A prominent New England researcher has rekindled a debate about the standards the Food and Drug Administration uses to speed approval of life-saving drugs, arguing that federal officials should insist on more rigorous proof that the drugs save lives and improve patients' health.
Dr. Clifford J. Rosen, director of the Maine Center for Osteoporosis Research and Education, issued the challenge to avoid a repeat of the public health safety crisis triggered by reports of heightened heart risks with Avandia, a once-popular diabetes drug.
Rosen is taking aim at whether the FDA should use shortcuts to approve drugs for type 2 diabetes, an ailment that afflicts 18 million Americans. But FDA critics say the agency should raise its approval standards for a broad range of drugs that it now speeds to consumers by lowering standards for proof that it works.
But such a seemingly simple change, demanding convincing proof of reductions in stroke, heart attack, or mortality rates, could lengthen the approval process by years. The debate is raising questions about the FDA's mission: Should drug safety take a back seat to expediting potentially life-saving remedies to market? Too often, critics say, the FDA uses substitute measures – like falling cholesterol levels or shrinking tumors – to speed approval times for AIDS drugs and therapies that combat cancer, heart disease, and high blood pressure. In Avandia's case, the FDA approved the pill based on its ability to effectively lower blood sugar levels that had zoomed too high in patients with diabetes.
In the future, Rosen said, federal regulators should insist upon solid proof that diabetes remedies enhance long-term quality of life by reducing risks for such serious side effects as heart attacks, eye and kidney ailments, and amputations.
Rosen came to his conclusion, which was published recently in a New England Journal of Medicine commentary, after serving as chair for an FDA advisory panel considering Avandia heart risks that emerged eight years after FDA approval.
For pharmaceutical manufacturers, changing the status quo could result in a dramatic increase in the number of patients they would need to enroll in clinical trials that would become longer and more expensive.
Rosen's article generated a raft of critical responses, which the New England Journal sped into publication on Aug. 29. Rosen agreed with critics that designing the trials he suggests would be a "formidable task," but argues that delaying new drug approvals in favor of more comprehensive studies would benefit patients in the long run.
The FDA approved Avandia in 1999 based on a handful of clinical trials. The trials that tested Avandia alone, rather than in combination with another drug, ran from eight weeks to one year and involved 2,315 patients. The stricter approval standards envisioned by Rosen would have required GlaxoSmithKline to test its drug in at least 5,000 patients, and to lengthen Avandia clinical trials to at least four years – enough time to measure quality-of-life changes that are meaningful to patients and insurers.
"Does it reduce cardiovascular morbidity? Does it reduce blood pressure? Does it improve quality of life? Well-being? Indices that matter," Rosen said. "Not just, 'OK, it lowers blood sugar.'
"Dr. Robert Temple, FDA's top clinical trial design guru, said applying such a sweeping change to diabetes drugs could have the unintended consequence of slowing to a crawl approvals of life-saving drugs. "Do you really want to not have any more blood-sugar lowering drugs until you do that?" Temple said.
The FDA did make such a evidentiary "sea change" decades ago, when it decided that improving bone density wasn't enough to gain approval for new osteoporosis remedies: The FDA wanted to see reduced bone fracture rates.
A clinical trial demonstrating improvements to bone density can cost less than $100 million and include as few as 1,000 patients. A late-stage clinical trial looking at bone fracture rates takes up to 6,000 patients, each followed for three years, measures more clinical outcomes, and can cost up to $700 million, Rosen said.
In the early 1990s, the agency embraced surrogate markers as a way to speed promising and life-saving HIV and cancer therapies to patients. It asserted its willingness to pull drugs off the market if the early predictive measures proved wrong, said Dr. Jane E. Henney, a former FDA commissioner.
The FDA's history with some surrogate markers has been spotty.
A recent flub halted the development of Pfizer's torcetrapib. The experimental product was designed to raise high density lipoprotein, a form of "good" cholesterol. The surrogate end point should have reliably predicted improved patient health.
"They do the trial. Instead of improving everything, it goes the other way," FDA's Temple said about heightened mortality associated with the drug. "Does that mean HDL isn't a good marker for cardiovascular risk? That we've been wrong all along? Another possibility is that HDL means more than one thing: There are different kinds of HDL."
But Temple defends the use of blood-sugar lowering for diabetes drug approvals, saying that early measure is closely tied to the drugs' long-term effectiveness.
David Nathan, director of the Diabetes Center at Massachusetts General Hospital, agrees. Lowering HbA1c, a key blood sugar, is "well-established" at reducing risks for such problems as eye disease, without committing 10 years and hundreds of millions of dollars to conduct lengthier clinical trials.
The problem, critics and supporters agree, is how to avoid getting "burned" by surrogates that ultimately don't pass muster.
Getting the starkest, black-and-white picture of whether a drug truly works would involve testing the experimental therapy against placebo, or dummy pills. Decades ago, such studies were conducted because some academics argued that lowering blood pressure levels may not be a good thing. Instead, the data showed effectively treating high blood pressure was good for patients, Temple said. These days, such a clinical trial would be unethical; depriving patients with diabetes of life-saving drugs would lead to serious health problems. "In some situations, it can be very hard to study the thing you most want to study," FDA's Temple said.
Dr. Jerry Avorn, a professor of medicine at Harvard Medical School, said the FDA could side step the controversy by giving conditional approval of drugs, based on such speedier surrogate markers as lowering cholesterol, for statins like Lipitor. Then, three years later the agency could require evidence the drugs "actually make people better."
No proof, the drug gets yanked and federal insurers, like Medicare, could rescind payment for such therapies. Avorn said that a such a change offered potential savings: "All you have to do is look at the billions of dollars that got spent on Avandia for so many years with an unclear clinical benefit and, perhaps, harm," he said.