Nexium-occurrence Of Drug Interaction With Clopidogrel (plavix) Summary

NEXIUM-Occurrence of Drug Interaction with Clopidogrel (Plavix) Summary •

On January 26, 2009, the U.S. Food and Drug Administration (FDA) issued an early communication regarding an ongoing safety review of clopidogrel. Sanofi-Aventis and Bristol-Myers Squibb, the manufacturers of clopidogrel (Plavix®), have agreed to work with the FDA to conduct studies to obtain additional information regarding the effect of genetic factors and other drugs (including proton pump inhibitors [PPIs]) on the effectiveness of clopidogrel. Until further information is available from these studies, the FDA recommends that healthcare providers should continue to prescribe clopidogrel because it prevents blood clots that can lead to a heart attack or stroke. In addition, healthcare providers should re-evaluate the necessity for starting or continuing treatment with PPIs, including Losec, in patients taking clopidogrel (risk versus benefit). 1

•

Given the scientific limitations and contradictory nature of data currently available, AstraZeneca agrees that more research is needed before any conclusions can be drawn about a potential interaction between clopidogrel and PPIs.

•

AstraZeneca's ongoing monitoring of clinical study data and adverse event reports from marketed use, have not raised safety concerns to date on the use of clopidogrel in conjunction with Nexium® (esomeprazole) or Losec® (omeprazole).

•

It is the position of AstraZeneca that its PPI products, Nexium® and Losec® should be used according to their UK product licence, inclusive of indications and dosing recommendations, with allowances for variations in these in individual circumstances under the supervision of physicians and other health care professionals.

•

In 2008, an expert consensus document prepared by the American College of Cardiology Foundation (ACCF) Task Force, the American College of Gastroenterology (ACG), and the American Heart Association (AHA), recommended that all patients who are receiving NSAIDs, aspirin, dual antiplatelet therapy, or concomitant anticoagulant therapy and who are at risk for gastrointestinal injury should receive gastroprotective prophylactic treatment with a PPI. Concomitant therapy with a PPI and clopidogrel appears to result in less GI bleeding.

•

Esomeprazole is the S-isomer of omeprazole, which is a racemic mixture of the (R)- and (S)- isomers. The metabolism of esomeprazole is less dependent on CYP2C19 than omeprazole.9

•

Plavix is a prodrug. The active metabolite, a thiol derivative, is formed by oxidation of clopidogrel to 2-oxo-clopidogrel and subsequent hydrolysis. The oxidative step is regulated primarily by cytochrome P450 isoenzymes 2B6 and 3A4 and to a lesser extent by 1A1, 1A2 and 2C19.

•

Several studies have found that patients with the CYP2C19*2 allele have high clopidogrel platelet reactivity (variable response to treatment), which has been associated with a higher rate of subsequent cardiovascular events.12,13,14,15 AstraZeneca UK Limited Horizon Place 600 Capability Green Luton Bedfordshire LU1 3LU

Tel +44 (0) 1582 836000 Fax +44 (0) 1582 838000 www.astrazeneca.co.uk

Registered in England No. 03674842 Registered Office 15 Stanhope Gate London W1K 1LN AstraZeneca UK Limited is a subsidiary company of AstraZeneca PLC

•

Some in vitro data suggests a potential drug interaction between PPIs and clopidogrel due to metabolism by the cytochrome P450 pathway. However, there has been relatively little evidence of any clinically significant interaction between clopidogrel and PPIs from prospective studies.8

•

A summary of the data regarding esomeprazole and a potential drug interaction with clopidogrel is provided in the table below. More detailed information on these studies is provided in the clinical data section below.

TABLE: Summary of the Data Regarding Esomeprazole and a Potential Drug Interaction with Clopidogrel Study Study Design Results Proton Pump Inhibitors (PPI) Evaluated Pharmacodynamic Studies Siller-Matula Ex vivo Esomeprazole and pantoprazole were Esomeprazole Pantoprazole et al16 pharmacodynamic shown to have no effect on platelet study. inhibition by clopidogrel in patients with coronary artery disease undergoing percutaneous coronary intervention. Sibbing et al Ex vivo Platelet aggregation was significantly Omeprazole Pantoprazole pharmacodynamic higher in omeprazole-treated patients vs Esomeprazole study. patients not treated with a PPI. Pantoprazole and esomeprazole did not adversely affect platelet response to clopidogrel. Clinical Outcome Data: Post hoc Analyses Tsiaousis et Post hoc analysis There was no difference between PPIs were assessed as a al of the 3C study. treatment groups (with or without PPI) class, and no data are with regard to death and myocardial available for individual infarction during the first year following drugs. coronary stenting. Dunn et al Post hoc analysis The efficacy of clopidogrel was similar in PPIs were assessed as a of the CREDO patients taking PPIs compared to those class, and no data are trial. not taking PPIs. available for individual drugs. Post hoc analysis PPI therapy had no effect on the clinical PPIs were assessed as a Simon et al of FAST-MI. effect of clopidogrel. In addition, class, and no data are omeprazole had no effect on the clinical available for individual effect of clopidogrel. drugs other than omeprazole. O’Donoghue Post hoc analysis There was a benefit with prasugrel over PPIs were assessed as a et al of TRITON-TIMI clopidogrel in patients with acute class, and no data are 38. coronary syndrome undergoing available for individual percutaneous intervention, whether or not drugs. they were also on a PPI. Clinical Outcome Data: Retrospective Database Analyses Pezalla et al Retrospective The 1-year acute myocardial infarction PPIs were assessed as a analysis of the rates were significantly lower in the class, and no data are medical and control group compared to the high PPI available for individual pharmacy exposure group. drugs. databases at Aetna. Aubert et al A retrospective The 1-year incidence rates of major PPIs were assessed as a cohort study using cardiovascular events were significantly class, and no data are the National higher in patients treated with clopidogrel available for individual MedCo Integrated and a PPI compared to clopidogrel alone. drugs. Database. Stanak et al A retrospective All PPIs studied had a similar inhibitory Omeprazole Rabeprazole database analysis effect on clopidogrel when taken Lansoprazole using the National concomitantly. Pantoprazole MedCo Integrated AstraZeneca UK Limited Horizon Place 600 Capability Green Luton Bedfordshire LU1 3LU

Tel +44 (0) 1582 836000 Fax +44 (0) 1582 838000 www.astrazeneca.co.uk

Registered in England No. 03674842 Registered Office 15 Stanhope Gate London W1K 1LN AstraZeneca UK Limited is a subsidiary company of AstraZeneca PLC

Study

Ho et al

Juurlink et al

Ramirez et al

Study Design Database Retrospective cohort study using the VA Hospital System medical records and pharmacy database.

