The Prasugrel Backlash: Another Black Eye for FDA's Advisory Committee Process

FDA is the first to acknowledge that a lot of things went wrong during the run-up to the prasugrel advisory committee meeting, not the least of which is how the agency handles allegations of intellectual bias among its members. For drug sponsors, it may mean that the rules of engagement have changed.

by Kate Rawson

In the eyes of critics of the Food & Drug Administration, a lot of things went wrong during the advisory committee review of Eli Lilly & Co./Daiichi Sankyo Co. Ltd.’s platelet inhibitor prasugrel (Effient).

An advisory committee member that had been skeptical of prasugrel’s efficacy was "disinvited" after Lilly raised the potential for intellectual bias. Not a single drug safety expert participated in the meeting—despite an application whose approvability seems to hinge on how the agency balances strong efficacy data against the potential for significant risks. And, last but not least, the committee heard only from one high-level official about prasugrel’s benefits and risks, despite internal dissent within FDA about a potential cancer signal.

Talk about a perfect case study for critics of FDA and industry. They see an agency that suppresses internal dissent, an advisory committee process that can too easily be stacked for or against an application, and an industry that has too much influence and control. Prasugrel seems to confirm all those complaints. No wonder a congressional committee has opened up an investigation into the "disinvitation" of a prasugrel advisory committee member.

FDA, of course, sees it differently, and the agency has logical explanations for every aspect of the committee review—including the role of simple human error in "disinviting" one member. But the fact that FDA officials even have to make explanations illustrates the credibility crisis that the agency—and its advisory committee system—continue to face.

Ironically, the February 3 meeting itself went very smoothly. Prasugrel is a high-profile, commercially vital application, and when the committee adjourned—having voted overwhelming in favor of approval, it looked like a huge sign of progress from industry’s perspective: a demonstration that an FDA advisory committee’s discussion of a tricky product application can be smooth and productive.

What the prasugrel review shows now, however, is that even the smoothest committee review does not necessarily exclude the possibility of a backlash after the meeting is over.

For Lilly, the post-meeting controversy means that the company is likely to experience a further delay in a final decision on prasugrel until FDA responds to the congressional inquiry. But it also underscores the much bigger problem for industry and FDA as a whole: restoring the integrity of a cornerstone of the approval process. (Also see "Avandia and the Commercial Impact of FDA's Credibility Gap" - Pink Sheet, 1 Sep, 2007.)

It is hard enough to get a product through an advisory committee. Now sponsors have to prepare for the possibility that a review can be too smooth—and even the appearance of meddling by industry can prove toxic. That standard may not be fair, but industry would do well to learn from Lilly’s example. FDA has admitted error in its handling of the committee, but—in hindsight—Lilly also erred in working too hard to avoid a rocky review.

Sanjay Kaul: The "Disinvited" Member

These days, the last thing FDA wants to do is abuse its advisory committee members, or take their service for granted. While the job is prestigious—although perhaps less so than in recent years—it is also time-consuming. And there aren’t real financial incentives: FDA does pay its members as special government employees, but meetings take experts away from their more lucrative practices.

Plus, there’s the "fishbowl" phenomenon: recent public attention on the FDA approval process and reporting requirements under new conflicts of interest guidelines have left some members feeling as though they are constantly under the microscope. For all those reasons, FDA has been having a tough time recruiting new members. (Also see "FDA Advisory Committees on the Brink: More Meetings, Fewer Members" - Pink Sheet, 1 Oct, 2008.)

The advisory committee that considered prasugrel has not faced those staffing challenges. Cardio-Renal has maintained a relatively full committee roster, and with the addition of three members in January, is just one of two committee without vacancies. (The other is Anti-Infective Drugs.)

One of those three new members is Cedars-Sinai Medical Center cardiologist Sanjay Kaul. A guest speaker at past advisory committee meetings, Kaul is known for being tough on industry—he expressed concern early on with drug-eluting stents, for example, and questioned whether the benefits versus conventional stents was overestimated. Other FDA advisory committee members have used the word "contrarian" to describe Kaul.

Prasugrel was to be Kaul’s first appearance as a full voting member of the Cardio-Renal advisory committee.

He never made it.

After being cleared to participate and added to the roster, Kaul was "disinvited" at the last minute due to a potential intellectual bias that FDA advisory committee staff said could not be resolved in time for the meeting. Instead of making the Los Angeles-to-Maryland trip and risk being recused after arriving, Kaul was contacted by advisory committee staff and encouraged not to attend.

"That is not the correct procedure," Center for Drug Evaluation & Research director Janet Woodcock acknowledges. "At every step of the way, some error was made by multiple parties, not invoking or following the correct way of doing this that led to the fact that he was disinvited from the advisory committee meeting."

