Reports

This report is based on medical evidence presented at sanctioned medical congress, from peer reviewed literature or opinion provided by a qualified healthcare practitioner. The consumption of the information contained within this report is intended for qualified Canadian healthcare practitioners only.

70th Annual Meeting of the American College of Rheumatology

Washington, DC / November 10-15, 2006

For the last two years, physicians have been justifiably concerned about the cardiovascular (CV) safety of cyclooxygenase (COX)-2 inhibitors for the treatment of chronic pain and inflammation associated with arthritic conditions such as osteoarthritis and rheumatoid arthritis (RA). A well-publicized study was presented at ACR 2004, and subsequently published in 2005, which linked the COX-2 inhibitor rofecoxib to an increased risk of CV disease. Shortly after publication, the agent was withdrawn from worldwide markets and the resultant action had a dampening effect on the use of other compounds within the class.

MEDAL Program Findings

In this newly published study, presented simultaneously at ACR 2006 in Washington, DC, and at the 2006 Annual Meeting of the American Heart Association (AHA) in Chicago, Illinois, investigators of the MEDAL (Multinational Etoricoxib and Diclofenac Arthritis Long-Term) Program found a reassuringly low rate of both CV and gastrointestinal (GI) adverse effects when etoricoxib was compared to the NSAID diclofenac. The study was also simultaneously published in The Lancet (Cannon et al. Lancet 2006;368(9549):1771-81).

In the MEDAL Program, the investigators analyzed pooled data from three trials involving 34,701 patients who had been randomly assigned to receive either etoricoxib or diclofenac. Among these patients, 24,913 had osteoarthritis and 9787 had RA. Those on the COX-2 inhibitor were further randomized to receive either 60 or 90 mg daily, while all of those on the NSAID received 150 mg daily.

The investigators were primarily evaluating thrombotic CV events in a non-inferiority designed study, which sought to determine whether the study agent, in this case etoricoxib, was equivalent or “non-inferior” to the study comparator diclofenac. The investigators defined the upper boundary as <1.30 for the hazard ratio (HR) for thrombotic CV events in the per-protocol analysis. Patients were treated for an average of 18 months.

Adverse-event Profiles

During the study period, thrombotic CV events occurred in 320 patients in the COX-2 inhibitor group and in 323 patients in the NSAID group, resulting in a rate of occurrence of 1.24 per 100 patient-years and 1.30 per 100 patient-years in the NSAID group, with an HR 0.95 for etoricoxib compared with diclofenac. Upper GI clinical events such as perforation, bleeding, obstruction and ulcer occurred at a rate of 0.67 per 100 patient-years in the COX-2 inhibitor cohort and 0.97 per 100 patient-years for those on the NSAID, for an HR of 0.69. However, the two treatment groups had similar rates of complicated upper GI events at 0.30 per 100 patient-years for etoricoxib and 0.32 for diclofenac.

“Our results showed that patients with arthritis treated with the COX-2-selective NSAID etoricoxib and those given the traditional NSAID diclofenac had nearly identical rates of thrombotic CV events,” the authors reported. They encouraged rheumatologists to use the information judiciously and to take patients’ medical and family histories into account when prescribing for arthritis. “In clinical practice, the choice of anti-inflammatory agent needs to take into consideration the risk for thrombotic CV and GI events, as well as congestive heart failure and other renovascular effects, and efficacy.” They noted that certain renovascular end points such as congestive heart failure, as well as discontinuations because of hypertension, were higher with etoricoxib than with diclofenac.

According to co-author Dr. Loren A. Laine, Professor of Medicine, Department of Gastroenterology, University of Southern California Keck School of Medicine, and Chief of Gastroenterology, Los Angeles County and University of Southern California Medical Center, “COX-2 inhibitors do not increase CV risk when compared to other traditional NSAIDs. The findings show the importance of stratifying patients by risk, so that we can treat them safely and effectively.” For example, patients with a prior GI clinical event are at high risk of subsequent events and patients with established CV disease are at risk of CV events. Taking into account patients’ risks for such events and individualizing therapy will help physicians treat their arthritis effectively and minimize the risk of adverse events, he told delegates.

