Differential Dose-Response to Oral Xemilofiban After Antecedent Intravenous Abciximab
Administration for Complex Coronary Intervention
Background Placebo-controlled randomized trials of parenteral platelet glycoprotein (GP) IIb/IIIa receptor antagonists have demonstrated reduced ischemic complications of coronary angioplasty. Orally active GP IIb/IIIa blockers are being developed to allow more sustained receptor antagonism with potential for long-term secondary prevention. Sequential therapy with abciximab followed by an oral IIb/IIIa antagonist has not previously been reported. The clinical safety and pharmacodynamics of a sequential therapeutic strategy are unknown.
Methods and Results Of 74 consecutive patients enrolled in a placebo-controlled, dose-ranging pharmacokinetic/pharmacodynamic study of xemilofiban, a new oral nonpeptide GP IIb/IIIa antagonist, after elective intracoronary stent placement, 17 patients received abciximab during stent deployment as a weight-adjusted intravenous bolus and 12-hour infusion at the discretion of the investigator. Ex vivo platelet aggregation in response to 20 μmol/L ADP and 4 μg/mL collagen was measured over time after the first dose of either xemilofiban (5, 10, 15, or 20 mg) or placebo (ticlopidine) administered 8 to 18 hours after termination of abciximab and again after 1 week of twice-daily oral administration of study drug. At baseline, patients who had received abciximab had lower platelet aggregation in response to both agonists (P<.001). A significant dose-response relationship to xemilofiban was observed. Patients who had received abciximab had lower ADP-induced (P≤.010) and collagen-induced (P≤.029) platelet aggregation after xemilofiban. This pharmacodynamic interaction was no longer evident at 1 week. No significant clinical bleeding events or blood product transfusions were observed in this trial.
Conclusions Both the magnitude and the duration of pharmacodynamic response to xemilofiban were enhanced by prior abciximab treatment. The potentiated pharmacodynamic response was not evident after 1 week. This observation has implications for the safety and efficacy of sequential parenteral-oral GP IIb/IIIa blockade therapy and may be useful in deriving dose regimens for orally administered compounds.
Large-scale, placebo-controlled randomized trials involving parenteral administration of platelet glycoprotein IIb/IIIa (GP IIb/IIIa) receptor antagonists have demonstrated efficacy of these agents in reducing early (≤30 days) ischemic complications of percutaneous coronary intervention.1 2 Patients with unstable coronary ischemic syndromes and those undergoing procedures with new technologies such as rotational/directional atherectomy or bail-out stenting may derive particular benefit from adjunctive parenteral GP IIb/IIIa blockade.3 4 5 6 The currently available parenteral GP IIb/IIIa receptor antagonist is a monoclonal IgG Fab antibody fragment (c7E3, abciximab [ReoPro] Centocor, Inc) that binds the GP IIb/IIIa receptor with high affinity, resulting in a long biological half-life.7 Prior pharmacodynamic studies of abciximab have demonstrated that ex vivo platelet aggregation in response to 20 μmol/L ADP recovers from nearly complete inhibition during drug infusion to >50% of baseline values within 12 hours of drug discontinuation. Similarly, bleeding time normalizes within the same time frame.7 Orally active GP IIb/IIIa receptor blockers are currently under development to allow more sustained receptor antagonism,8 thus offering the potential for greater long-term benefit and secondary prevention of recurrent ischemic events. Sequential therapy with abciximab followed by xemilofiban, an oral GP IIb/IIIa antagonist, might be desirable. However, pharmacological differences beyond the route of administration exist between these agents. In particular, abciximab has a short plasma half-life but produces slowly reversible blockade of the GP IIb/IIIa receptor, whereas xemilofiban has a relatively longer plasma half-life and a shorter duration of action at the target receptor. The pharmacodynamics and clinical safety of a sequential treatment strategy have not been described.
This report describes our preliminary experience in 17 patients who received parenteral abciximab during a complex coronary intervention and were subsequently treated with xemilofiban. An important pharmacodynamic interaction between these agents is defined.
From October 1, 1995, through February 1, 1996, 74 patients were enrolled into a placebo-controlled, dose-ranging pharmacokinetic-pharmacodynamic study of xemilofiban, a new orally active, nonpeptide antagonist of the RGD binding site on the platelet GP IIb/IIIa receptor after elective intracoronary stent placement. This study was approved by the institutional review board of each institution, and informed consent was obtained from each patient. Of this larger group, 17 patients received intravenous abciximab during the stent procedure at the discretion of the investigator based on clinical indications as a weight-adjusted bolus (0.25 mg/kg) and 12-hour (0.6 mg/h) intravenous infusion (Table⇓). All patients received aspirin 325 mg PO daily beginning at least 1 day before the procedure and throughout the study period. Oral xemilofiban was initiated on the morning after the interventional procedure (8 to 18 hours after abciximab infusion was discontinued) in doses of 5, 10, 15, or 20 mg on a twice-daily schedule and was continued for 2 weeks. Patients who were randomized to placebo received ticlopidine 250 mg PO twice daily beginning immediately after the procedure in addition to a placebo tablet administered twice daily. The minimum dosing interval between cessation of abciximab and initiation of xemilofiban therapy was prospectively defined as ≥8 hours to provide a safety threshold.
