Should The Fda Lower The Drug Testing Threshold?
For Dr. Patricia K. Coyle, the testimony was wrenching. Speaker after speaker described how paralyzing Lou Gehrig's disease had relentlessly robbed them or their loved ones of the ability to walk, feed themselves, or even hold a pencil. Tearfully, they pleaded in May with Coyle and five other neurologists on a Food & Drug Administration advisory committee to clear the way for approval of a new drug, Myotrophin--a genetically engineered, insulinlike growth factor that might slow the disease's advance.
But for Coyle and the other panel members, the facts were less persuasive. The drug's manufacturer, biotech startup Cephalon Inc., simply hadn't proved its case. "There was no way you could review that data and say unequivocally that this agent worked," says Coyle, a professor at the State University of New York at Stony Brook. The vote against the drug was 6-3, with only nonphysicians dissenting. By Aug. 11, the FDA is expected to decide whether to approve the drug--with the panel's thumbs-down weighing heavily.
Whether the panel acted correctly is a matter of heated dispute. Myotrophin is a case study in how well-meaning regulators and companies with little capital can err when dealing with diseases for which there is little or no treatment. Startups such as Cephalon may be aiming too low in designing studies that fail to provide enough evidence to justify approval for such drugs. However, critics charge that regulators set too high a bar for drugs targeted at poorly understood diseases by applying the same tough standards used for routine medications, such as preferring two studies showing the same results.
Lou Gehrig's sufferers are caught in the middle. "I want the opportunity to see if it would work for me," says Shelbie M. Oppenheimer, a 30-year-old former day-care center director from New Hope, Pa., who was diagnosed with Lou Gehrig's disease in 1996. "I'm not an idiot. If it doesn't work, I'm not going to keep taking it."
Precedent may be on her side. The FDA has broken with tradition to approve some AIDS medicines on the basis of one trial. They've gone on to become life savers. And the agency eventually approved Warner-Lambert's Cognex medicine for Alzheimer's sufferers, even though three studies showed the drug offered only marginal, if any, improvements from some of the fatal disease's symptoms.
Five years ago, Myotrophin's future seemed bright. Researchers at Cephalon, based in West Chester, Pa., began research in 1992 to see if the drug, which promotes nerve growth, could stymie a disease that afflicts up to 30,000 Americans. Scientifically dubbed amyotrophic lateral sclerosis (ALS), its cause is mostly unknown--though its consequences are clear: irreversible wasting of muscles, usually leading to death from respiratory failure in less than five years.
After initial tests determined the drug was safe, Cephalon had to prove that it worked. But that was trickier. Soon after researchers launched their studies--on 266 patients in North America in early 1993, and on 183 Europeans later that year--they found that any improvement was modest. Some had better muscle tone or more ease in swallowing, for instance, and a measurable slowing of some patients' decline was reported. Still, in mid-1995, based on the North American results, Cephalon claimed "highly statistically significant effects" with patients showing 26% "less deterioration" than those on a placebo. Chief Executive and company founder Frank Baldino Jr. heralded "good news for a lot of people."
Soon came the bad news. In October, 1995, Cephalon reported that in the European trial, Myotrophin fell short of treatment goals. More disturbing, the European death rate was notably higher on Myotrophin than on the placebo--14.5% vs. 8.5%. "I wouldn't even conclude that it is a marginal drug," says Dr. Sid Gilman, head of the FDA advisory committee. "I suspect that it is ineffective."
LET'S MAKE A DEAL. Why the difference between the two studies? Cephalon executives say the European patients were sicker than American counterparts. But the FDA could not overlook the European results, particularly since the agency generally wants to see two studies before it will approve a new drug. With one study showing some effectiveness, however, the regulators wanted to keep the drug available as tests continued. So in mid-1996, the advisory committee ruled that a limited number of patients outside the formal studies could receive the drug until a final decision is made. However, the committee insisted that Cephalon provide convincing new data.
By last February, though, Cephalon decided it had done enough. The company and its partner, Chiron Corp., filed with the FDA for approval to sell Myotrophin. Baldino told regulators that Cephalon had spent $180 million on research and would have a tough time justifying further investment. "It's almost impossible for us to idly throw $20 million or $30 million at another study," Baldino told regulators.
Instead, Cephalon offered a deal. If regulators let it start selling the drug--some analysts say Myotrophin could fetch up to $10,000 a year per patient--Cephalon would do further research. It proposed combining Myotrophin with the only approved ALS medicine, Rilutek, so that doctors could see the drug's usefulness in the field. Rilutek, a Rhone-Poulenc Rorer drug approved in December, 1995, extends life for ALS sufferers by an average of three months but offers no improvement in physical functions. Backers theorized a combination of the two drugs could do both modestly.
But the advisory panel had a narrow charge--to determine whether the drug showed "substantial" effectiveness. Without two convincing trials, it hadn't. What's more, Cephalon's approach irked committee members. "I thought their presentation was a bit on the sloppy side," says Coyle. One committee member charges that "Rather than doing the science...the company wanted to call in the most pitiful cases and use the emotional pressure of people with a very bad disease." Baldino insists the company already had presented sufficient data and says the patients testified without company encouragement.
Now comes the FDA. The agency could overrule its advisers and approve the drug, a rare move. It could grant conditional approval, reserving the right to order withdrawal if future studies prove disappointing. Or it could say no--a decision likely to meet with opposition from congressional members pressing for approval. Senator Orrin G. Hatch (R-Utah) wrote to the agency: "There is an occasion here to provide patients with a drug they need at no risk to public health, and I hope the FDA will be able to seize that opportunity."
Doctors who treat ALS also urged a speedy O.K. at the hearings. Even the slightest benefit is better than nothing, they said, and the typical two-study demand may be too high a hurdle. "We're not talking false hope. We're talking modest hope," says California Pacific Medical Center neurologist Dr. Deborah F. Gelinas.
HARD LESSONS. Whatever the outcome, the case offers drugmakers hard lessons. Lesson one: Large trials, though costly, may pay off. "I'm quite sure it would have been statistically significant if they had two times as many patients," says Johns Hopkins University neurologist Dr. Ralph W. Kuncl. Lesson two: Heed advisory committee requests for more data, even if you believe you've done enough.
For regulators, the case raises troubling questions. Should standards be lower for deadly diseases that lack treatments? Even now, the Senate is weighing a bill that makes it clear the FDA has the authority to approve drugs based on a single trial. And when should compassion override rules? Says University of Pennsylvania bioethicist Arthur Caplan: "It is morally relevant not only what the evidence is, but what the plight is of the people who might be helped." With an admittedly weak treatment and a horrific ailment before it, the FDA now must decide what makes for the smartest medicine.
To continue reading this article you must be a Bloomberg Professional Service Subscriber.
If you believe that you may have received this message in error please let us know.