Population-based nested case-control study using the records of the Ontario Public Drug Program in Ontario, Canada.

Retrospective data analysis from the University of Pittsburgh NHLBI (National Heart, Lung, and Blood Institute) Dynamic Registry.

Results

Concomitant clopidogrel and PPI therapy at any time point during follow-up was associated with an increased risk of death or rehospitalization for acute coronary syndrome (ACS; primary endpoint); rehospitalization for ACS and revascularization procedure (secondary endpoints). Concomitant treatment with clopidogrel and PPI was not associated with an increased risk of all-cause mortality. There was a significant association between current use of PPIs and readmission for an acute myocardial infarction. When the investigators stratified the results by PPI, current use of omeprazole, lansoprazole and rabeprazole were associated with readmission for acute myocardial infarction, but pantoprazole was not. In patients undergoing PCI and discharged on clopidogrel therapy, the concomitant use of PPI’s did not result in adverse cardiovascular outcomes at one year.

Proton Pump Inhibitors (PPI) Evaluated Esomepazole Omeprazole Rabeprazole Lansoprazole Pantoprazole

Esomeprazole was not included in this study.

Omeprazole Rabeprazole Lansoprazole Pantoprazole

Esomeprazole was not included in this study. PPIs were assessed as a class, and no data are available for individual drugs.

Metabolism Esomeprazole is completely metabolised by the cytochrome P450 system (CYP). The major part of the metabolism of esomeprazole is dependent on the polymorphic CYP2C19, responsible for the formation of the hydroxy- and desmethyl metabolites of esomeprazole. The remaining part is dependent on another specific isoform, CYP3A4, responsible for the formation of esomeprazole sulphone, the main metabolite in plasma. 2 . In vitro and in vivo studies have shown that esomeprazole is not likely to inhibit CYPs 1A2, 2A6, 2C9, 2D6, 2E1 and 3A4. 3 Esomeprazole inhibits CYP2C19, the major esomeprazole metabolising enzyme. Thus, when esomeprazole is combined with drugs metabolised by CYP2C19, such as diazepam, citalopram, imipramine, clomipramine, phenytoin etc., the plasma concentrations of these drugs may be increased and a dose reduction could be needed. This should be considered especially when prescribing esomeprazole for on-demand therapy. Concomitant administration of 30 mg esomeprazole resulted in a 45% decrease in clearance of the CYP2C19 substrate diazepam. Concomitant administration of 40 mg esomeprazole resulted in a 13% increase in trough plasma levels of phenytoin in epileptic patients. It is recommended to monitor the plasma concentrations of phenytoin when treatment with esomeprazole is introduced or withdrawn. Omeprazole (40 mg once daily) increased voriconazole (a CYP2C19 substrate) Cmax and AUCτ by 15% and 41%, respectively. AstraZeneca UK Limited Horizon Place 600 Capability Green Luton Bedfordshire LU1 3LU

Tel +44 (0) 1582 836000 Fax +44 (0) 1582 838000 www.astrazeneca.co.uk

Registered in England No. 03674842 Registered Office 15 Stanhope Gate London W1K 1LN AstraZeneca UK Limited is a subsidiary company of AstraZeneca PLC

Concomitant administration of 40mg esomeprazole to warfarin-treated patients in a clinical trial showed that coagulation times were within the accepted range. However, post-marketing, a few isolated cases of elevated INR of clinical significance have been reported during concomitant treatment. Monitoring is recommended when initiating and ending concomitant esomeprazole treatment during treatment with warfarin or other coumarine derivatives. In healthy volunteers, concomitant administration of 40 mg esomeprazole resulted in a 32% increase in area under the plasma concentration-time curve (AUC) and a 31% prolongation of elimination half-life(t1/2) but no significant increase in peak plasma levels of cisapride. The slightly prolonged QTc interval observed after administration of cisapride alone, was not further prolonged when cisapride was given in combination with esomeprazole. Esomeprazole has been shown to have no clinically relevant effects on the pharmacokinetics of amoxicillin or quinidine. Studies evaluating concomitant administration of esomeprazole and either naproxen or rofecoxib did not identify any clinically relevant pharmacokinetic interactions during short-term studies. 2 Information regarding the routes of hepatic metabolism for esomperazole and the various PPIs is provided in the table below. TABLE: Metabolism of Proton Pump Inhibitors Proton Pump Inhibitor Nexium

Lansoprazole

Losec 5

Pantoprazole 6

Rabeprazole 7

Routes of Hepatic Metabolism

Esomeprazole is extensively metabolised in the liver by the cytochrome P450 (CYP) enzyme system. The metabolites of esomeprazole lack antisecretory activity. The major part of esomeprazole’s metabolism is dependent upon the CYP2C19 isoenzyme, which forms the hydroxy and desmethyl metabolites. The remaining amount is dependent on CYP3A4 which forms the sulphone metabolite. 4

Lansoprazole is extensively metabolised by the liver and the metabolites are excreted by both the renal and biliary route. The metabolism of lansoprazole is mainly catalysed by the enzyme CYP2C19. The enzyme CYP3A4 also contributes to the metabolism. Omeprazole is completely metabolised by the cytochrome P450 system, mainly in the liver. The major part of its metabolism is dependent on the polymorphically expressed, specific isoform CYP2C19 (S-mephenytoin hydroxylase), responsible for the formation of hydroxyomeprazole, the major metabolite in plasma. Pantoprazole is extensively metabolised in the liver through the cytochrome P450 system. The main metabolic pathway is demethylation, by CYP2C19, with subsequent sulfation; other metabolic pathways include oxidation by CYP3A4. In vitro studies with human liver microsomes indicated that rabeprazole sodium is metabolised by isoenzymes of CYP450 (CYP2C19 and CYP3A4).

