Over the past few years, a debate has intensified over the extent to which drugmakers should disclose all clinical trial data pertaining to their medicines. The issue has drawn bitter complaints from some physicians and consumer watchdogs, among others, who argue anything less than complete disclosure may deprive the medical community of important info needed for treating patients.
Consider the case of Mirapex. The drug, which is sold by Boehringer Ingelheim, was initially approved in 1997 to treat Parkinson’s disease and, six years ago, the approval was extended to combat Restless Legs Syndrome. At that time, cardiac failure was listed in the labeling among dozens of adverse events that were noticed during clinical trials but for which causation was never determined (look here).
Since then, cardiac failure has been cited in connection with the drug. A study that was funded by Pfizer, which was comparing its own Dostinex against other medicines, found a statistically significant increased risk of heart failure with Mirapex. The results were presented at the European Society of Cardiology meeting in Sweden two years ago (see this) and published online last November in Pharmacological Research (read here and here).
Yet another study appeared a few months ago in Pharmacoepidemiology & Drug Safety. Supported, in part, by Boehringer Ingelheim, the study reached the same conclusion about both Mirapex and the Pfizer drug. But the text yielded some revealing passages that suggested Boeheringer Ingelheim had been tracking cardiac failure for some time without any public disclosure (look here and here).
The introduction to the study, which was entitled ‘Dopamine Agonist Use and The Risk of Heart Failure,’ noted that, “recently, adverse events of heart failure have been observed in association with the dopamine agonist pramipexole (the chemical name of Mirapex) in randomized trials.” The authors cited Boehringer “internal data” and a “personal communication from Dr. Bartels.”
The reference was to Dorothee Bartels, who is corporate vice president and head of global epidemiology at the drugmaker. She provided the results of a pooled analysis from randomized Mirapex trials. Specifically, the introduction mentions a “signal of a potential risk of heart failure arose in pooled data from 26 placebo-controlled, randomized trials” in Parkinson’s and RLS patients.
Those trials involved 4,157 patients who were taking Mirapex and another 2,280 on a placebo, but the outcome was too small to draw reliable conclusions, the authors wrote, and raised questions about the potential for the risk to occur with other dopamine agonists. This led them to conduct the study. Yet those 26 trials were never made available. It is also not clear how far back they go in time.
We asked the corresponding author, Samy Suissa, a professor of epidemiology, biostatistics and medicine at McGill University, if Boehringer provided underlying data. He wrote us that “the only data I have are in the paper – rates, numbers, etc.” We should note that Suissa received research grants and honorarium from Boehringer over the last three years.
There was, however, one curious disclaimer at the bottom of the study. The authors maintained they had ultimate control over “all aspects” of the research, including control over publications. “During the study, however, any differences about the presentation or interpretation of findings that arose between the authors and the company were resolved through honest scientific debate,” they added.
We asked a Boehringer spokeswoman when the drugmaker learned of the risk of cardiac failure in the pooled trials, whether a report about the findings was compiled and, if so, would a copy be made available. But those questions went unanswered during two separate efforts. Instead, we were sent copies of the product labeling along with a brief statement.
She wrote us that labeling for Mirapex and Mirapex ER, a newer version approved in 2010, “include language related to cardiac events, and specifically heart failure or cardiac failure. Post-marketing surveillance studies continue to assess the safety and efficacy of our products… We continue to work closely with the FDA to ensure our products are appropriately labeled and are used safely.”
The labeling, which was revised last year, provides the same language that appeared in the 2006 labeling, with the addition of one line: “In a pharmacoepidemiological study, pramipexole (Mirapex) use was associated with an increased risk of cardiac failure compared with non-use.” This appears under the post-marketing experience section. There is no mention, however, of two studies or the undisclosed data (see here and here).
In Europe, meanwhile, the 2011 labeling mentions cardiac failure as a possible effect (read this). The European Commission Guideline on Summary of Product Characteristics notes that adverse events are not added to the European Union labeling unless they at least are suspected to be causally related to the drug (see page 11 here).
The mysterious fate of the internal data troubled one expert, who has regularly advocated for greater dissemination of clinical trial results and last year suggested the creation of an “external coordinating organization,” which would be independent and have legal resources to contract with drug and device makers for providing access to patient-level data, documents and case reports (read this).
“I am disturbed that there seems to be a pooling of internal data indicating a potential problem with heart failure and there has been no opportunity for independent review,” wrote Harlan Krumholz, a cardiologist, a professor of medicine, epidemiology and public health at the Yale University School of Medicine, and a board of governor of the Patient-Centered Outcomes Research Institute.
He reviewed the studies and related documentation at our request. “This is an ideal situation where sharing of individual patient-level data with the entire research community, or at least with independent investigators, can provide information about the risk that was conveyed in the company’s personal communication to their consultant,” he continued.
“Such sharing would deflect concerns that the company is hiding something and allow a fair evaluation of the risk. With no record in the medical literature and no sharing of the data, we are left to wonder about these risks and the reason that the company has not been more forthcoming. I applaud them for doing analyses on harm, but they need to go further.
“The privilege of selling a product should be accompanied by the responsibility to share data you have that is germane to the risks and benefits of the drug. Patients deserve access to information that could influence their decisions about clinical strategies. If it is true that the company has data from their trials that is relevant to concerns about the drug’s safety profile and that data has not been shared, then I hope they will move quickly to make it available.”
At around the same time that Boehringer was accumulating the pooled data, by the way, the drugmaker was tagged by the FDA for improperly promoting Mirapex. A warning letter from 2008 noted the drugmaker prepared promotional materials that omitted risk information and an untitled letter in 2009 pointed to sponsored links on Google that also failed to contain risk information.