FDA Label Change Fails to Stem Uncertainty Over PPI Use with Clopidogrel

Three years after the US Food and Drug Administration (FDA) first raised a red flag about concomitant use of clopidogrel and the proton pump inhibitor (PPI) omeprazole, physicians are still at odds over whether the advice should influence clinical practice. A recent survey of US cardiologists by TCTMD reveals that most do not believe the interaction matters and many have not changed how they treat patients.

Clopidogrel (Plavix; Bristol-Myers Squibb/Sanofi Pharmaceuticals, Bridgewater, NJ) is a prodrug activated by the cytochrome P450 system, in which the gene CYP2C19 plays a role in determining active metabolite levels. Simultaneous use of other drugs metabolized through the same pathway, such as PPIs, has been shown in pharmacodynamic studies to reduce clopidogrel’s efficacy. And while observational studies have provided mixed evidence, no randomized controlled trials have demonstrated an effect on clinical outcomes.

In a telephone interview with TCTMD, Deepak L. Bhatt, MD, of Brigham and Women's Hospital (Boston, MA), shared what had inspired him to summarize the possible drug-drug interaction for clinicians in a paper published January 17, 2012, in Circulation. Dr. Bhatt served as lead investigator of the Clopidogrel and the Optimization of Gastrointestinal Events Trial (COGENT), the only large randomized trial to take on the issue.

“The topic remains timely—and confusing and controversial,” he related. “That is, doctors [including not only cardiologists but] primary care physicians and gastroenterologists, really are uncertain what to make of the . . . mountain of data out there regarding proton pump inhibitors and antiplatelet therapy.”

Neal S. Kleiman, MD, of the Methodist DeBakey Heart and Vascular Center (Houston, TX), told TCTMD in a telephone interview that physicians do in fact listen to the FDA’s guidance and that he also hears from patients who “have read something here or there and are concerned about it.” Yet he expressed doubts over whether the label change was justified.

The ‘Mountain of Data’

As early as 2006, an observational study of 105 patients documented a reduction in the biological action of clopidogrel via vasodilator-stimulated phosphoprotein (VASP) assay (Gilard M. J Thromb Haemost. 2006;4:2508-2509). “Demonstrating an interaction between these two drugs, which are so routinely associated in cardiology, would have major implications for the management of the risk of thrombosis following angioplasty, but will require the results of a randomized study that is currently under way,” the paper notes.

The subsequent OCLA (Omeprazole CLopidogrel Aspirin) trial, by the same group of researchers, was published 2 years later in 2008, again showing evidence that omeprazole impairs clopidogrel metabolism (Gilard M. J Am Coll Cardiol. 2008;51:256-260).

Yet in nearly 4,000 patients, the randomized COGENT trial found no adverse clinical interaction between clopidogrel and omeprazole in patients receiving dual antiplatelet therapy for ACS and/or PCI. The study also showed that the PPI actually offers benefit by substantially reducing upper gastrointestinal bleeding compared with placebo (Bhatt DL. N Engl J Med. 2010; 363:1909-1917).

And as recently as April 2012, another study—this time randomizing patients to a variety of PPIs in a cross-over design—found that coadministration of dexlansoprazole or lansoprazole reduced levels of clopidogrel active metabolite and inhibited platelet function on VASP assay to a lesser extent compared with esomeprazole or omeprazole (Frelinger AL III. J Am Coll Cardiol. 2012;59:1304-1311.)

Many Cardiologists Unconvinced But Taking Action

Of 98 respondents to the TCTMD survey, 95% were aware that in 2009 the FDA updated its label for clopidogrel. Specifically, the label states: “Concomitant use of drugs that inhibit CYP2C19 (eg, omeprazole) should be discouraged.”

Both Drs. Bhatt and Kleiman—along with only 38% of those surveyed by TCTMD—reported being unaware at the time of their interviews that the agency expanded the clopidogrel label’s scope late last year to include the PPI esomeprazole.

As of December 2011, the label states: “Avoid using omeprazole or esomeprazole with Plavix. Omeprazole and esomeprazole significantly reduce the antiplatelet activity of Plavix. When concomitant administration of a PPI is required, consider using another acid-reducing agent with minimal or no CYP2C19 inhibitory effect on the formation of clopidogrel active metabolite.”

A full 81% percent of physicians said that they did not believe PPIs can cause a clinically meaningful reduction in clopidogrel’s efficacy. Yet many are still cautious, with nearly 38% of respondents acknowledging that the FDA labeling has affected their clinical strategy.

Some Say Association ‘Spurious’

The trouble now is the lack of decisive evidence, said Dr. Kleiman.

