ARTICLE IN BRIEF
By a unanimous vote, the 17 members of an FDA advisory committee voted in mid-October against an expanded indication for rasagiline, saying it found no compelling evidence that the drug delays progression of Parkinson disease.
The first serious effort to establish that a drug for any neurodegenerative disorder can modify the course of that disease appears to have failed.
By a unanimous vote, the 17 members of an FDA advisory committee voted in mid-October against an expanded indication for rasagiline, saying it found no compelling evidence that the drug delays progression of Parkinson disease (PD).
But in comments to Neurology Today, two of the neurologists on the committee, as well as four other physicians familiar with its decision, expressed varying views on whether the problem was with the evidence or with the drug itself.
“This would have been the first drug approved as ‘disease modifying’ for any condition in neurology,” said the acting chair of the committee, Nathan B. Fountain, MD, professor of neurology at the University of Virginia in Charlottesville. “We certainly don't want to miss a drug that has even a small disease-modifying benefit, but most members of the committee thought the data presented just didn't have sufficient power to demonstrate a difference if one exists.”
He added, however, “I've been on several FDA advisory committees, and the data presented here were by far the most complex.”
The complexity of the data on rasagiline, sold under the trade name Azilect by Teva Pharmaceuticals, could be seen in the 172-page package of studies and analysis presented to the Peripheral and Central Nervous System Drugs Advisory Committee in advance of its Oct. 17 meeting.
THE ADAGIO DATA
At the center of the debate was the ADAGIO (Attenuation of Disease Progression with Azilect Given Once-Daily) trial data. Published in the New England Journal of Medicine in 2009, the double blind 18-month trial involved 1,176 people with untreated Parkinson disease.
Rather than employ a straightforward placebo vs. active-agent design, the trial pioneered one meant to disentangle a merely symptomatic benefit from one that actually slows the progression of the disease over time. In an attempt to discern such a benefit, ADAGIO used a so-called “delayed start” design, in which half of the participants start the drug from day one and the other half are put on placebo. Halfway through the trial, however, the placebo group is switched to the active agent. If the drug is truly disease modifying, the theory goes, those who started on placebo should never catch up to those who started on the drug once both are put on active therapy; the benefits of beginning the drug earlier should persist.
While a significant benefit was seen for those who started early on the 1 mg dose, none was seen for those who started early on the 2 mg dose, for reasons that have never been adequately explained. That discordance is what led FDA statisticians to conclude, in analyses submitted to the committee, that the case had not been made for rasagiline being disease modifying. Some advisory committee members agreed with that statistical assessment.
An earlier smaller trial, TEMPO (The rasagiline in Early Monotherapy for Parkinson's disease Outpatients) was not designed to show a disease-modifying benefit, but offered hints that one might be present, which led researchers to mount ADAGIO.
One member of the FDA advisory committee told Neurology Today that the drug might in fact be disease modifying, but that the evidence presented simply wasn't strong enough to convince him or other members.
“I do prescribe this medication a lot because it's safe, it's well tolerated, and it does improve symptoms,” said Samuel Frank, MD, associate professor of neurology at Boston University. “But I don't think the evidence presented at the meeting confirmed that it's disease modifying. There was a mixed signal that it may slow the progression. We just don't know for sure.”
Although not on the committee, a neurologist who specializes in the treatment of PD said its vote was not surprising. “There's really little or no evidence that rasagiline is disease modifying or neuroprotective,” said William J. Weiner, MD, professor and chair of neurology at the University of Maryland School of Medicine and director of the Maryland Parkinson's Disease and Movement Disorders Center. Dr. Weiner is also a member of the Neurology Today editorial advisory board.
While agreeing that the evidence was not strong enough to warrant an expanded indication, the principal investigator of ADAGIO expressed disappointment at the committee's unanimous negative vote.
“I think it's more likely than not that rasagiline does have disease-modifying effects and slows clinical progression,” said C. Warren Olanow, MD, professor and chairman emeritus of neurology at the Mount Sinai School of Medicine in New York City. “Given that the 2 mg dose didn't work, the panel had no other choice but to vote no. But from my point of view, I still think that rasagiline is the closest we have come to a disease-modifying drug to date.”
Dr. Olanow added: “I'm worried this decision will deter research and investment into finding a disease-modifying treatment for Parkinson's. I think the statisticians at the FDA are asking for a perfect study, and I don't think it's going to be possible to deliver a perfect study. This was the most rigorous study we've ever had in this area, by far. Who is going to be willing to do another trial like this?”
A former director of the FDA Division of Neuro-Pharmacological Drug Products agreed that ADAGIO came close to proving the case for rasagiline being disease modifying — just not close enough.
“Just because a jury doesn't feel there's enough evidence to convict a criminal doesn't mean he didn't do it,” said Paul Leber, MD, who first proposed the “delayed start” design in a 1996 paper, and who now serves as an industry consultant based in Rockville, MD. “The 1 mg evidence was pretty good. The problem was the 2 mg result just didn't jibe with it. The jury didn't have much choice.”
Despite the negative vote, the director of the Parkinson's Disease and Movement Disorders Program at the Henry Ford Hospital in West Bloomfield, Michigan, expressed confidence that other ongoing efforts will find a treatment that significantly delays the progression of the disease.
“A good biomarker, as a surrogate of the disease process, might offer a way to track disease progress better than clinical indices,” said Peter LeWitt, MD. But, he added, “In all likelihood, an effective neuroprotective treatment would be obvious in a typical clinical trial regardless of availability of a good biomarker.”
On the question of a Parkinson biomarker, Dr. Olanow commented, “A biomarker would be great, but we don't have one and it may be decades before we do. Indeed, it may be easier to prove neuroprotection than to prove that something is a valid biomarker, which leaves us back in the same conundrum.”
To rise out of the statistical haze that has enveloped rasagiline, Dr. Fountain said, “The ideal solution would be to have a very potent drug with a quick onset of action. The problem is we don't have any drugs for Parkinson's where we see a huge effect quickly. Even if you had only a mild or moderate effect on disease progression, you might still be able to support that with validated biomarkers.”
Dr. Weiner offered another possible route toward finding disease-modifying drugs. “I really think that Parkinson's is not a single disease; it's a clinical syndrome,” Dr. Weiner said. “I think there will turn out to be many different causes for development of the phenotype of Parkinson's. The future of trials is we're going to divide out the genetic variations and test drugs for specific subgroups.”