Background

In 2008, an expert consensus document prepared by the American College of Cardiology Foundation (ACCF) Task Force, the American College of Gastroenterology (ACG), and the American Heart Association (AHA), recommended that all patients who are receiving NSAIDs, aspirin, dual antiplatelet therapy, or concomitant anticoagulant therapy and who are at risk for gastrointestinal injury should receive gastroprotective prophylactic treatment with a PPI. 8 In the consensus document, the group noted that concomitant therapy with a PPI and clopidogrel appears to result in less GI bleeding. Despite some in vitro data that suggests a potential drug interaction between PPIs and clopidogrel due to metabolism by AstraZeneca UK Limited Horizon Place 600 Capability Green Luton Bedfordshire LU1 3LU

Tel +44 (0) 1582 836000 Fax +44 (0) 1582 838000 www.astrazeneca.co.uk

Registered in England No. 03674842 Registered Office 15 Stanhope Gate London W1K 1LN AstraZeneca UK Limited is a subsidiary company of AstraZeneca PLC

the cytochrome P450 pathway, there has been relatively little evidence of any clinically significant interaction between clopidogrel and PPIs from prospective studies.

Esomeprazole is the S-isomer of omeprazole, which is a racemic mixture of the (R)- and (S)isomers. The metabolism of esomeprazole is less dependent on CYP2C19 than omeprazole. 9 Human liver microsome models using liver samples from three surgery donors showed that the two CYP2C19-dependent routes (5-O-desmethyl and hydroxy formation) accounted for 73% (46 and 27%, for 5-O-desmethyl and hydroxy metabolites, respectively) of the total hepatic metabolism for esomeprazole and 98% (4.2 and 94%, respectively) of hepatic metabolism for Romeprazole. 10 Metabolism to the CYP3A4-mediated omeprazole sulfone accounted for 27 and 2.0% of the sum of metabolism for the S- and R- isomers, respectively. Other cytochromes, CYP2C9, 2A6, and 2D6, accounted for insignificant amounts of metabolism. Clopidogrel is a prodrug. The active metabolite, a thiol derivative, is formed by oxidation of clopidogrel to 2-oxo-clopidogrel and subsequent hydrolysis. The oxidative step is regulated primarily by cytochrome P450 isoenzymes 2B6 and 3A4 and to a lesser extent by 1A1, 1A2 and 2C19. 11 Several studies have evaluated the effect of cytochrome P450 polymorphisms on platelet reactivity in patients treated with clopidogrel and patient outcome. The studies have found that patients with the CYP2C19*2 allele have high clopidogrel platelet reactivity (variable response to treatment), which has been associated with a higher rate of subsequent cardiovascular events. 12 , 13 , 14 , 15 Clinical Data

Pharmacodynamic Data Siller-Matula et al 16 studied the effects of pantoprazole and esomeprazole on platelet inhibition by clopidogrel in 300 patients with coronary artery disease (CAD) undergoing percutaneous coronary intervention (PCI). All patients had been on clopidogrel (75 mg / day) and aspirin (100 mg / day) for ≥ 5 days (3 months on average) at the time of inclusion into the study and had received a 600 mg clopidogrel loading dose at the start of clopidogrel treatment. Platelet reactivity to clopidogrel was measured through vasodilator-stimulated phosphoprotein phosphorylation (VASP) assays and ADP-induced aggregometry. The results are presented in the tables below. TABLE I: Comparison of Platelet Inhibition in PCI Patients Receiving Clopidogrel with and without a PPI. Clopidogrel Clopidogrel + PPI P - value (n = 74) (n = 226) PRI assessed using VASP 49% 51% 0.724 ADP-induced aggregometry 41 U 45 U 0.619 PCI = percutaneous coronary intervention; PPI = proton pump inhibitor; PRI = platelet reactivity index; VASP = vasodilator-stimulated phosphoprotein phosphorylation

AstraZeneca UK Limited Horizon Place 600 Capability Green Luton Bedfordshire LU1 3LU

Tel +44 (0) 1582 836000 Fax +44 (0) 1582 838000 www.astrazeneca.co.uk

Registered in England No. 03674842 Registered Office 15 Stanhope Gate London W1K 1LN AstraZeneca UK Limited is a subsidiary company of AstraZeneca PLC

TABLE II: Comparison of Platelet Inhibition in PCI Patients Receiving Clopidogrel with Esomeprazole, Pantoprazole, or Without a PPI. Esomeprazole Pantoprazole Without a PPI* (n = 74) (n = 74) (n = 152) PRI assessed using VASP 54% 50% 49% ADP-induced aggregometry 42 U 47 U 41 U P – value versus esomeprazole or pantoprazole was 0.382. PCI = percutaneous coronary intervention; PPI = proton pump inhibitor; PRI = platelet reactivity index; VASP = vasodilator-stimulated phosphoprotein phosphorylation

The investigators conducted a multivariate logistic regression analysis of the results and found that minor baseline differences in demographic data between the groups (male gender, intake of statins, ACE inhibitors, or calcium channel blockers) did not influence the results. The authors concluded that treatment with esomeprazole or pantoprazole was not associated with an impaired platelet inhibition response to clopidogrel. Sibbing et al 17 conducted an ex vivo study to evaluate the impact of concomitant PPI and clopidogrel therapy on adenosine diphosphate (ADP)-induced platelet aggregation in coronary artery disease (CAD) patients with previous percutaneous coronary intervention (PCI). Blood samples from 268 consecutive patients treated with dual antiplatelet therapy (clopidogrel and aspirin) were analysed. ADP-induced platelet aggregation was measured by multiple electrode platelet aggregometry. Of the 268 patients assessed, 26.8% were on a PPI (pantoprazole, n = 162; omeprazole, n = 64, or esomeprazole, n = 42). Results are provided in the table below. TABLE III: Median Platelet Aggregation with or without Concomitant Proton Pump Inhibitor Therapy. 17

