Mouse study provides possible mechanism for cardiac risks with COX-2 inhibitors

April 19, 2002

Fri, 19 Apr 2002 17:00:00

Philadelphia, PA - A new study in mice could help shed light on why COX-2 inhibitors might increase the risk of thrombotic events. The chief investigator, pharmacologist Dr Garret A Fitzgerald (University of Pennsylvania), says: "Our results afford a credible mechanism by which a cardiovascular hazard might occur in that minority of individuals predisposed to thrombosis." However, he stressed to heartwire , "a mouse study is not reason enough for people to stop taking COX-2 inhibitors," and that much more clinical work would be required to determine whether the murine data can be extrapolated into humans.

Reporting their findings in Science, Fitzgerald, together with first author Dr Yan Cheng (University of Pennsylvania) and colleagues, explain how they wanted to clarify the interplay of the two products of the cyclooxygenases - thromboxane (TxA2) and prostacyclin (PGI2) - in the cardiovascular system.

COX-1, the form of cyclooxygenase found in platelets, makes TxA2, which causes blood vessels to constrict and platelets aggregate - an early step in clot formation. COX-2, by contrast, is expressed in blood vessels and is a major source of PGI2, which dilates blood vessels and prevents the activation of platelets.

 

A mouse study is not reason enough for people to stop taking COX-2 inhibitors.

 

Concerns about the potential prothrombotic effects of COX-2 inhibitors first arose with rofecoxib (Vioxx® - Merck) in the Vioxx Gastrointestinal Outcomes Research (VIGOR) trial, when the rate of MIs in patients taking rofecoxib was higher than in those taking the comparator NSAID naproxen. Merck had long claimed that this was because naproxen is cardioprotective; however, a meta-analysis published in JAMA last year found that both rofecoxib and celecoxib (Celebrex® - Pharmacia) may increase the risk of cardiovascular thrombotic events, and concluded that further prospective randomized controlled trials in cardiovascular patients needed to be done to clarify this issue. Many experts were critical of this meta-analysis, however, including Fitzgerald, who told heartwire at the time that it was "a statistically dubious, retrospective arrangement of prospectively acquired data across several trials." He agreed, however, that the cardiovascular effects of the coxibs needed to be further elucidated.

Last week, events moved on, when Merck was forced to alter the US label for Vioxx, stating that caution should be exercised when it is used in patients with a medical history of ischemic heart disease, including angina or MI. The move came as a new crop of COX-2 inhibitors hit the market, amidst growing concerns that any risk of thrombotic events may be a class effect.

Balance between the two COX products "maintains cardiovascular homeostasis"

In an attempt to mimic in vivo in mice what happens when COX-2 inhibitors are administered to humans, Fitzgerald and his colleagues inflicted injury on the carotid artery with a catheter in animals bred to lack receptors for PGI2. They showed that both the vascular response and the corresponding activation of platelets were both markedly exaggerated. Similarly, mice that overexpressed receptors for TxA2 also produced these exaggerated effects. The effect was diminished when researchers either deleted the TxA2 receptor or blocked the receptor using an experimental drug. When the researchers deleted both the TxA2 receptor and the PGI2 receptor, it cancelled out the effect of inactivating the PGI2 receptor alone.

In an accompanying "Perspective," Sir John Vane (William Harvey Research Institute, St Bart's Hospital Medical College, London, UK) says: "These results provide clear evidence that PGI2 modulates the cardiovascular actions of TxA2 in vivo. They further reinforce the notion that the balance between these two eicosanoids maintains cardiovascular homeostasis."

Fitzgerald told heartwire that the research has two implications: first, if this effect turns out to be true in humans, it would be a class effect; and second, "it would only be relevant to people at greater risk of thrombosis." This is because knockout PGI2 mice do not naturally develop thrombosis, he explained, but "if you induce thrombosis in these animals by other means, the lack of this receptor renders them more susceptible to thrombosis." Also, in humans, COX-2 inhibitors only depress PGI2 by "about 70-80%, whereas in the knockout mice models, this figure was obviously 100%."

When asked what type of patient may be at such a greater risk of thrombosis, Fitzgerald replied, "That's a good question. Defining and identifying people at increased risk of thrombosis by this mechanism needs further research." He added, however, that he believes the vast majority of people with osteoarthritis who use COX-2 inhibitors are not at increased risk of thrombotic events.

Parecoxib, etoricoxib data may hold key to class effect question

In his "Perspective," Vane asks, "Was it an imbalance between PGI2 and TxA2 that led to the increased thrombotic events associated with rofecoxib in the VIGOR trial?" Noting that there is no published evidence for a cardiovascular hazard associated with COX-2 inhibitors other than rofecoxib, he suggested to heartwire that the explanation could lie in the selectivity of the different agents. "Rofecoxib is by far the most selective COX-2 inhibitor, with 70-100 times more selectivity for COX-2 over COX-1 compared with about 10-20 times the selectivity in the case of celecoxib."

 

We need these data to get a sense of what's going on. At the moment, all we are left with is speculation.

 

Fitzgerald says that, for this reason, everyone is keen to see data published on IV parecoxib (Dynastat® - Pharmacia/Pfizer), which is a prodrug to the oral valdecoxib (Bextra® - Pharmacia/Pfizer), "because these drugs are similar in selectivity to rofecoxib," and also data on etoricoxib (Arcoxia® - Merck). All there is so far are data on valdecoxib in low-risk patients, which would not be expected to show any thrombotic effects, he said. "We need these data to get a sense of what's going on. At the moment, all we are left with is speculation."

Fitzgerald adds that VIGOR remains the only study to show this thrombotic risk associated with a COX-2 inhbitor, "and the door is still open as to whether this was a real effect." Both he and Vane reiterate the fact that the VIGOR results could have been due to chance, or aspects of the trial design. Fitzgerald says it might also still prove to be true that naproxen is cardioprotective, but Vane says there is no evidence to support this assertion.

In conclusion, Fitzgerald says: "We should not allow the perception to get too far ahead of the evidence. All we have so far is a plausible mechanism, and proof of principle that this might be relevant in vivo. I don't think what we have is a reason per se, to say that taking coxibs is a cardiovascular risk factor." However, he added that in patients who have had a prior myocardial infarction or stroke, the best therapy would be aspirin "unless there is a reason not to take it."

Fitzgerald reiterated calls he has made previously for studies into coxibs used in combination with aspirin. "We don't know if the aspirin would erode the GI benefit of the coxib, or prevent any potential cardiovascular hazard." He adds that despite initial hesitancy by the companies manufacturing coxibs, he believes the message that coxib/aspirin studies are needed "has finally gotten across." However, he said he did not know how close companies are to actually beginning such studies.



Related links

1. [HeartWire > News; Apr 12, 2002]

2. [HeartWire > News; Jan 11, 2002]

3. [HeartWire > News; Nov 7, 2001]

4. [HeartWire > MediaPulse; Oct 9, 2001]

5. [HeartWire > News; Sep 27, 2001]

6. [HeartWire > MediaPulse; Aug 22, 2001]

7. [HeartWire > News; Aug 21, 2001]

8. [Satellite education > CyberSessions (non-CME); Aug 21, 2001]


Comments

3090D553-9492-4563-8681-AD288FA52ACE
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.
Post as:

processing....