New Analysis Extends Dabigatran MI Signal to Other Thrombin Inhibitors

March 30, 2012

March 30, 2012 (Chicago, Illinois) — A new meta-analysis of trials of oral direct thrombin inhibitors--which included two studies of the new anticoagulant dabigatran (Pradaxa, Boehringer Ingelheim) and three studies of an older, discontinued drug, ximelagatran (Exanta, AstraZeneca)--shows an increased rate of MI in those receiving these agents as compared with warfarin.

This effect was statistically significant, "suggesting a class effect," said Dr Ramin Artang (University of Nebraska Medical Center, Omaha), who reported the findings earlier this week at the American College of Cardiology (ACC) 2012 Scientific Sessions. "Clinicians involved in the care of patients with coronary artery disease may need to exercise caution in the use of these agents," he noted. And as this is believed to be one of the first analyses to extend the finding of an increased risk of MI with dabigatran to other thrombin inhibitors, it has implications for newer agents further back in development, Artang notes.

"I am very happy that after 50 to 60 years of warfarin, we have alternatives, and I have very good experience with dabigatran so far. But the trend [for MI] is there, and in real life we have thousands and thousands of patients, some of whom have coronary disease, whom we are putting on dabigatran and perhaps in the future other thrombin inhibitors. That is a concern. It's important to keep an eye out. I therefore believe we need to have registries on dabigatran patients so we can report this easily, and MI should be made a secondary end point for new thrombin inhibitors in development, so we don't need to dig into articles to find it," he told heartwire .

The trend [for MI] is there, and in real life we have thousands and thousands of patients, some of whom have CAD, whom we are putting on dabigatran and perhaps in the future other thrombin inhibitors.

Most experts polled by heartwire following the presentation continued to stress that any effect of dabigatran on MI is "very small"; it would not deter them from using it and should be seen in the overall context of its benefits and risks. This stance is consistent with recent new AF recommendations from the Canadian Cardiovascular Association and antithrombotic guidelines from the ACCP, which have made the jump beyond previous advice to indicate that new oral anticoagulants--including dabigatran--be "preferred" to warfarin, as opposed to just being "alternatives."

Any MI Signal "Very Small"; No Need to Change Recommendations

Heart Rhythm Society president Dr Anne M Gillis (University of Calgary, AB), who was one of the cochairs of the session at which Artang presented his results, said his new meta-analysis would not cause her to change practice. The other cochair, Dr Julia H Indik (University of Arizona College of Medicine, Tucson), agreed, calling Artang's presentation "thought provoking; we'll learn more over time, but I would agree with Dr Gillis. At this point, the risk/benefit ratio still favors using dabigatran, and we just need to continue to monitor."

Dr Harry R Büller (Academic Medical Center, Amsterdam, the Netherlands)--a venous thromboembolism (VTE) expert who uses the new oral anticoagulants for thrombosis indications--is similarly inclined. "We have to keep in mind the magnitude observed. Under the assumption that there is a small [MI] signal, it should not be blown out of proportion, because it actually is very, very small, and you should see that in the total amount of advantages. For example, if you look at the intracranial hemorrhages numerically, that's a much larger problem," he told heartwire .

Likewise, in another session at ACC--entitled "New anticoagulants in AF, what to do in clinical practice"--Dr John Camm (St Georges Hospital, London, UK) said: "It is now clear that there is a small but definite signal of MI with dabigatran vs warfarin, but this is far outweighed by its benefits." Nevertheless, Camm acknowledged, "This has made some clinicians wary."

Signal Is There With the "Older Cousin," Too, Ximelagatran

Artang and colleagues performed an analysis of all trials to date that they could identify in which an oral direct thrombin inhibitor was compared with warfarin, for any indication. As well as the two dabigatran trials--RE-LY for prevention of stroke in AF and RECOVER for the treatment of acute deep vein thrombosis (DVT)--they also included three trials of ximelagatran, which Artang described as "an older cousin of dabigatran."

