Clopidogrel-PPI Interaction: Mechanistic Studies That Led to FDA Warning Now Published

Sue HughesSue Hughes

September 21, 2010

September 21, 2010 (Jacksonville, Florida) — The pharmacokinetic/dynamic studies that contributed to the FDA's decision to request a public-health warning about the interaction between clopidogrel and omeprazole have now been published [1].

The data, published online on September 15, 2010 in Clinical Pharmacology and Therapeutics, consist of four randomized crossover studies in healthy volunteers, which show a significant interaction between clopidogrel and omeprazole on the amount of clopidogrel active metabolite formed and the degree of platelet aggregation. This interaction was not mitigated by staggering the doses of the two drugs or by raising the clopidogrel dose. However, pantoprazole had much less of an interaction with clopidogrel.

Lead author Dr Dominick Angiolillo (University of Florida College of Medicine, Jacksonville) commented to heartwire : "Our results suggest that there is clearly an interaction with omeprazole that supports the boxed warning. Patients who need some gastric protection when taking clopidogrel should therefore avoid omeprazole but can take pantoprazole or a non-PPI agent such as ranitidine."

He added: "We need to think more carefully about whether a patient on clopidogrel actually needs some gastric protection before prescribing any such agent. But if they do, our data should reassure physicians that there is a PPI (pantoprazole) that can be used safely."

Enough to Convince the Skeptics?

But cardiologists who have previously voiced skepticism over the clopidogrel-PPI interaction still remain unconvinced that these data translate into any clinically relevant effect.

Summarizing these concerns, Dr Michelle O'Donoghue (Brigham and Women's Hospital, Boston, MA) told heartwire : "I believe the data are now very compelling that certain PPIs, in particular omeprazole, appear to attenuate some of the in vitro antiplatelet effects of clopidogrel. Dr Angiolillo's work elegantly demonstrates this interaction and provides more evidence to suggest that this interaction rests at the level of the CYP2C19 enzyme. However, the question that continues to weigh on the minds of clinicians is whether this in vitro interaction translates into a higher risk of cardiovascular events in clopidogrel-treated patients. Although it seems intuitive that the pharmacodynamic effects and clinical benefit of clopidogrel should move in tandem, this has not always been found to be the case."

As background, Angiolillo explained to heartwire : "Because clinical studies have shown mixed and conflicting results about a reaction between clopidogrel and PPIs, the FDA requested that Bristol-Myers Squibb conduct pharmacokinetic/dynamic studies to the most rigorous scientific standards. These are those studies."

The studies involved 282 healthy subjects. Four different studies were conducted, all crossover in design with placebo controls:

  • Clopidogrel (300-mg loading dose/75-mg/day maintenance dose) and omeprazole (80 mg) administered simultaneously.

  • Clopidogrel (300-mg loading dose/75-mg/day maintenance dose) with omeprazole (80 mg) given 12 hours later.

  • Increasing the clopidogrel dose to 600-mg loading/150-mg/day maintenance and omeprazole (80 mg) administered simultaneously.

  • Clopidogrel (300-mg loading dose/75-mg/day maintenance dose) and pantoprazole (80 mg) administered simultaneously.

Results (compared with clopidogrel alone) showed that exposure of the active metabolite of clopidogrel, clopi-H4, was consistently decreased by 40% to 47% when clopidogrel and omeprazole were coadministered, and this was associated with increases in the exposure of unchanged clopidogrel of 37% to 51%. Similarly, the maximal platelet aggregation (MPA) induced by ADP and vasodilator-stimulated phosphoprotein (VASP)–platelet-reactivity index (PRI) were significantly increased and the inhibition of platelet aggregation (IPA) response to clopidogrel decreased when omeprazole was coadministered.

The differences in the exposure of clopi-H4 and platelet response were observed irrespective of the timing of omeprazole administration. Although doubling of the clopidogrel doses to 600 mg/150 mg increased the exposure of clopi-H4, significant differences in exposure and platelet response were still observed.

When clopidogrel and pantoprazole were coadministered, a smaller, although still significant, decrease in the exposure of clopi-H4 was observed, as were smaller differences in MPA, IPA, and VASP-PRI.

Results of Clopidogrel-PPI Interaction Studies

Outcome Study 1:

Clopidogrel 300/75 mg, omeprazole 80 mg

Study 2:

Clopidogrel 300/75 mg, omeprazole 80 mg (12 h later)

Study 2:

Clopidogrel 600/150 mg, omeprazole 80 mg

Study 4:

Clopidogrel 300/75 mg, pantoprazole 80 mg

Reduction in clopidogrel active metabolite (%) 40 47 41 14
Increase in maximal platelet aggregation induced by 5-µmol/L ADP (%) 8.0 5.6 8.1 4.3
Increase in VASP-PRI (%) 20.7 27.1 19.0 3.9

The authors point out that the different results for omeprazole and pantoprazole suggest that the interaction between clopidogrel and omeprazole is not due to decreased clopidogrel absorption secondary to an elevation of gastric pH consequent to the action of the PPI but rather a metabolic drug–drug interaction at the site of the CYP2C19 enzyme, with clopidogrel as the victim and omeprazole as the perpetrator.