Ex vivo platelet aggregation in response to both 20 μmol/L ADP and 4 μg/mL collagen was assessed at 2, 4, 6, 8, and 12 hours after the first dose of study drug and again at 7 days of continuous oral therapy. Blood was collected in 3.8% sodium citrate tubes, and platelet aggregation was measured in platelet-rich plasma by the turbidimetric method using single-lot 20 μmol/L ADP/L (Biodata) or single-lot 4 μg/mL horse collagen (Chronolog). Platelet aggregation was quantified as the maximal change in light transmission occurring within 5 minutes of addition of agonist. Platelet-rich plasma was not adjusted or prepared with uniform platelet counts. Thus, a potential confounding effect of widely variable platelet counts as a source of variability in the platelet aggregation studies cannot be excluded.
Between-treatment differences in platelet aggregation and differences with respect to antecedent abciximab were tested by submission of values from each scheduled observation time to ANCOVA. The ANOVA model for baseline values included randomized treatment (ticlopidine or xemilofiban 5, 10, 15, or 20 mg), antecedent abciximab use (yes or no), and the interaction of randomized treatment and abciximab. The ANCOVA model for postbaseline values included randomized treatment, antecedent abciximab (yes or no), the interaction of randomized treatment and abciximab, and baseline platelet aggregation.
In principle, platelet aggregation values are bounded by 0% and 100%. (In practice, values slightly over 100% are sometimes reported.) The bounded nature of platelet aggregation values violates an assumption of ANCOVA, so for purposes of model fitting and hypothesis testing, logits of the platelet aggregation values were used. For the logit transformation, values of ≥100% were set to 99.5%, and values of 0% were set to 0.5%.
For each xemilofiban dose group, the least mean squares response from the ANCOVA was compared with the ticlopidine least mean squares response by use of a t test based on the residual mean squares error from the ANCOVA.
Because of the exploratory nature of this analysis, no correction was made for multiple comparisons.
A significant dose-response relationship to incremental doses of xemilofiban was observed for both ADP- and collagen-induced platelet aggregation (Figs 1⇓, top, and 2, top).⇓ The duration of action of a single dose of xemilofiban (8 to 10 hours) was similar to prior observations.9
With respect to ADP-induced platelet aggregation, significant randomized treatment effects were seen at all postdose sampling times after the initial dose of study drug and at 1 week of follow-up (P≤.032). Patients who had received abciximab had significantly lower baseline platelet aggregation values (P<.001) than non–abciximab-treated patients and had lower ADP-induced platelet aggregation than non–abciximab-treated patients in the same treatment groups at 2 and 4 hours after the first dose of study drug even after control for baseline levels of aggregation (P≤.035). Randomized treatment by abciximab interaction was not statistically significant at any observation time (P≥.050), indicating that effects of xemilofiban and abciximab are approximately additive in the logit scale.
With respect to collagen-induced platelet aggregation, significant randomized treatment effects were seen after the initial dose of study drug in non–abciximab-treated patients at 4 hours (P=.007) and approached significance at 2 hours (P=.060) and 12 hours (P=.057) postdose. Patients who had received abciximab had lower collagen-induced platelet aggregation values than corresponding non–abciximab-treated patients in the same treatment groups at 2, 4, 6, and 12 hours after the initial dose of study drug (P≤.017) even after control for baseline levels of aggregation. Randomized treatment by abciximab interaction was significant at 12 hours after the initial dose of study drug (P=.024) and before the study drug at week 1 (P=.007).
There were no episodes of significant hemorrhage or transfusion in the 17 patients with sequential GP IIb/IIIa receptor blockade therapy. Patients were discharged the day after the interventional procedure and received oral xemilofiban or placebo (ticlopidine) for 2 weeks after discharge. There were no recurrent myocardial ischemic events in any of the 17 combination-treated (abciximab/xemilofiban) patients at an average follow-up of 56 days (51.0 to 84.0 days) (median and 25th and 75th percentiles) after stent deployment. In the 57 patients who did not receive abciximab at the time of stent placement, there were no significant bleeding complications in follow-up, and 2 patients experienced subacute stent thrombosis at 4 and 7 days after stent deployment. These patients were receiving xemilofiban in doses of 10 and 15 mg, respectively, twice daily.
The administration of xemilofiban, a new orally active platelet GP IIb/IIIa antagonist, to patients after complex coronary intervention was associated with a stepwise dose-response inhibition of ex vivo platelet aggregation in response to 20 μmol/L ADP and 4 μg/mL collagen. Both the magnitude and duration of pharmacodynamic response to xemilofiban were enhanced in patients who had received intravenous abciximab during the preceding interventional procedure.
