BETHESDA, MD 27 Jun 2014—FDA does not need the results of a randomized, controlled cardiovascular outcomes trial before deciding whether to approve methylnaltrexone bromide or naloxegol oxalate for the treatment of opioid-induced constipation in adults with chronic noncancer pain; instead, well-designed postapproval observational studies of adequate duration would suffice. That’s the view of two thirds of the 24 advisers who met June 11–12 in Silver Spring, Maryland, at the behest of the regulatory agency.
The agency sought help from its Anesthetic and Analgesic Drug Products Advisory Committee in navigating a controversy that began eight years ago with the emergence of a cardiovascular safety signal during clinical testing of the related agent alvimopan—a signal that might have been a fluke.
In summarizing the committee’s discussion on the strength of the eight-year-old almivopan safety signal after a review of the trial data, Chairperson Randall P. Flick, with the Mayo Clinic in Rochester, Minnesota, reported the group as “mixed in its view” almost equally.
At one end of the spectrum was temporary committee member Peter W. F. Wilson, with the Emory Clinical Cardiovascular Research Institute in Atlanta: “I think this is a strong signal.”
At the other end of the spectrum was temporary committee member Milton Packer, Stoffel Distinguished Chair in Cardiology at the University of Texas Southwestern Medical Center in Dallas: “I don’t think there’s a signal.”
At the root of the current quandary surrounding methylnaltrexone bromide, naloxegol oxalate, and related agents are the seven myocardial infarctions reported in 2006 among patients with chronic noncancer pain who were participating in a 12-month safety study of alvimopan, a peripherally active µ-opioid–receptor antagonist, for the treatment of opioid-related bowel dysfunction. No myocardial infarction occurred in any of the patients assigned to the placebo group.
Potential risk. Temporary committee member John R. Teerlink, director of the heart failure program at the San Francisco VA Medical Center, called alvimopan “the canary in the coal mine”—an indicator of cardiovascular risk as a class effect of peripherally active µ-opioid antagonists. Yet he acknowledged that the safety signal reported with alvimopan was “relatively weak” in strength.
Alvimopan came to market in 2008 for a short-term use: accelerating the time to gastrointestinal recovery after partial bowel resection surgery in adults.
To ensure that only those patients receive alvimopan, FDA said, a risk evaluation and mitigation strategy (REMS) was necessary.
One of the formal goals of that REMS is to “mitigate the potential risk of myocardial infarction” by ensuring that alvimopan therapy involves no more than 15 doses and occurs only in a hospital inpatient setting.
At the time of alvimopan’s approval, FDA stated that a causal relationship between alvimopan and myocardial infarction had not been established.
To this day, the agency still describes the finding from the long-term safety study of low-dose alvimopan as a numeric “imbalance” in cardiovascular events between the drug and placebo groups.
The imbalance again became an issue in 2012 as FDA’s Division of Gastroenterology and Inborn Errors Products reviewed Salix Pharmaceutical Inc.’s application to market methylnaltrexone bromide for the treatment of opioid-induced constipation in adults with chronic noncancer pain.
Like alvimopan after oral administration, methylnaltrexone interferes with the binding of opioids to µ-opioid receptors in the gastrointestinal tract.
Salix reported in financial statements that the division “expressed a concern that there might be a risk associated with the chronic use of [µ]-opioid antagonists in patients that are taking opioids for chronic pain.”
Further, to understand this potential risk, “the Division has communicated that a very large, well-controlled, chronic administration trial will have to be conducted to assess the safety of any [µ]-opioid antagonist prior to market approval for the treatment of patients with OIC [opioid-induced constipation] who are taking opioids for chronic, non-cancer pain.”
The company stated it filed an appeal with FDA, which decided to consult its outside advisers regarding the need for the trial.
Potential obstacles to an answer. A cardiovascular outcomes trial of naloxegol oxalate, another peripherally active µ-opioid–receptor antagonist, would require 10,000–20,000 enrollees and take 7–10 years to complete, AstraZeneca’s William Mezzanotte told the committee.
Mezzanotte, vice president for global medicines development, said that during the four trials comprising the Phase III program for naloxegol, a total of 2100 patients at 440 sites in 12 countries were randomly assigned to receive the drug or an alternative.
The “screen failure rate” was 65%, he reported, implying that investigators screened 6000 patients for enrollment.
Results from the two identical 12-week double-blind, randomized, placebo-controlled trials in the Phase III program were released online on June 4 and in print on June 19 by The New England Journal of Medicine.
According to the article, 18–26% of the patients assigned to the naloxegol and control groups did not finish the studies.
Company consultant William B. White, a hypertension specialist at the Pat and Jim Calhoun Cardiology Center, part of the University of Connecticut Health Center in Farmington, reported that the distribution of treatment-related adverse cardiovascular events was balanced among the studies’ groups.
He also reported that adverse cardiovascular events occurred more commonly in the patients at higher risk for such events due to preexisting cardiovascular disease or diabetes mellitus or to multiple risk factors than in the patients with a single risk factor.
AstraZeneca, Mezzanotte said, has proposed conducting a postapproval observational study instead of a preapproval cardiovascular outcomes trial.
The proposed observational study would use electronic health records from multiple validated sources, he said.
Temporary committee member Wilson expressed concern in general as to whether patients with opioid-induced constipation who are assigned to the placebo group in a cardiovascular outcomes trial would continue beyond the first several weeks. After all, opioid-induced constipation is highly symptomatic, he said, and patients receiving the placebo would still be constipated and feel it.
FDA’s Robert J. Temple, deputy director for clinical science in the Center for Drug Evaluation and Research, acknowledged that “people do drop out of symptomatic trials.”
The committee, said Chairperson Flick, in general believed that for peripherally active µ-opioid–receptor antagonists other than alvimopan, a safety-related controlled clinical trial can be conducted after, rather than before, FDA approves the drug for use in patients with opioid-induced constipation. This study should be one year in duration to enable assessment of cardiovascular outcomes associated with long-term use of the drug.