Treatment

Median Platelet Aggregation (interquartile range) 220.0 (143.8 – 388.8) AUC 295.5 (193.5 – 571.2) AUC 226.0 (150.0 – 401.5) AUC 209.0 (134.8 – 384.8) AUC

No PPI Treatment (n = 732) Omeprazole (n = 64) Pantoprazole (n = 162) Esomeprazole (n = 42)

p-value 0.001 0.69 0.88

PPI: proton pump inhibitor; p-values for individual PPIs vs no PPI; AUC: area under the curve (arbitrary units *min).

The investigators concluded that platelet aggregation was significantly higher in omeprazoletreated patients versus patients not treated with a PPI. Pantoprazole and esomeprazole did not adversely affect platelet response to clopidogrel. 17 Clinical Outcome Data Post hoc Analyses Tsiaousis et al 18 conducted a post hoc analysis of the Combined Clopidogrel and aspirin resistance in Coronary artery stenting study (3 C study) to examine the effect of PPIs on the effectiveness of dual therapy with clopidogrel and aspirin. The investigators evaluated the 1-year incidence of death and MI in 612 consecutive patients with stable or unstable CAD following elective coronary stenting. Of the 612 patients, 59.6% (n = 365) were co-prescribed a PPI. There were no significant differences in baseline characteristics of the two groups related to clinical, angiographic, or coronary stenting related data. Results are provided in the table below. TABLE IV: 1-Year Incidence Rates of Death and Myocardial Infarction. Outcome Concomitant PPI No PPI Treatment n = 247 Treatment n = 365 Incidence of 3.6% 3.2% Death

AstraZeneca UK Limited Horizon Place 600 Capability Green Luton Bedfordshire LU1 3LU

Tel +44 (0) 1582 836000 Fax +44 (0) 1582 838000 www.astrazeneca.co.uk

18

Relative Risk (95% Confidence Interval) 1.1 (0.7 – 1.4) p = 0.8 Registered in England No. 03674842 Registered Office 15 Stanhope Gate London W1K 1LN AstraZeneca UK Limited is a subsidiary company of AstraZeneca PLC

Incidence of Myocardial Infarction

6.3%

6.5%

1 (0.8 – 1.3) p = 0.9

PPI: proton pump inhibitor

There was no difference between treatment groups with regard to death and MI during the first year following coronary stenting. The investigators concluded that PPIs do not have any impact on the effectiveness of anti-platelet drug therapy during the first year following coronary stenting. 18

Dunn et al 19 conducted a post hoc analysis of the CREDO study to assess the effects of baseline PPI use on patients undergoing or at high likelihood of undergoing PCI. Patients were randomised to a loading dose followed by either 1 year or 4 weeks of clopidogrel and daily aspirin. Patients received either clopidogrel + PPI or clopidogrel alone, or placebo + PPI or placebo alone. The primary composite endpoints were death, MI, or urgent target vessel revascularization (UTVR) at 28 days and death, MI, or stroke at 1 year. Cox proportional hazard or logistic regression models were used to identify the risk associated with the use of PPIs in patients taking clopidogrel, placebo, and in the overall population. Results are provided in the table below.

AstraZeneca UK Limited Horizon Place 600 Capability Green Luton Bedfordshire LU1 3LU

Tel +44 (0) 1582 836000 Fax +44 (0) 1582 838000 www.astrazeneca.co.uk

Registered in England No. 03674842 Registered Office 15 Stanhope Gate London W1K 1LN AstraZeneca UK Limited is a subsidiary company of AstraZeneca PLC

TABLE V: 28-Day and 1-Year Rates of Major CV Events in Patients Receiving Clopidogrel with and without a PPI. 28 Days OR (95% CI) p-value 1 Year OR (95% CI) p-value Death/MI/UTVR Death/MI/Stroke Clopidogr el with PPI (n = 176)

Clopidogrel without PPI (n = 877) Placebo with PPI

18/176 (10.3%) 47/877 (5.4%)

Placebo without PPI

23/176 (13.4%) 1.794 (0.997, 3.227)

0.051

18/190 (9.5%)

(n = 190) 65/873 (7.5%)

19

66/877 (7.7%)

1.633 (1.015, 2.627)

0.043

1.554 (1.031, 2.341)

0.035

28/190 (15.0%) 1.400 (0.814, 2.407)

0.221 94/873 (11.1%)

(n = 873)

CV: cardiovascular; PPI: proton pump inhibitor; MI: myocardial infarction; UTVR: urgent target vessel revascularization; OR: odds ratio; CI: confidence interval Results were not presented separately for any specific PPI. The authors concluded that at 1 year, baseline PPI use was associated with an increase in cardiovascular events in patients receiving clopidogrel as well as the overall population. They also concluded the efficacy of clopidogrel was similar in patients taking PPIs compared to those not taking PPIs (clopidogrel reduced adverse events to a similar degree at 1 year in both patient groups). 19 Simon et al evaluated the relationship of allelic variants of genes that modulate clopidogrel’s absorption (ABCB1), metabolic activation (CYP3A5 and CYP2C19), and biologic activity (P2RY12 and ITGB3) with cardiovascular events during 1 year of follow-up. Cardiovascular events included risk of death from any cause, nonfatal stroke, or MI. The analysis included 2208 patients presenting with acute MI and treated with clopidogrel who were consecutively enrolled in a nationwide French Registry of Acute ST-Elevation and Non-ST-Elevation Myocardial Infarction (FAST-MI). The analysis showed that patients with CYP2C19 loss-of-function alleles had an elevated rate of subsequent cardiovascular events. The investigators also assessed the effect of CYP2C19 and CYP2C19*17 and the concomitant prescribing of PPIs on clopidogrel efficacy. The analysis showed that PPI therapy had no effect on the clinical effect of clopidogrel. Results are provided in the table below. TABLE VI: Effect of Concomitant PPI therapy on the clinical effect of clopidogrel. 12 Patients without outcome event Patients with outcome event (n = 1914) (n = 294) Any PPI n (%) Omeprazole n (%)

p-value

1397 (73)

