Prasugrel (Page 5 of 7)

12.5 Pharmacogenomics

There is no relevant effect of genetic variation in CYP2B6, CYP2C9, CYP2C19, or CYP3A5 on the pharmacokinetics of prasugrel’s active metabolite or its inhibition of platelet aggregation.

13 NONCLINICAL TOXICOLOGY

13.1 Carcinogenesis, Mutagenesis, Impairment of Fertility

Carcinogenesis -No compound-related tumors were observed in a 2-year rat study with prasugrel at oral doses up to 100 mg/kg/day (>100 times the recommended therapeutic exposures in humans (based on plasma exposures to the major circulating human metabolite). There was an increased incidence of tumors (hepatocellular adenomas) in mice exposed for 2 years to high doses (>250 times the human metabolite exposure).

Mutagenesis — Prasugrel was not genotoxic in two in vitro tests (Ames bacterial gene mutation test, clastogenicity assay in Chinese hamster fibroblasts) and in one in vivo test (micronucleus test by intraperitoneal route in mice).

Impairment of Fertility — Prasugrel had no effect on fertility of male and female rats at oral doses up to 300 mg/kg/day (80 times the human major metabolite exposure at daily dose of 10-mg prasugrel).

14 CLINICAL STUDIES

The clinical evidence for the effectiveness of Prasugrel tablet is derived from the TRITON-TIMI 38 (TRial to Assess Improvement in Therapeutic Outcomes by Optimizing Platelet Inhibitio N with Prasugrel) study, a 13,608-patient, multicenter, international, randomized, double-blind, parallel-group study comparing Prasugrel tablets to a regimen of clopidogrel, each added to aspirin and other standard therapy, in patients with ACS (UA, NSTEMI, or STEMI) who were to be managed with PCI. Randomization was stratified for UA/NSTEMI and STEMI.

Patients with UA/NSTEMI presenting within 72 hours of symptom onset were to be randomized after undergoing coronary angiography. Patients with STEMI presenting within 12 hours of symptom onset could be randomized prior to coronary angiography. Patients with STEMI presenting between 12 hours and 14 days of symptom onset were to be randomized after undergoing coronary angiography. Patients underwent PCI, and for both UA/NSTEMI and STEMI patients, the loading dose was to be administered anytime between randomization and 1 hour after the patient left the catheterization lab. If patients with STEMI were treated with thrombolytic therapy, randomization could not occur until at least 24 hours (for tenecteplase, reteplase, or alteplase) or 48 hours (for streptokinase) after the thrombolytic was given.

Patients were randomized to receive Prasugrel tablets (60-mg loading dose followed by 10-mg once daily) or clopidogrel (300-mg loading dose followed by 75-mg once daily), with administration and follow-up for a minimum of 6 months (actual median 14.5 months). Patients also received aspirin (75-mg to 325-mg once daily). Other therapies, such as heparin and intravenous glycoprotein IIb/IIIa (GPIIb/IIIa) inhibitors, were administered at the discretion of the treating physician. Oral anticoagulants, other platelet inhibitors, and chronic NSAIDs were not allowed.

The primary outcome measure was the composite of cardiovascular death, nonfatal MI, or nonfatal stroke in the UA/NSTEMI population. Success in this group allowed analysis of the same endpoint in the overall ACS and STEMI populations. Nonfatal MIs included both MIs detected solely through analysis of creatine kinase muscle-brain (CK-MB) changes and clinically apparent (investigator-reported) MIs.

The patient population was 92% Caucasian, 26% female, and 39% ≥65 years of age. The median time from symptom onset to study drug administration was 7 hours for patients with STEMI and 30 hours for patients with UA/NSTEMI. Approximately 99% of patients underwent PCI. The study drug was administered after the first coronary guidewire was placed in approximately 75% of patients.

Prasugrel tablets significantly reduced total endpoint events compared to clopidogrel (see Table 5 and Figure 3). The reduction of total endpoint events was driven primarily by a decrease in nonfatal MIs, both those occurring early (through 3 days) and later (after 3 days). Approximately 40% of MIs occurred peri-procedurally and were detected solely by changes in CK-MB. Administration of the clopidogrel loading dose in TRITON-TIMI 38 was delayed relative to the placebo-controlled trials that supported its approval for ACS. Prasugrel tablets produced higher rates of clinically significant bleeding than clopidogrel in TRITON-TIMI 38 [see Adverse Reactions (6.1)]. Choice of therapy requires balancing these differences in outcome.

The treatment effect of Prasugrel tablet was apparent within the first few days, and persisted to the end of the study (see Figure 3). The inset shows results over the first 7 days.

Figure 3: Time to first event of CV death, MI, or stroke (TRITON-TIMI 38).

The Kaplan-Meier curves (see Figure 3) show the primary composite endpoint of CV death, nonfatal MI, or nonfatal stroke over time in the UA/NSTEMI and STEMI populations. In both populations, the curves separate within the first few hours. In the UA/NSTEMI population, the curves continue to diverge throughout the 15 month follow-up period. In the STEMI population, the early separation was maintained throughout the 15 month follow-up period, but there was no progressive divergence after the first few weeks.