Canadian Perspective

“The findings should be helpful for Canadian rheumatologists,” commented Dr. Alfred Cividino, Associate Clinical Professor of Medicine, McMaster University, Hamilton, Ontario. “This is the largest COX-2 inhibitor comparator study looking at CV end points. The findings showed no difference regarding CV safety in thrombotic CV events and a lower incidence of GI events than diclofenac.”

Results of the MEDAL Program help to address the initial concerns raised in 2004 regarding rofecoxib, he added. “Canadian physicians were concerned that these findings applied to COX-2 inhibitors across the board. Many of us went back to using traditional NSAIDs in the treatment of arthritis. Before MEDAL, we had limited data from which to gauge COX-2 inhibitors’ safety. In MEDAL, we saw 600 CV events in over 34,000 patients and etoricoxib was similar in risk to diclofenac.” Dr. Cividino stressed that it would be too soon to say whether the safety findings could be generalized to other compounds in this class or whether they were specific to etoricoxib.

Concurrent Presentation of Data at AHA

The MEDAL Program results were presented simultaneously at the AHA in Chicago. In that setting, the major concern was whether the MEDAL Program was definitive. Dr. Christopher P. Cannon, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts, indicated that the data were more than reassuring. He suggested that the large multicentre trial settled a critical issue for cardiologists, rheumatologists and others who are trying to balance risks and benefits of commonly used therapies.

“This trial addressed the fundamental question of this class: Does the increase in COX-2 selectivity lead to an increase in thrombotic CV events compared with a traditional nonsteroidal agent? And the answer is no,” Dr. Cannon stated. He emphasized that many of the features of the MEDAL Program support this conclusion. Specifically, etoricoxib is highly selective for COX-2, while diclofenac, unlike many other non-selective NSAIDs such as naproxen, has no sustained effect on platelets. Diclofenac was also a reasonable comparator because it is the most prescribed NSAID in the world.

“We also looked at other events, including arterial events and the more common events, and the more common end point of CV death and stroke, with identical findings. There were robust numbers of events, totalling nearly 430 events in the two arms and not differing on this hard end point over the three years of follow-up,” Dr. Cannon reported.

In a commentary delivered immediately after the data were presented at the AHA, Dr. Robert Califf, Duke University Medical Center, Durham, North Carolina, agreed that the study can be considered “a relatively definitive trial for the direct comparison that was done.” He suggested that there were several features of the study design that give it authority. Most importantly, there was an independent data analysis committee that provided results “with another pair of eyes.” In addition, he was impressed with the large number of CV events with which the two agents could be compared. According to Dr. Califf, the number of events is a far more important variable than the number of patients or the duration of follow-up when comparing the relative protection of two strategies. By confirming protection in a patient population proven to be at high risk from a high number of events, this study provides critical information “in understanding how to advise patients and their doctors about the choices they need to make,” he observed.

Both Drs. Cannon and Califf cautioned that many important questions about the relative CV safety of NSAIDs remain, including the relative risks and benefits of NSAIDS such as naproxen that affect platelets. However, Dr. Califf concurred with Dr. Cannon in concluding that MEDAL answered the question it was designed to address. He praised MEDAL investigators for not only demonstrating that etoricoxib provides CV safety that is at least comparable to diclofenac, but also that it objectively demonstrated similar pain relief and a lower rate of GI events.

Reference: Christopher P Cannon, Sean P Curtis, Garret A FitzGerald, Henry Krum, Amarjot Kaur, James A Bolognese, Alise S Reicin, Claire Bombardier, Michael E Weinblatt, Désirée van der Heijde, Erland Erdmann, Loren Laine, for the MEDAL Steering Committee. Cardiovascular outcomes with etoricoxib and diclofenac in patients with osteoarthritis and rheumatoid arthritis in the Multinational Etoricoxib and Diclofenac Arthritis Long-term (MEDAL) programme: a randomised comparison. Lancet 2006;368(9549):1771-81.

Note: At the time of printing, etoricoxib is not available in Canada.