Addressing the high doses of PPI used in the studies, Angiolillo commented to heartwire : "We did this because we wanted to truly address if there was any interaction. If we used a lower dose we might have missed it. There are already other data suggesting an interaction with lower doses, and more specific studies are ongoing."

Are These Results Clinically Relevant?

Others who reviewed the new studies for heartwire all appeared to remain unconvinced of the clinical relevance of the new data.

Dr Peter Berger (Geisinger Health Systems, Danville, PA) said he was "confused" by the new studies, as they would indeed suggest that a clinically relevant interaction between clopidogrel and omeprazole might be present. But he noted that there was "no hint" of any interaction between clopidogrel and omeprazole taken at the same time in the >3600 patient COGENT trial, and no interactions were evident between PPIs and clopidogrel in subgroup analyses of the PLATO and TRITON randomized trials. "I believe that more has to be learned about the relationship between ex vivo platelet-function testing and clinical outcome before it should be used to guide clinical practice," he commented.

I believe that more has to be learned about the relationship between ex vivo platelet-function testing and clinical outcome before it should be used to guide clinical outcome.

Berger added that he was even trying to figure out whether a PPI ought to be prescribed more frequently, even prophylactically, to patients on antiplatelet therapy. "There are now three randomized placebo-controlled trials that have shown many fewer adverse [gastrointestinal] GI side effects when a PPI is administered with aspirin (two studies) or with aspirin and clopidogrel (one study). And GI side effects are not only uncomfortable or even painful, they are a common cause of noncompliance with aspirin," he said.

O'Donoghue also pointed out that other drugs, including atorvastatin, have been shown to attenuate some of the antiplatelet effects of clopidogrel and yet this interaction does not appear to have clinical significance. She concedes: "Given that pantoprazole is available and is not a strong inhibitor of CYP2C19, I agree that it appears to be a cautious alternative for clopidogrel-treated patients requiring a PPI." But she added: "We need to pursue ongoing studies to better understand the nature of the relationship between platelet-function testing and clinical outcomes. This will be of critical importance as more clinicians begin to incorporate platelet-function testing into their practices."

I am not altering my antiplatelet strategies due to these . . . observations.

Dr Deepak Bhatt (VA Boston Healthcare System, MA) also voiced caution about making clinical decisions based on these studies: "This is a very carefully done study, and I believe the results. I think the bigger issue, though, is that while PPIs and CYP450 genetic polymorphisms affect the response to clopidogrel, the intrinsic variability of clopidogrel is greater than that due to either of these factors. Therefore, I think their clinical impact on a population level is minimal."

He added: "The idea of tailoring therapy based on an individual's platelet response remains appealing but also unproven. So, for the time being, I am not altering my antiplatelet strategies due to these pharmacokinetic/pharmacodynamic observations, nor am I changing around the PPI regimens on my patients (although I do try to make sure they really need to be on a PPI at all)."

Bhatt also questioned whether these results with 80 mg of omeprazole could be extrapolated to the 20-mg dose used most often by cardiologists or primary-care physicians. But he said: "If a doctor is really worried about it, these data suggest pantoprazole is unlikely to interact with clopidogrel. There is, however, always a risk of switching medications around in a patient who is doing well--it can create confusion, dosing errors, and lead to nonadherence."

Applicable to High-Risk Patients Only?

In response to these comments, Angiolillo said he agreed that the clinical implications of these data were not clear and the variability of clopidogrel may mean that there will be an impact in some patients but not in others. He suggested that the clinical impact may be applicable only to high-risk patients, which was also suggested in a recent meta-analysis by Dr Gilles Montalescot (Hôpital Pitié-Salpêtrière, Paris, France) et al.

The studies were funded by Sanofi-Aventis and Bristol-Myers Squibb. Angiolillo reports receiving honoraria/consulting fees/research grants from Bristol Myers Squibb, Sanofi-Aventis, Eli Lilly, Daiichi Sankyo, the Medicines Company, Portola, Novartis, Medicure, Accumetrics, Arena Pharmaceuticals, AstraZeneca, GlaxoSmithKline, Otsuka, Schering-Plough, Eisai, and Johnson & Johnson.  Several coauthors of the paper are employees of Sanofi-Aventis or Bristol-Myers Squibb. O'Donoghue reports grants or honoraria from GlaxoSmithKline, Eisai, Eli Lilly, and Daiichi Sankyo. Berger has served as a consultant to AstraZeneca, Boehringer Ingelheim, Eli Lilly/Daiichi-Sankyo, Medicure, and Ortho McNeil and has received research funding for Geisinger Clinic from Thrombovision, Helena, Accumetrics, AstraZeneca, Haemoscope, the Medicines Company, and Corgenix/Aspirinworks; he owns equity in Lumen. Bhatt has previously reported receiving institutional research support from AstraZeneca, Bristol-Myers Squibb, Eisai, Ethicon, Heartscape, Sanofi-Aventis, and the Medicines Company.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.