209 (71)

0.75

1006 (53)

141 (48)

0.12

PPI: proton pump inhibitor

The authors concluded that there was no effect of omeprazole or other PPI therapy on the response to clopidogrel in these patients. O’Donoghue et al 20 conducted a post hoc analysis of the TRITON-TIMI 38 study to compare the efficacy of prasugrel with clopidogrel in patients with acute coronary syndrome (ACS) undergoing PCI who were or were not on medications metabolised through the cytochrome P450 system. AstraZeneca UK Limited Horizon Place 600 Capability Green Luton Bedfordshire LU1 3LU

Tel +44 (0) 1582 836000 Fax +44 (0) 1582 838000 www.astrazeneca.co.uk

Registered in England No. 03674842 Registered Office 15 Stanhope Gate London W1K 1LN AstraZeneca UK Limited is a subsidiary company of AstraZeneca PLC

The primary endpoint of the analysis was cardiovascular death, myocardial infarction or stroke. A total of 3,662 (27%) patients from the study were treated with a PPI. PPIs were assessed as a class, and no data are available for individual drugs. There was a benefit with prasugrel over clopidogrel in patients, whether or not they were also on a PPI. The hazard ratio (HR) and 95% confidence interval was 0.81, 0.62 – 1.04 for patients on a PPI and 0.81, 0.68 – 0.95 (P int = 1.00) for patients not on a PPI. The authors concluded that prasugrel improved outcomes in patients after PCI in ACS, irrespective of use of drugs hypothesised to alter clopidogrel metabolism. Retrospective Database Analyses Following the publication of a study by Gilard et al 21 which reported that omeprazole significantly decreased the inhibitory effect of clopidogrel on the platelet P2Y12 receptor, Pezalla et al 22 conducted a retrospective analysis to determine the clinical impact of a drug interaction between clopidogrel and proton pump inhibitors (PPIs). These data were published as a letter to the editor and consequently have not been subjected to peer review. The medical and pharmacy databases at Aetna were examined for cases of acute MI in plan members under the age of 65 who were compliant with their clopidogrel therapy. The patients were studied for a period of 1 year and were divided into 3 groups: no PPI exposure (control group), low PPI exposure, and high PPI exposure. The terms low exposure and high exposure were not defined. PPIs were assessed as a class, and no data are available for individual drugs. The 1-year acute MI rates were 1.38% (66/4,800 patient members) in the control group, 3.08% (22/712) in the low PPI exposure group and 5.03% (number of patient members not provided by the investigators) in the high exposure group (p < 0.05 for the high PPI exposure group vs. the control group). Further analysis of the data revealed significant comorbidity differences between the groups that may account for these findings: more patients in the high PPI exposure group had pre-existing hypertension, diabetes and a slightly elevated overall severity of illness. Therefore, the investigators conducted a subanalysis to control for these differences by selecting a subset of patients from each group who had preexisting ischemic heart disease, congestive heart failure, hypertension, hyperlipidemia, and diabetes prior to starting clopidogrel. The results of this subanalysis are provided in the table below. TABLE VII: 1-Year Acute MI Rates in Risk-Adjusted Patients Receiving Clopidogrel with and without a PPI. Control Group (no PPI) Low PPI Exposure High PPI Exposure Patients with all identified 384 90 536 risk factors, n

Age (years) Male gender, n (%) Patients with acute MI events, n (%) 95% confidence interval

58.24 ± 4.68 257 (66.92) 10 (2.60%)

58.32 ± 5.79 58 (64.4) 9 (10%)

57.12 ± 5.80 365 (68.09) 61 (11.38%)

1.01 – 4.19

3.81 – 16.19

8.69 – 14.07

MI = myocardial infarction; PPI = proton pump inhibitor

The difference in acute MI rates between the control and high PPI exposure groups remained significant (p < 0.05). Noting the limitations of a claims-based analysis, the authors concluded that “evidence is pointing toward a potentially significant interaction between PPIs and clopidogrel that may decrease the ability of clopidogrel to prevent acute MI events.” In a reply to the Letter from Pezzala et al, Gilard et al commented that randomised trials with clinical end points must be performed to answer remaining questions such as: 1) Are all PPIs equal? AstraZeneca UK Limited Horizon Place 600 Capability Green Luton Bedfordshire LU1 3LU

Tel +44 (0) 1582 836000 Fax +44 (0) 1582 838000 www.astrazeneca.co.uk

Registered in England No. 03674842 Registered Office 15 Stanhope Gate London W1K 1LN AstraZeneca UK Limited is a subsidiary company of AstraZeneca PLC

2) When concomitant therapy with clopidogrel and PPIs is necessary, what is the best approach? (i.e. should a vasodilator-stimulated phosphoprotein [VASP] phosphorylation test be performed and the clopidogrel dose adjusted accordingly?). Aubert et al 23 conducted a retrospective cohort study using the National MedCo Integrated Database in order to investigate the potential impact of PPIs on the effectiveness of clopidogrel in preventing coronary artery restenosis. Patients who underwent stent placement at some time during 2005 – 2006, started taking clopidogrel, and were ≥ 80% compliant with their medication for the year following stent placement, were selected for the study (n = 14,383). The patients were followed for the 1-year incidence of major cardiovascular events (hospitalization for stroke, MI, angina, or coronary artery bypass graft [CABG]). The patients were divided into groups based on who took clopidogrel alone (n = 9,862) versus those took clopidogrel with a PPI (n = 4,521), and results were adjusted for baseline differences in age, gender, and comorbidities. The results are shown in the table below. TABLE VIII: 1-Year Incidence Rates of Major CV Events in Patients Receiving Clopidogrel with and without a PPI. Incidence of a major CV event Clopidogrel Clopidogrel + PPI Adjusted OR; CI (n = 9,862) (n = 4,521) No previous CV events 21.2% 32.5% 1.79; 1.62 – 1.97 Previous CV events 26.2% 39.8% 1.86; 1.63 – 2.12 CV = cardiovascular; PPI = proton pump inhibitors; OR = odds ratio; CI = confidence interval

