By Blake N. Shultz and Gregory Curfman
Despite a troubling history, rofecoxib (Vioxx) may be making a comeback.
The voluntary withdrawal of rofecoxib (Vioxx) from the market in September 2004 marked the end of a controversial era for a once highly profitable and widely used drug. It also marked the beginning of years of high-profile product-liability litigation that would cost Merck billions.
Rofecoxib was approved by the U.S. Food and Drug Administration (FDA) in 1999 for the treatment of acute, osteoarthritis-induced pain and was subsequently also approved for the treatment of pain associated with rheumatoid arthritis and migraine. Despite this broad indication, post-approval studies showed that the drug was associated with an increased risk of a variety of cardiovascular and thrombotic complications, most significantly myocardial infarction. Adding to the controversy, documents obtained in the subsequent litigation showed that Merck engaged in misrepresentation of study results that downplayed the risks of rofecoxib.
But over a decade later, in November 2017, TRM-201 (rofecoxib) received orphan drug designation for the treatment of pain caused by hemophilic arthropathy. Preliminary, observational data suggested that rofecoxib may be beneficial in this debilitating clinical condition, which is caused by spontaneous bleeding into joints, with repeated hemarthroses leading to severe, disabling, chronic inflammatory arthritis.
To explore the potential risks and benefits, researchers are in the planning stages of a randomized clinical trial of rofecoxib in this condition. The proposal to bring rofecoxib back to market was met with skepticism from some members of the academic community. Broadly, concerns revolved around the potential risks of off-label use and the need for clinical trial transparency about the cardiovascular risk of rofecoxib.
While the cardiovascular risk profile of rofecoxib should not be understated, these criticisms should be placed in context. The risk of death from ischemic heart disease in patients with hemophilia has been reported to be lower than in a control population. Furthermore, the safety profile of rofecoxib has not been rigorously assessed in patients with hemophilic arthropathy, and the application of risk evaluation and mitigation strategies (REMS) also needs to be considered. For example, limiting prescriptions to patients with low cardiac risk and regular cardiovascular screening would be expected to mitigate cardiovascular risk. Because side effects increase with duration of use, drug cycling or holidays may be used to reduce risk (similar to those used to reduce the significant side effects of long-term levodopa therapy for Parkinson’s disease).
Although studies have shown that the cardiovascular risk profile of rofecoxib may be worse than that of other NSAIDs, the magnitude of this risk is modest and the above strategies may provide mitigation. For example, one large cohort study found that the odds ratios for acute myocardial infarction were 1.48 (95% CI, 1.00-2.26) for ibuprofen and 1.58 (95% CI, 1.07-2.17) for rofecoxib. The modest difference in the point estimates of risk with the potential for additional mitigation strategies suggests that sufficient clinical equipoise exists to justify a clinical trial. Development and appropriate prescribing of this medication are worthy of further study.
The need for novel pain treatments for patients with hemophilic arthropathy is substantial. Current pharmaceutical management of hemophilic arthropathy-related pain relies on long-term opioid use and intra-articular corticosteroid injections, both of which come with significant risks. In addition to the risk of addiction, studies have shown that an opioid supply of at least 180 days over a 3.5-year period was significantly associated with increased risk of myocardial infarction versus no therapy (adjusted incidence rate ratio, 2.66 [95% CI, 2.30 to 3.08]), and that this risk persisted for any duration of therapy. The risk of myocardial infarction from long-term opioid therapy is comparable to that observed with rofecoxib, supporting the argument that sufficient equipoise exists to justify conducting a clinical trial.
Intra-articular corticosteroid injections may also be associated with potentially serious complications, including bone loss and joint destruction, which led Kompel et al. to recently suggest that intra-articular corticosteroid injections are “perhaps not as safe as we thought.”
Concerns about off-label prescribing if rofecoxib were re-introduced into the market similarly may be over-stated. Some critics suggest that because drugs that treat common conditions, such as osteoarthritis pain in the case of rofecoxib, have high rates of off-label use, the risk of market reentry is particularly pronounced. They also point to examples of nonadherence with REMS commonly utilized by the FDA to control the use of drugs with known safety risks. However, as a counterexample, consider the REMS for thalidomide. Initially used to treat anxiety, insomnia, and morning sickness, thalidomide was found to be a powerful teratogen after it caused thousands of fetal deaths and deformities across the world. Today, however, it is used in the United States under a REMS as a first-line therapy for multiple myeloma. Recent studies suggest that complete compliance with birth control requirements under the REMS program was 96.3%, and that compliance remained very high throughout a full course of therapy.
Prior authorization for insurance coverage and other formulary management strategies have also been cited as methods to control off-label use of rofecoxib, although skepticism has been raised due to the ability of companies to circumvent these barriers. However, the threat of civil enforcement and malpractice liability are potent deterrents to unsafe prescribing practices.
Without further study of the safety of rofecoxib in the hemophilic arthropathy population, and without rigorous evaluation of REMS and other mitigation strategies for rofecoxib, the potential benefits of the drug’s return to market should not be readily dismissed. While concerns related to cardiovascular risks are serious and cannot be overlooked, the clinical problem of hemophilic arthropathy is severely debilitating for patients and current therapies are not only inadequate but also pose substantial risks of their own. It is entirely plausible that rofecoxib may provide more benefit with less risk than current therapies in patients with hemophilic arthropathy. We believe that the principle of clinical equipoise is clearly met and that a clinical trial of rofecoxib in patients with hemophilic arthropathy is appropriate and should be undertaken promptly to address this significant unmet medical need.