With the retrospective and observational analyses, “it’s probably impossible to take out all the confounders,” he commented. “We have theoretical reasons there might be an interaction. And then we have a randomized study—which was incomplete, underpowered, and with short follow-up because they ran out of money—that looked like it was headed in the opposite direction.”

When asked how heavily to weigh the in vitro evidence, which he characterized as “pretty convincing,” Dr. Kleiman answered, “Despite what you read about [functional testing], we don’t know what the real threshold is [for actually increasing events, or] if there is a threshold. . . . We have a pretty good idea, but we don’t know [if it’s consistent] in all patients.”

Considering all these issues “gets pretty complex,” he said, “and we’re pretty good at oversimplifying it and making the case more strongly than there’s evidence to support. I don’t even know that platelet aggregation is the best thing to look at, [though r]ight now, it’s the only thing to look at. The in vitro story is pretty consistent but its clinical implications are pretty inconsistent.”

Dr. Bhatt also decribed the ex vivo evidence as “clear.”

“I don’t think that’s debatable,” he said. The question is, . . . does it translate into clinical differences? Within the context of COGENT, and its limited statistical power for cardiovascular events, we didn’t see any signal of that. But that doesn’t mean that if someone had done a 30,000-patient study for 1 year that there wouldn’t be some signal. . . . So I’m not dismissing this interaction either.”

Sorin J. Brener, MD, of Weill Cornell Medical College (New York, NY), told TCTMD in an interview that the “easy explanation” for the apparent risk in observational trials is that patients taking PPIs were sicker to begin with. He noted that the PPI-clopidogrel saga serves as an example of how one can “be fooled by spurious associations.”

In a telephone interview, Shaun G. Goodman, MD, MSc, of St. Michael’s Hospital (Toronto, Canada), agreed. He recently coauthored a subanalysis of the PLATO trial that found PPIs are associated with increased risk of cardiovascular events not only when paired with clopidogrel but also with the newer antiplatelet agent ticagrelor, which had been thought immune to such an interaction (Goodman SG. Circulation. 2012;125:978-986).

“It’s a good reminder—this observational study just like the other ones—that we have to be really cautious about making, as we tend to do, a leap from an association where something might be a marker or confounding variable to a true cause and effect relationship,” he said.

While Dr. Goodman acknowledged that it is probably fair to “err on the side of caution” in the face of observational evidence backing an interaction between certain PPIs and clopidogrel, he described the actions of the US FDA and other regulatory bodies around the world as “knee jerk.” The pendulum is now swinging toward an understanding that the issue is more complex than previously thought, he said.

However, all the physicians interviewed by TCTMD said the prospect of a dedicated clinical trial to settle the debate is unlikely.

How Much Evidence Is Enough?

“The truth is,” said Dr. Kleiman, “I’m not convinced that the label change [on PPIs or the ‘black box’ warning about the influence of CYP2C19 on clopidogrel metabolism in general are] justified. But you have to ask yourself, if you’re the FDA, what level of evidence do you need on a safety issue to make a statement? And what level of evidence do you need to make a strong statement like a boxed warning? When there’s a lot of pressure on [the agency] to scrutinize for side effects that are recognized after a drug is approved, it just may be that their threshold is low.

“From a social point of view, I understand that. But . . . is the highest level of evidence met? No,” he continued, adding, “So depending on what you perceive your job to be, how you define truth is variable.”

In a telephone interview, Mary Ross Southworth, PharmD, deputy director of the Division of Cardiovascular & Renal Products at the FDA’s Center for Drug Evaluation & Research (Silver Spring, MD), explained how the agency arrived at its warnings.

“When clopidogrel was first approved in 1997, we didn’t know a lot about how it was processed in the body . . . into its active metabolite,” she said. CYP2C19 only entered the conversation 10 years later, and eventually as evidence began building on the PPI omeprazole, the FDA issued its first public communication in January 2009 prior to any labeling changes.

The agency also asked clopidogrel’s manufacturer to conduct additional studies on omeprazole. When the results came out in late 2009, the FDA formally updated the label (Angiolillo DJ. Clin Pharmacol Ther. 2011;89:65-74). Additional industry-conducted research has come out over the past 3 years, such that the most recent label contains results from drug interaction studies on dexlansoprazole, lansoprazole, and pantoprazole; all appear to have less effect on clopidogrel’s antiplatelet activity than omeprazole.

According to Dr. Southworth, esomeprazole was added to the label because it is 1 of the 2 enantiomers—mirror-image chemical structures—that comprise omeprazole and, as such, may behave similarly.

