inhibition of sess safety of coronary stenting with use of platelet gly- signaling in cell adhesion. Cell. 1992;69:11-25. platelet glycoprotein IIb/IIIa with eptifibatide in pa- coprotein-IIb/IIIa blockade. Lancet. 1998;352:87-92. 7. Jang U, Lincoff AM, Plow EF, Topol EJ. Cell adhe- tients with acute coronary syndromes. N Engl J Med. 21. Abdelmeguid AE, Topol EJ. The myth of the myo- sion molecules in coronary artery disease. J Am Coll 1998;339:436-443. cardial ‘infarctlet’ during percutaneous coronary re- Cardiol. 1994;24:1591-1601. 15. The ESPRIT Investigators. Novel dosing regimen vascularization procedures. Circulation. 1996;94: 8. Phillips DR, Charo IF, Parisi LV, Fitzgerald LA. The of eptifibatide in planned coronary stent implanta- 3369-3375. platelet membrane glycoprotein IIb/IIIa complex. tion (ESPRIT): a randomised, placebo-controlled trial. 22. Califf RM, Abdelmeguid AE, Kuntz RE, et al. Myo- Blood. 1988;71:831-843. Lancet. 2000;356:2037-2044. necrosis after revascularization procedures. J Am Coll 9. Phillips DR, Charo IF, Scarborough RM. GP IIb- 16. O’Shea JC, Madan M, Cantor WJ, et al. Design Cardiol. 1998;31:241-251. IIIa: the responsive integrin. Cell. 1991;65:359-362. and methodology of the ESPRIT trial: evaluating a novel 23. Gibson C, Cohen D, Cohen E, et al. Eptifibatide 10. Charo IF, Kieffer N, Phillips DR. Platelet mem- dosing regimen of eptifibatide in percutaneous coro- improves coronary flow reserve following stent place- brane glycoproteins. In: Colman RW, ed. Hemostasis nary intervention. Am Heart J. 2000;140:834-839. ment: an ESPRIT substudy [abstract]. Circulation. 2000; and Thrombosis: Basic Principles and Clinical Prac- 17. Topol EJ, Mark DB, Lincoff AM, et al, for the 102:II-366.

Downloaded From: https://jamanetwork.com/ on 10/02/2021