Results were not presented separately for any specific PPI. The authors concluded serious adverse outcomes within 1-year of therapy initiation may result from the drug interaction between clopidogrel and PPIs. Further results extrapolated from the MedCo Integrated database were presented during the Society for Cardiovascular Angiography and Interventions (SCAI) meeting in May 2009. 24 The Clopidogrel Medco Outcomes Study compared a group of 6,828 patients who were concurrently taking a PPI (omeprazole, pantoprazole, esomeprazole or lansoprazole) and clopidogrel to a group of 9,862 patients who were only taking clopidogrel. In the study, patients taking clopidogrel and any of the PPIs experienced a 51% increase in the combined risk of hospitalization for heart attack stroke, unstable angina, or repeat revascularization, including: 70% increase in the risk of heart attack or unstable angina 48% increase in the risk of stroke or stroke-like symptoms 35% increase in the need for a repeat coronary procedure Relative risk calculated for the primary endpoint showed that the use of any of the PPIs in conjunction with clopidogrel increases the risk of CV outcomes: PPI

MACE* rate (%) Omeprazole 25.1 Esomeprazole 24.9 Pantoprazole 29.2 Lansoprazole 24.3

*MACE: major adverse cardiovascular events

AstraZeneca UK Limited Horizon Place 600 Capability Green Luton Bedfordshire LU1 3LU

Hazard ratio p 1.39 1.57 1.61 1.39

Tel +44 (0) 1582 836000 Fax +44 (0) 1582 838000 www.astrazeneca.co.uk

<0.0001 <0.0001 <0.0001 <0.0004

Registered in England No. 03674842 Registered Office 15 Stanhope Gate London W1K 1LN AstraZeneca UK Limited is a subsidiary company of AstraZeneca PLC

If patients were not on clopidogrel, there was no significantly increased risk of CV events in the patients on a PPI versus not on a PPI. There was no statistically significant increased risk for CV events in patients taking H2-blockers. SCAI believes more research is needed on this topic. They also stated that in some patients the use of PPIs may still be warranted, based on the patient's medical problems, and should be taken at the direction of the patient's cardiologist, gastroenterologist, or primary physician. However, given the thousands of patients who receive stents each year, coupled with the significant risks demonstrated in this study, SCAI recommends the use of alternative medications for GI symptoms in patients with stents when appropriate. Ho et al 25 conducted a retrospective cohort study to assess the outcomes of patients taking clopidogrel with or without a PPI following hospitalization for ACS. The investigators analysed the medical records and pharmacy dispensing data of 8,205 patients with acute MI or unstable angina who were prescribed clopidogrel following discharge from 127 Veteran’s Affairs (VA) hospitals between October 1, 2003 and January 31, 2006. The primary outcome of the study was the combined endpoint of all-cause mortality or rehospitalisation for ACS (MI or unstable angina) following the index hospital discharge. Secondary outcomes included: re-hospitalisation for ACS; revascularization procedures, percutaneous coronary intervention, or coronary bypass graft surgery; and all-cause mortality flowing the index ACS hospitalization. Patients with a history of gastrointestinal bleeding prior to, or any bleeding events during, the index hospitalization were excluded from the analysis. Patients prescribed H2-antagonists were also excluded from the analysis. Of the 8,205 patients included in the study, 5244 (63.9%) were prescribed a PPI and clopidogrel upon hospital discharge and/or during follow-up; 2961 (36.1%) were prescribed clopidogrel without a PPI. Patients in the clopidogrel plus PPI group were older, had significantly more comorbidities (including, but not limited to, diabetes, prior MI, heart failure, prior coronary bypass surgery, lower left ventricular ejection fraction, cerebrovascular disease, renal disease, liver disease), and had significantly less low-dose aspirin use after discharge compared to patients in the clopidogrel without PPI group. PPIs included in the study included omeprazole (n = 3132; 59.7%), rabeprazole (n = 151; 2.9%), lansoprazole (n = 22; 0.4%), and pantoprazole (n = 15; 0.2%). A large number of patients received a variety of these PPIs during the follow-up period (n = 1924; 36.7%). Esomeprazole was not included in the study. Median follow-up after the index hospital discharge was 521 days. Results are provided in the table below.

AstraZeneca UK Limited Horizon Place 600 Capability Green Luton Bedfordshire LU1 3LU

Tel +44 (0) 1582 836000 Fax +44 (0) 1582 838000 www.astrazeneca.co.uk

Registered in England No. 03674842 Registered Office 15 Stanhope Gate London W1K 1LN AstraZeneca UK Limited is a subsidiary company of AstraZeneca PLC

TABLE IX: Adverse Outcomes Following Hospital Discharge for Acute Coronary Syndrome (ACS).

No. (%) of Events Clopidogrel Clopidogrel with PPI without PPI (n = 2961) (n = 5244)

Outcome

Unadjusted OR (95% CI)

Adjusted OR (95% CI)a

Primary Outcome Death or rehospitalization for ACS Secondary Outcomes

Rehospitalization for ACS Revascularization procedure Death (all-cause)

615 (20.8)

1561 (29.8)

1.62 (1.45 – 1.80)

1.25 (1.11 – 1.41)

205 (6.9) 353 (11.9) 493 (16.6)

764 (14.6) 815 (15.5) 1042 (19.9)

2.29 (1.95 – 2.69) 1.36 (1.19 – 1.55) 1.24 (1.10 – 1.40)

1.86 (1.57 – 2.20) 1.49 (1.30 – 1.71) 0.91 (0.80 – 1.05)

PPI: proton pump inhibitor; OR: odds ratio; CI: confidence interval a: adjusted for all baseline characteristics except male sex.