The FDA message is not a blanket statement against coadministration, she stressed. “We know PPIs are beneficial and they’re needed in some people. Just be prudent about which ones you choose,” Dr. Southworth advised.

Importantly, studies have shown that even when clinicians stagger omeprazole and clopidogrel doses throughout the day, antiplatelet activity is still impaired. “That’s why we’re so definitive about [our recommendations], because it doesn’t seem like there’s much you can do to get rid of the risk,” Dr. Southworth noted.

As to whether the FDA would ever revoke its warning, she seemed doubtful. “Although we are always interested in looking at new data, the interaction is pretty well characterized at this point, so I don't see too much changing unless another PPI [manufacturer] does another study and we decide to put that in the label,” she said.

“But right now, you have several options other than omeprazole [and esomeprazole] to choose from, so it seems like a reasonable situation to be in,” Dr. Southworth concluded.

Respondents Express Frustration

A few physicians who responded to the TCTMD survey indicated they completely avoid using PPIs, or at least the ones in question, in combination with clopidogrel.

“I have concerns based on the observational data. The FDA warning does not change that,” one commented. “The recent report from the European registry examining predictors of stent thrombosis has added to my concerns, although the PLATO analysis suggests PPIs may be a marker of risk rather than acting directly through clopidogrel. The question remains unsettled.”

Many physicians reported they saw no firsthand signs of an interaction but felt obligated to alter their practice. Several, however, were far less sanguine.

“While not convinced that there are any real clinical consequences, I have changed my prescribing pattern to reflect these recommendations,” said one respondent.

“I believe their [recommendations] to be in error,” replied a different doctor. Even so, “I try to avoid certain PPI Rx with Plavix because it raises too many questions from [referring physicians], pharmacies, and patients.” Another griped: “FDA is full of sound and fury . . . symphony of idiots.”

Complicating matters, “patients prefer less costly omeprazole to other PPIs,” said a respondent, who cited the mixed evidence of an effect. “FDA statement seems premature and unwarranted in retrospect. Is its statement on statins and memory loss a similar mistake?” Others suggested H-2 blockers as an alternative to PPIs.

Malpractice is also a concern, wrote another respondent: “Even though I believe there is not an issue, given the government warning, I am worried if a problem occurred I would be [liable].”

For his part, Dr. Brener asserted that the label is no longer relevant because there are now alternative antiplatelet drugs. “Prasugrel, even though it’s a prodrug, is very little affected by CYP2C19. Ticagrelor and the newer drugs like elinogrel are not at all, because they don’t have anything to do with CYP2C19,” he commented.

Involving the Patient

Dr. Goodman said that it often comes down to having a discussion with the patient. Physicians can say, “’Here’s the information that’s available. . . . If you need to be on both of these agents, here’s the potential upside and here’s the potential downside, but we don’t know for sure,” he related. “I think that [now] there’s a little flexibility in clinical practice, whereas in the beginning patients were calling to ask if they should be dropping this class of drugs completely. But [the label] is there in black and white, and it is hard to ignore. You worry about litigation and a number of things, even if you’ve [been upfront] with your patients.”

The FDA label, Dr. Bhatt said, “certainly has left physicians a bit confused . . . In fairness, there was a lot of indiscriminate PPI use [before], not necessarily by cardiologists but just in patients coming into the hospital [more generally.] PPIs were really being prescribed like candy for a while, and in some cases it wasn’t clear what the indication was.”

Dr. Bhatt noted that “a little bit of restraint is never bad with medications, because they can always have side effects.” Beyond their potential interaction with clopidogrel, PPIs may affect bone health and infection risk, he mentioned.

Similarly, Dr. Kleiman said that clinicians “ought to be careful all the time. If people don’t need a drug you shouldn’t give it to them. . . . No one likes to take extra medicines.”

 

---

Sources:
1. Moukarbel GV, Bhatt DL. Antiplatelet therapy and proton pump inhibition: Clinician update. Circulation. 2012;125:375-380.

2. US Food and Drug Administration. 2011 clopidogrel label. http://www.accessdata.fda.gov/drugsatfda_docs/label/2011/020839s055lbl.pdf. Published December 20, 2011. Accessed April 16, 2012.

3. US Food and Drug Administration. 2009 clopidogrel label. http://www.accessdata.fda.gov/drugsatfda_docs/label/2009/020839s040lbl.pdf. Published May 5, 2009. Access April 16, 2012.

 

 

Related Stories:

• PLATO Analysis: PPIs Problematic with Both Ticagrelor, Clopidogrel
• Consensus Statement Supports Judicious Use of PPIs Plus Antiplatelet Therapy
• COGENT Published: No Adverse Interaction Between Omeprazole, Clopidogrel