Prasugrel tablets reduced the occurrence of the primary composite endpoint compared to clopidogrel in both the UA/NSTEMI and STEMI populations (see Table 5). In patients who survived an on-study myocardial infarction, the incidence of subsequent events was also lower in the Prasugrel tablets group.

Table 5: Patients with Outcome Events (CV Death, MI, Stroke) in TRITON-TIMI 38

	Patients with events		From Kaplan-Meier analysis
	Prasugrel tablets (%)	Clopidogrel (%)	Relative Risk Reduction (%) a (95% CI)	p-value
UA/NSTEMI	N=5044	N=5030
CV death, nonfatal MI, or nonfatal stroke	9.3	11.2	18.0 (7.3, 27.4)	0.002
CV death	1.8	1.8	2.1 (-30.9, 26.8)	0.885
Nonfatal MI	7.1	9.2	23.9 (12.7, 33.7)	<0.001
Nonfatal Stroke	0.8	0.8	2.1 (-51.3, 36.7)	0.922
STEMI	N=1769	N=1765
CV death, nonfatal MI, or nonfatal stroke	9.8	12.2	20.7 (3.2, 35.1)	0.019
CV death	2.4	3.3	26.2 (-9.4, 50.3)	0.129
Nonfatal MI	6.7	8.8	25.4 (5.2, 41.2)	0.016
Nonfatal Stroke	1.2	1.1	-9.7 (-104.0, 41.0)	0.77

^a RRR = (1-Hazard Ratio) x 100%. Values with a negative relative risk reduction indicate a relative risk increase.

The effect of Prasugrel tablets in various subgroups is shown in Figures 4 and 5. Results are generally consistent across pre-specified subgroups, with the exception of patients with a history of TIA or stroke [see Contraindications (4.2)]. The treatment effect was driven primarily by a reduction in nonfatal MI. The effect in patients ≥ 75 years of age was also somewhat smaller, and bleeding risk is higher in these individuals [see Adverse Reactions (6.1)]. See below for analyses of patients ≥75 years of age with risk factors.

Figure 4: Subgroup analyses for time to first event of CV death, MI, or stroke (HR and 95% CI; TRITON-TIMI 38) – UA/NSTEMI Patients.

Figure 5: Subgroup analyses for time to first event of CV death, MI, or stroke (HR and 95% CI; TRITON-TIMI 38) – STEMI Patients.

Prasugrel tablet is generally not recommended in patients ≥75 years of age, except in high-risk situations (diabetes mellitus or prior MI) where its effect appears to be greater and its use may be considered. These recommendations are based on subgroup analyses (see Table 6) and must be interpreted with caution, but the data suggest that Prasugrel tablets reduces ischemic events in such patients.

Table 6: Subgroup Analyses for Time to First Event of CV Death, MI, or Stroke: Patients < or ≥75 Years of Age, ± Diabetes, ± Prior History of MI, All ACS Patient Population

	Prasugrel tablets		Clopidogrel		Hazard Ratio (95% CI)
	N	% with events	N	% with events
Age ≥75
Diabetes — yes	249	14.9	234	21.8	0.64 (0.42, 0.97)
Diabetes — no	652	16.4	674	15.3	1.1 (0.83, 1.43)
Age <75
Diabetes — yes	1327	10.8	1336	14.8	0.72 (0.58, 0.89)
Diabetes — no	4585	7.8	4551	9.5	0.82 (0.71, 0.94)
Age ≥75
Prior MI — yes	220	17.3	212	22.6	0.72 (0.47, 1.09)
Prior MI — no	681	15.6	696	15.2	1.05 (0.80, 1.37)
Age <75
Prior MI — yes	1006	12.2	996	15.4	0.78 (0.62, 0.99)
Prior MI — no	4906	7.7	4891	9.7	0.78 (0.68, 0.90)

There were 50% fewer stent thromboses (95% C.I. 32% — 64%; p<0.001) reported among patients randomized to Prasugrel tablets (0.9%) than among patients randomized to clopidogrel (1.8%). The difference manifested early and was maintained through one year of follow-up. Findings were similar with bare metal and drug-eluting stents.

In TRITON-TIMI 38, prasugrel reduced ischemic events (mainly nonfatal MIs) and increased bleeding events [see Adverse Reactions (6.1)] relative to clopidogrel. The findings are consistent with the intended greater inhibition of platelet aggregation by prasugrel at the doses used in the study [see Clinical Pharmacology (12.2)]. There is, however, an alternative explanation: both prasugrel and clopidogrel are pro-drugs that must be metabolized to their active moieties. Whereas the pharmacokinetics of prasugrel’s active metabolite are not known to be affected by genetic variations in CYP2B6, CYP2C9, CYP2C19, or CYP3A5, the pharmacokinetics of clopidogrel’s active metabolite are affected by CYP2C19 genotype, and approximately 30% of Caucasians are reduced-metabolizers. Moreover, certain proton pump inhibitors, widely used in the ACS patient population and used in TRITON-TIMI 38, inhibit CYP2C19, thereby decreasing formation of clopidogrel’s active metabolite. Thus, reduced-metabolizer status and use of proton pump inhibitors may diminish clopidogrel’s activity in a fraction of the population, and may have contributed to prasugrel’s greater treatment effect and greater bleeding rate in TRITON-TIMI 38. The extent to which these factors were operational, however, is unknown.