Multivariable analysis (adjusted odds ratios) revealed that concomitant clopidogrel and PPI therapy at any time point during follow-up was associated with an increased risk of death or rehospitalization for ACS (primary endpoint); rehospitalization for ACS and revascularization procedure (secondary endpoints). Concomitant treatment with clopidogrel and PPI was not associated with an increased risk of all-cause mortality. The use of PPI without clopidogrel was not associated with death or rehospitalization for ACS following the index hospital discharge (adjusted OR with 95% CI: 0.98; 0.85 – 1.13). In addition, there was no association between PPI dose (omeprazole and rabeprazole) and adverse outcomes. The number of patients who received lansoprazole and pantoprazole was too small to evaluate a possible association between dose and adverse outcomes. According to the authors, some possible limitations of the study include: unavailability of information regarding the use of over-the-counter medications, unavailability of medical treatment or prescription data outside the VA system, and inherent limitations with an observational study design. The authors concluded that the findings of this retrospective database analysis suggest that the use of PPI may be associated with the attenuation of benefits of clopidogrel after ACS. Juurlink et al 26 conducted a population-based nested case-control study using the records of the Ontario Public Drug Program in Ontario, Canada to evaluate the potential drug interaction between PPIs and clopidogrel. Patients aged ≥ 66 years old who filled a prescription for clopidogrel between April 1, 2002 and December 31, 2007 within 3 days of hospital discharge following an acute MI were included in the cohort. The investigators followed patients who received clopidogrel for a maximum of 90 days following hospital discharge or until they were readmitted for another acute MI. Exposure to PPIs was defined as current (within 30 days), previous (within 31 – 90 days), or remote (within 91 – 180 days). Of 13, 636 patients who were identified, 782 of them were readmitted for an acute MI within 90 days of their previous hospital discharge. Of the 782 patients, 734 of them were matched to at least one control. The investigators found a significant association between current use of PPIs and readmission for an acute MI (adjusted OR 1.27, 95% CI 1.03 – 1.57). There was no association between previous or remote use of a PPI and readmission for acute MI. When the investigators stratified the results by PPI, current use of omeprazole, lansoprazole and rabeprazole were associated with readmission for acute MI (adjusted OR 1.40, 95% CI 1.10 – 1.77), but pantoprazole was not (adjusted OR 1.02 95% CI 0.70 – 1.47). Esomeprazole was not included in the study. Ramirez et al 27 evaluated the effect of proton pump inhibitor and clopidogrel combination therapy on adverse clinical outcomes after percutaneous coronary intervention (PCI). Data from the University of Pittsburgh NHLBI (National Heart, Lung, and Blood Institute) Dynamic Registry waves 4 (2004) and 5 (2006) was analysed. Patients (n = 535) discharged alive after their index PCI and prescribed clopidogrel were evaluated. The patients were divided into two groups: PPI users (n = 138; 25.8%) and nonusers (n = 397; 74.2%). Data regarding rates of death, AstraZeneca UK Limited Horizon Place 600 Capability Green Luton Bedfordshire LU1 3LU

Tel +44 (0) 1582 836000 Fax +44 (0) 1582 838000 www.astrazeneca.co.uk

Registered in England No. 03674842 Registered Office 15 Stanhope Gate London W1K 1LN AstraZeneca UK Limited is a subsidiary company of AstraZeneca PLC

myocardial infarction (MI), and repeat revascularization 1-year post-PCI were collected. Baseline characteristics were similar between the two patient groups (age, baseline prevalence of diabetes, hypertension, renal dysfunction, smoking habits, or procedural success). PPIs were assessed as a class, and no data are available for individual drugs. Results are provided in the table below. TABLE X: Cumulative 1-Year Outcomes in Patients Receiving Clopidogrel with and without Proton Pump Inhibitor (PPI). Adverse Outcome No PPI PPI p-value Death

Myocardial Infarction Coronary Artery Bypass Graft Death/Myocardial Infarction Repeat percutaneous coronary intervention Repeat Revascularization

(n = 397) 5.9% 4.2% 4.1% 9.6% 10.1%

(n = 138) 3.0% 3.7% 3.1% 6.7% 13.4%

0.18 0.83 0.53 0.32 0.23

14.2%

15.8%

0.65

There were no differences between treatment groups with regard to univariate rates of death, MI, death/MI, or repeat revascularization at one year. The authors concluded that in patients undergoing PCI and discharged on clopidogrel therapy, the concomitant use of PPI’s did not result in adverse cardiovascular outcomes at one year. Given the scientific limitations and contradictory nature of data currently available, AstraZeneca agrees that more research is needed before any conclusions can be drawn about a potential interaction between clopidogrel and PPIs. AstraZeneca's ongoing monitoring of clinical study data and adverse event reports from marketed use, have not raised safety concerns to date on the use of clopidogrel in conjunction with Nexium® or Losec®. It is the position of AstraZeneca that its PPI products, Nexium® and Losec® should be used according to their UK product licence, inclusive of indications and dosing recommendations, with allowances for variations in these in individual circumstances under the supervision of physicians and other health care professionals.

