FDA Needs To Put Meaning Behind “Clinically Meaningful,” Lipicky Says

FDA's increasing requirements for clinical meaningfulness in efficacy measures make it more difficult to predict the agency's regulatory actions, former division director Ray Lipicky says.

"I think FDA is getting harder to predict and is using the 'I am not convinced this is clinically meaningful' as an argument for being unsure of the 'risk' side of things," Lipicky told "The Pink Sheet" after making similar comments at a Sept. 18 Citigroup conference in Washington, D.C.

Lipicky, now an industry consultant, recently reviewed a company's data prior to FDA submission and "thought it was a shoo in." However, FDA found the application "not approvable." "I missed something. I still don't know what I missed," Lipicky told Citigroup, admitting that his own ability to predict FDA regulatory decisions has diminished since his time at the agency.

"This business of clinical relevance is sneaking in more and more and I have no idea where that's coming from. I don't know where in the agency it has become important."

"I know in one recent case that...question of 'clinically meaningful' came from people [at the agency] who obviously used to say to me 'you can't ask that question.' I don't know why it's there now. It's changing," Lipicky said.

He pointed to an example of an anti-arrhythmia drug where, although the drug had met the previously agreed-upon endpoint, FDA ultimately "could not be sure the endpoint was 'clinically meaningful.'" Lipicky noted that the agency was trying to balance the efficacy against a mortality risk.

FDA issued a "not approvable" letter for Sanofi-Aventis' Multaq (dronedarone) for treatment of atrial fibrillation/atrial flutter on Aug. 31. The NDA was based on the EURIDIS and ADONIS trials with a combined endpoint of all-cause hospitalizations or deaths compared to placebo. A separate dronedarone trial (ANDROMEDA) was stopped early due to excess non-cardiac mortality.

Another recent example is Pozen's MT-100. The firm stopped development of the migraine therapy after members of FDA's Peripheral & Central Nervous System Drugs Advisory Committee said a migraine indication for patients without nausea would not be "clinically meaningful" (¹ (Also see "Migraine Two-Hour Pain Response Efficacy Standard Suggested By Committee" - Pink Sheet, 8 Aug, 2005.), p. 10).

According to Lipicky, greater focus at the agency on risk-based "clinical meaningfulness" is obfuscating FDA's rationale behind binary decisions on drug applications.

The difficulty about this emerging standard is that "'clinical meaningfulness' is an undefined concept," Lipicky said, noting the lack of clarity surrounding measures of clinical benefit such as exercise tolerance (i.e. 10 seconds longer walking on a treadmill), a decrease in SPID scores for pain medicines and decrease in depression scores.

While such measures differentiate active drugs from placebo, "I have absolutely no idea how one would 'validate' such scores as to their clinical meaningfulness," Lipicky said.

"If a drug were shown to be as safe as placebo and some questionable clinical benefit were shown, I think FDA would say okay. But, the safety databases are not large enough, and FDA needs something to hide behind - [such as saying] the effect is not clinically meaningful," Lipicky said. "So it's fuzzy thinking."

In the post- Vioxx era, FDA has frequently been charged with taking a more risk-averse stance in developing clinical trial requirements and approving drugs.

However, former FDA Office of Compliance regulatory counsel Frank Sasinowski noted that the changes in FDA's requirements, even if they do reflect a reaction to a recent drug safety crisis, are part of the natural evolution of drug regulation.

"We've always been in the post-something world; we're in the post-thalidomide world, we're in the post- Rezulin , we're in the post- Duragesic , we're in the post- Seldane ....Now we're in the post-Vioxx world," the Hyman, Phelps & McNamara partner said.

"Events such as Rezulin, Seldane or Vioxx cause a momentary spike in the rate of acceleration of that slope of ever-increasing requirements, but it's...part of the normal incrementalism that is part of the regulatory review process," he continued. "I think there is incrementalism occurring in safety and efficacy domains. I don't see it as unusual. I don't see it as subject to political whim. It's the march of science."

"Vioxx to me is a precedent-setting business," Lipicky maintained. "It was sort of [the] higher-level agency making decisions...to examine something from a risk point of view. We don't want to discuss benefit as if there was some absolute risk that one knew was unacceptable and Vioxx settled it," he said.

"Safety is one of those things...that nobody knows how to approach. There is very little documentation written," Lipicky added.

Panelists agreed that as the agency's comfort with uncertainty - either in terms of safety or efficacy - has diminished and requirements for pre-approval studies have increased, the issue of clinical meaningfulness has remained "really amorphous." Thus far, the only form of clarity seems to be coming from inferences from recent actions on applications.

Former CDER supervisory medical officer Tom Garvey noted at the Citigroup conference that a subset at the agency seems to be driving industry to more clearly define the criteria for "clinical meaningfulness."

There are several initiatives at FDA which are aimed at formalizing clinical study requirements and identifying clinically meaningful measures, including moves to find validated biomarkers and the agency's oncology endpoint initiative. The Biomarkers Consortium, a cooperative effort between industry and government offices, was publicly unveiled last week (² , p. 3).

During the most recent meeting of FDA's oncology endpoint project on ovarian cancer, FDA Office of Oncology Drug Products Director Richard Pazdur said an impact on progression-free survival would be clinically meaningful only if the trial was sized large enough for survival (³ (Also see "Ovarian Progression-Free Survival Acceptable If Trial Is Powered For Survival" - Pink Sheet, 15 May, 2006.), p. 19).

In relation to the approval of Bayer/Onyx' Nexavar (sorafenib), Pazdur noted that for oncologic approvals not based on survival, the focus should be on "the magnitude of change" on the endpoint. The agency cited Nexavar's clinically meaningful impact on PFS as a reason for approval of the kidney cancer therapy (⁴ (Also see "Magnitude Of Effect, Not Survival, Is Key To Approval For Oncologic Nexavar" - Pink Sheet, 2 Jan, 2006.), p. 10).

Lipicky suggested that rather than actually examining the magnitude of the treatment effect, the real problem for FDA is the inadequacy of safety databases.

"In the past, magnitude of benefit played no role in approval, but played a large role in how labeling got written," Lipicky said. "Now, when faced with an indeterminate 'safety' database, FDA is likely to say, 'the effect is so small that it cannot be approved.'"

Lipicky suggested that sponsors focus on large efficacy studies, which also produce large safety databases. "My advice to everyone is that you got to do big trials. You got to be thinking in the 20,000 range. And you've got to be thinking of blowing away the classical ways of developing trials. Your first dosing should be in patients," Lipicky said.

In the cardiovascular setting, if the trial is less than 10,000 to 15,000 patients "you're crazy," he added.

Sanofi plans to resubmit the dronedarone NDA in 2008 with data from the ongoing ATHENA trial. The firm recently expanded the ATHENA sample size from 3,700 to 4,300 patients to attain planned event rates.

Panelists advised that given operational differences between the agency's divisions, sponsors should focus on understanding the nuances of FDA's review decisions and try to improve communication with the agency, instead of worrying about top-level announcements.

"I can say this now that Dr. von Eschenbach isn't in the room...what goes on in the top policy levels doesn't affect the actual INDs, NDAs, and BLAs that are under review," Sasinowski said. Instead, sponsors should focus on communication with the review division.

Garvey acknowledged that regulatory certainty can come from a single source: FDA Office of Medical Policy Director Bob Temple. "Leave it to one person, and you know who that one person is. There's no question about it. It's been that way for some 30 odd years...Temple, he's the guy. That's the answer. If you can get to Temple, and Temple is still around, that's the answer."

- Turna Ray