One explanation for this observation is that despite the return in ex vivo platelet aggregation in response to ADP to >50% of baseline before xemilofiban administration, abciximab continues to occupy numerous GP IIb/IIIa receptors, thus making fewer receptors available for antagonism by the oral agent. The response curve to any given dose of an oral agent would thus be exaggerated for as long as this “background” receptor occupancy by abciximab exists. This observation has important implications for both efficacy and safety when a sequential treatment strategy involving intravenous abciximab followed by an orally active competitive GP IIb/IIIa inhibitor is used. The dose of an orally active agent could be reduced for a period of at least 48 to 72 hours after abciximab administration. The pharmacodynamic interaction of abciximab and xemilofiban appeared to resolve gradually over time and was not evident at 1 week in this study. As further data emerge showing clinical benefit of parenteral GP IIb/IIIa blockade with abciximab in reducing ischemic complications of coronary intervention (Epilog, unpublished data), expanded indications for use will follow.
Particularly in high-risk patient subsets, extended GP IIb/IIIa receptor blockade with an orally active compound may confer long-term clinical benefit through secondary prevention. The sequential administration of abciximab and orally active compounds such as xemilofiban would appear to be a logical therapeutic sequence. This combination may allow a smoother extension of the initial “passivation” period without reversal or interruption of inhibition of platelet aggregation for an additional period of time. Knowledge of the pharmacodynamic interaction of these agents as described in the present study may be of considerable benefit in deriving dose regimens of orally administered compounds.
Limitations of the Study
The major limitations of the present study include the facts that patients were not randomly assigned to receive intravenous abciximab and therefore differences between abciximab-treated and non–abciximab-treated patients may not be ascribed with certainty to abciximab, and that the small sample sizes per treatment arm make conclusions based on this analysis less definitive. In addition, the time interval between abciximab and study drug administration was not fixed, and platelet aggregation data before or during abciximab treatment were not obtained. Such data would have permitted determination of a true baseline measurement for each patient and enhanced the interpretation of both second baseline (before study drug) and time-related dose-response effects. Nonetheless, these limitations do not diminish the potential importance of the observed pharmacodynamic interaction between intravenous abciximab and xemilofiban, an oral competitive GP IIb/IIIa antagonist.
- Received April 1, 1996.
- Revision received June 17, 1996.
- Accepted June 20, 1996.
- Copyright © 1996 by American Heart Association
Tcheng JE, Harrington RA, Kottke-Marchant K, Kleiman NS, Ellis SG, Kereiakes DJ, Mick MJ, Navetta FI, Smith JE, Worley SJ, Miller JA, Joseph DM, Sigmon KN, Kitt MM, duMee CP, Califf RM, Topol EJ, for the IMPACT Investigators. Multicenter, randomized, double-blind, placebo-controlled trial of the platelet integrin glycoprotein IIb/IIIa blocker integrelin in elective coronary intervention. Circulation. 1995;91:2151-2157.
Lincoff AM, Califf RM, Anderson K, Weisman HF, Topol EJ, for the EPIC Investigators. Striking clinical benefit with platelet GP IIb/IIIa inhibition by c7E3 among patients with unstable angina: outcomes in the EPIC trial. Circulation. 1994;90(suppl I):I-21. Abstract.
Teirstein PS, Yakubov SJ, Thel MC, Wildermann N, Kereiakes DJ, Popma JJ, Lincoff MA, Tcheng JE, Califf RM, Topol EJ. Platelet IIb/IIIa blockade with integrelin: atherectomy patients. J Am Coll Cardiol. 1996;27:334A. Abstract.
Lefkovits J, Anderson K, Weisman H, Topol EJ, for the EPIC Investigators. Increased risk of non-Q MI following DCA: evidence for a platelet dependent mechanism from the EPIC trial. Circulation. 1994;90(suppl I):I-214. Abstract.
Zidar JP, Kruse KR, Thel MC, Kereiakes DJ, Muhlestein JB, Davidson CJ, Teirstein PS, Tenaglia A, Yakubov SJ, Popma JJ, Tanguay JF, Kitt MM, Lorenz TJ, Tcheng JE, Lincoff AM, Califf RM, Topol EJ. Integrelin for emergency coronary artery stenting. J Am Coll Cardiol. 1996;27:138A. Abstract.
Tcheng JE, Ellis SG, George BS, Kereiakes DJ, Kleiman NS, Talley JD, Wang AL, Weisman HF, Califf RM, Topol EJ. Pharmacodynamics of chimeric glycoprotein IIb/IIIa integrin antiplatelet antibody fab 7E3 in high-risk coronary angioplasty. Circulation. 1994;90:1757-1764.
Simpfendorfer C, Kottke-Marchant K, Topol EJ. First experience with chronic platelet GPIIb/IIIa receptor blockade: a pilot study of xemilofiban, an orally active antagonist in unstable angina patients eligible for PTCA. J Am Coll Cardiol. 1996;27:242A. Abstract.
Anders RJ, Alexander JC, Hantsbarger GL, Burns DM, Oliver SD, Cole G, Fitzgerald DJ. Demonstration of potent inhibition of platelet aggregation with an orally active GP IIb/IIIa receptor antagonist. J Am Coll Cardiol. February 1995;(special issue):117A. Abstract.