1

January 26, 2009 early communication about an ongoing safety review of clopidogrel bisulfate (marketed as Plavix). Available at: Hhttp://www.fda.gov/cder/drug/early_comm/clopidogrel_bisulfate.htmH. Accessed February 2, 2009. 2 Summary of Product Characteristics for Nexium 3 US Prescribing Information for Nexium. 4 Summary of Product Characteristics for Zoton Tabs, Wyeth Pharmaceuticals 5 Summary of Product Characteristics for Losec 6 US Prescribing Information, Protonix. Wyeth Pharmaceuticals. 7 Summary of Product Characteristics for Pariet. Eisai Co. Ltd. 8 American College of Cardiology Foundation Task Force on Clinical Expert Consensus Documents, Deepak L. Bhatt, James Scheiman, Neena S. Abraham, et al. ACCF/ACG/AHA 2008 Expert Consensus Document on Reducing the Gastrointestinal Risks of Antiplatelet Therapy and NSAID Use. J Am Coll Cardiol. 2008;52(18):1502–1517. 9 Lind T, Rydberg L, Kylebäck A, et al. Esomeprazole provides improved acid control vs. omeprazole in patients with symptoms of gastro-esophageal reflux disease. Aliment Pharmacol Ther. 2000;14:861-867. 10 Äbelö A, Andersson TB, Antonsson M, et al. Stereoselective metabolism of omeprazole by human cytochrome P450 enzymes. Drug Metabol Disp. 2000;28:966-972. 11 Summary of Product Charcteristics for Plavix, Sanofi Pharma Bristol-Myers Squibb SNC 12 Simon T, Verstuyft C, Mary-Krause M, et al. Genetic determinants of response to clopidogrel and cardiovascular events. N Engl J Med. 2009;360(4):363-375. AstraZeneca UK Limited Horizon Place 600 Capability Green Luton Bedfordshire LU1 3LU

Tel +44 (0) 1582 836000 Fax +44 (0) 1582 838000 www.astrazeneca.co.uk

Registered in England No. 03674842 Registered Office 15 Stanhope Gate London W1K 1LN AstraZeneca UK Limited is a subsidiary company of AstraZeneca PLC

13

Frere C, Cuisset T, Morange PE, et al. Effect of cytochrome P450 polymorphisms on platelet reactivity after treatment with clopidogrel in acute coronary syndrome. Am J Cardiol. 2008;101:1088-1093. 14 Trenk D, Hochholzer W, Fromm MF, et al. Cytochrome P450 2C19 681G>A polymorphism and high on-clopidogrel platelet reactivity associated with adverse 1-year clinical outcome of elective percutaneous coronary intervention with drug-eluting or bare-metal stents. J Am Coll Cardiol. 2008;51(20):1925-1934. 15 Collet JP, Hulot JS, Pena A, et al. Cytochrome P450 2C19 polymorphism in young patients treated with clopidogrel after myocardial infarction: a cohort study. Lancet. 2009;373:309-317. 16 Siller-Matula JM, Spiel AO, Lang IM, et al. Effects of pantoprazole and esomeprazole on platelet inhibition by clopidogrel. Am Heart J. 2009;157(1):148.e1 - 148.e5. 17 Sibbing D, Morath T, Stegherr J, et al. Impact of proton pump inhibitors on platelet response to clopidogrel treatment [Abstract]. J Am Coll Cardiol. 2009;53(10 suppl 1):A67, Abs 2515-736. 18 Tsiaousis GZ, Zairis MN, Patsourakos N, et al. Oral proton pump inhibitors and their impact on the effectiveness of dual antiplatelet therapy during the first year after elective coronary stenting [Abstract]. J Am Coll Cardiol. 2009;53(10 suppl 1):A335, Abs 1041-125. 19 Dunn SP, Macaulay TE, Brennan DM, et al. Baseline proton pump inhibitor use is associated with increased cardiovascular events with and without the use of clopidogrel in the CREDO trial [Abstract]. Circulation. 2008;118(18):S_815, Abs 3999. 20 O’Donoghue M, Wiviott SD, Murphy SA, et al. Absence of an interaction between drugs metabolized by cytochrome P450 enzymes and the benefit of treatment with prasugrel versus clopidogrel in patients with acute coronary syndromes undergoing percutaneous intervention: a TRITON-TIMI 38 analysis [Abstract]. J Am Coll Cardiol. 2009;53(10 suppl 1):A328, Abs 1032-143. 21 Gilard M, Arnaud B, Le Gal G, et al. Influence of omeprazole on the antiplatelet action of clopidogrel associated to aspirin. J Thromb Haemost. 2006;4:2508-2509. 22 Pezalla E, Day D, Pulliadath I. Initial assessment of clinical impact of a drug interaction between clopidogrel and proton pump inhibitors [Letter]. J Am Coll Cardiol. 2008;52(12):1038-1039. 23 Aubert RE, Epstein RS, Teagarden JR, et al. Proton pump inhibitors effect on clopidogrel effectiveness: the clopidogrel Medco outcomes study [Abstract]. Circulation. 2008;118(18):S_815, Abs 3998. 24 Stanek EJ et al Clopidogrel Medco Outcomes study Hhttp://medco.mediaroom.com/index.php?s=43&item=376H Hhttp://www.theheart.org/article/967075.doH 25 Ho PM, Maddox TM, Wang L, et al. Risk of adverse outcomes associated with concomitant use of clopidogrel and proton pump inhibitors following acute coronary syndrome. J Am Med Assoc. 2009;301(9):937-944. 26 Juurlink DN, Gomes T, Ko DT, et al. A population-based study of the drug interaction between proton pump inhibitors and clopidogrel. Can Med Assoc J. 2009;180(7):DOI:10.1503/cmaj.082001. 27 Ramirez JF, Selzer F, Chakaprani R, et al. Proton pump inhibitor and clopidogrel combination is not associated with adverse clinical outcomes after PCI: the NHLBI dynamic registry. J Am Coll Cardiol. 2009;53 (10 suppl 1):A27, Abs 2903-7.

AstraZeneca UK Limited Horizon Place 600 Capability Green Luton Bedfordshire LU1 3LU

Tel +44 (0) 1582 836000 Fax +44 (0) 1582 838000 www.astrazeneca.co.uk

Registered in England No. 03674842 Registered Office 15 Stanhope Gate London W1K 1LN AstraZeneca UK Limited is a subsidiary company of AstraZeneca PLC