Ximelagatran was briefly marketed in some European countries for prevention of VTE in orthopedic surgery in the early 2000s, but was withdrawn worldwide in 2006 when it was found to cause liver toxicity. Artang included in the meta-analysis THRIVE--a DVT/pulmonary-embolism (PE) trial with ximelagatran--and SPORTIF III and V, AF trials with the agent. In total, there were just over 30 000 patients in the whole meta-analysis, he noted.

They found an MI signal for thrombin inhibitors overall, with an HR of 1.27, "which is highly significant," compared with warfarin, he said. But he also acknowledged that if the RE-LY data are taken out of the meta-analysis, the signal for MI is no longer significant. "This is a limitation," he admitted. "We cannot get away from it--whichever statistical maneuver we use--RE-LY represents 57% of the cohort and 7% of the effects."

One audience member also pointed out that the only other trial in the meta-analysis that flags a significant increased risk of MI is THRIVE with ximelagatran but that this trial "only had 2000 patients in it." Artang said while this was true, the ratio of ximelagatran to warfarin given in that trial was 10:1, giving it more weight than the numbers would suggest.

"The idea is not to ignore the fact that you have 30 000 patients and two-thirds of them received a direct thrombin inhibitor and a third received warfarin, and you have a significant [excess] amount of MIs" in the former arm. "The trend is there. Let's keep an eye open and see," he noted.

Is the Explanation a Protective Effect of Warfarin? Probably Not

The same questioner also criticized the meta-analysis because it "mixed and matched" studies in different indications--ie, PE/DVT and AF. And "there has been a lot of speculation about the cardioprotective effects of warfarin, which skews your findings in favor of warfarin and against direct thrombin inhibitors."

Artang replied that there "is no doubt that warfarin has a protective effect [in terms of] MI," but that does not explain the whole issue with thrombin inhibitors; otherwise, MI signals would have been seen in other trials of new anticoagulants that are not thrombin inhibitors, such as the factor Xa inhibitors.

With the thrombin inhibitors you have a trend of more MI, but you don't see that with the factor Xa inhibitors, so it's not because warfarin is better; it's some other explanation.

Others agreed. In discussion with Artang at the end of the session, hematologist Dr Jørn Dalggard Nielsen (Rigshospitalet, Copenhagen, Denmark), commented: "With the thrombin inhibitors you have a trend of more MI, but you don't see that with the factor Xa inhibitors, so it's not because warfarin is better; it's some other explanation." Büller, the VTE expert, concurs. "There are theoretical, pharmacologic, and pharmacodynamic reasons" relating to where in the coagulation cascade the different agents act that might explain the differential effects, he told heartwire .

Artang was also challenged by someone who said they worked for Boehringer Ingelheim, the company that markets dabigatran, who stressed that the reanalysis of RE-LY--performed after the FDA expressed concern about MIs in the trial--found no increased risk of MI with dabigatran, so "what you say is simply not valid."

But Artang stood his ground: "I didn't do a meta-analysis on dabigatran alone--it was the whole group." He had, he said, also clearly stated that the MI signal with dabigatran in RE-LY was no longer statistically significant after the reanalysis, "but the meta-analysis with other thrombin inhibitors still falls out, regardless of what you do, on the statistically significant side."

Artang has no disclosures. Büller has consulted for, received research support from, and is on scientific advisory boards for Sanofi, Bayer HealthCare, Bristol-Myers Squibb, Daiichi-Sankyo, GlaxoSmithKline, Pfizer, Roche, Isis, and Thrombogenics. Camm has served as an advisor or consultant for Actelion Pharmaceuticals, ARYx Therapeutics, Bristol-Myers Squibb, Cardiome Pharma, CV Therapeutics, Daiichi Sankyo, Menarini Group, Merck, Novartis, Sanofi, Servier, and Xention; served as a speaker or a member of a speaker's bureau for Cardiome Pharma, Daiichi Sankyo, Menarini Group, Pfizer, and Sanofi; received grants for clinical research from Bristol-Myers Squibb, Daiichi Sankyo, Sanofi, and Server; served as a member of the data safety monitoring board for Bristol-Myers Squibb, Novartis, and Servier; and served as an expert witness for Johnson & Johnson, Sanofi, and Servier.


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.