A Lack of Clear Guidelines

Part of the problem stems from the fact that recusals for intellectual bias are relatively uncommon, and FDA does not have clear guidelines for vetting members for intellectual conflicts the way it does for financial ones. The agency has had general criteria for what constitutes intellectual bias in place since the early 1990s, but admits the document—just seven paragraphs long—is vague and requires case-by-case assessment.

Some cases are obvious, like the recusal of Drug Safety & Risk Management Advisory Committee member Sidney Wolfe from a recent meeting regarding the safety of propoxyphene drugs. In that instance, Wolfe, the director of Public Citizen’s Health Research Group, had petitioned FDA to remove the products from the market.

"I don’t think anyone would question that [Wolfe] should not have been allowed to sit on the committee, since he wrote the petition," Office of New Drugs director John Jenkins says. "We are carefully applying that policy for people on either side of a question. The same would apply to someone who had been very vocal in the media saying that prasugrel should be approved."

That description would fit Cleveland Clinic cardiologist Steve Nissen, the past chair of the Cardio-Renal advisory committee.

Nissen, a likely candidate to participate on the prasugrel advisory committee as an invited guest, says he "recused himself" from the meeting. He has both been vocal in his support for a prasugrel approval, and critical of FDA for delaying to give Lilly an answer on the application. (Also see "The Second Coming: Tysabri Ready For Blockbuster Status?" - Pink Sheet, 1 Jul, 2006.)

In 2008, FDA convened four joint pre-approval meetings, all of which were for potent opioids: a two-day meeting to consider Purdue Pharma LP’s tamper-resistant version of oxycodone (OxyContin) and a supplemental indication for Cephalon Inc.’s fentanyl buccal (Fentora); and a two-day meeting for King Pharmaceuticals Inc.’s abuse-resistant morphine (Embeda) and King/Pain Therapeutics Inc.’s abuse-resistant oxycodone (Remoxy).

By comparison, Jenkins says, prasugrel’s benefit-risk profile didn’t rise to the level where the agency needed to involve the drug safety advisory committee.

There are two major safety issues with prasugrel, both of which were discussed at length during the advisory committee meeting. The first was an increased risk of bleeding—a serious side effect, to be sure, but not unexpected given prasugrel’s mechanism of action, Jenkins says. "The bleeding risk is a well-recognized risk of the anti-platelet agents, so that wasn’t viewed as a risk that warranted a joint meeting."

A potential increased risk of cancer was another major safety issue, and the one that caused the biggest debate inside FDA during the review process. But by the time prasugrel got to the advisory committee, the agency felt it had that situation well under control, in large part because the data were reviewed by the Division of Drug Oncology Products, which questioned whether the signal was real.

"FDA didn’t come into the meeting thinking that there was a real signal of increased cancer risk on prasugrel based on our review, and we weren’t proposing a risk management plan beyond including the information in the labeling," Jenkins says. "For all those reasons, it wasn’t a joint meeting."

No Safety Experts For Prasugrel

Rather than invite the entire drug safety advisory committee to the prasugrel discussion, Jenkins acknowledges that FDA could have asked individual representatives to attend. The agency did just that for eight of the 24 pre-approval meetings in 2008, or one-third. (Also see "Open Disputes at FDA Advisory Committees: Get Ready for Institutionalized Cacophony " - Pink Sheet, 1 Jan, 2008.)

FDA says it is trying to do just that. "We’re trying to be very responsive to some of the concerns that have been raised over the last several years that we did not allow certain individuals to present their views at advisory committee meetings," Jenkins says. "So in situations where it is clear that there are very different views, and people are interested and willing to present, we allow for multiple presentations."

But as much as FDA would like to have multiple opinions at certain advisory committee meetings, Jenkins says the agency can’t always convince dissenters to speak up. "Sometimes the people who might have dissenting views may not want to present to the advisory committee, and we certainly can’t force them to present if they do not want to," Jenkins says.

That is precisely what Jenkins says happened at the advisory committee review of prasugrel.

Some reviewers were so concerned about a potential carcinogenicity signal seen during clinical trials that they recommended restricting prasugrel’s use to as little as one week, before patients switched over to the gold standard in anti-platelet therapy, clopidogrel (Plavix), marketed by Sanofi and Bristol-Myers Squibb & Co. (See "Cancer, Bleeding Risks At The Heart Of Prasugrel Review; Could Limiting Use Help?," The Pink Sheet DAILY, February 2, 2009.)

Why those reviewers chose not to speak to the committee is unclear. Some reviewers—like drug safety officer David Graham—are more than happy to offer dissenting views during advisory committee meetings. But others shun the spotlight, perhaps not wishing to go on the record with a different opinion, or nervous about exposing themselves to questions from the committee.

In the case of prasugrel, those dissenting views were spelled out in the briefing documents to the advisory committee, but when it came time for the actual meeting, only one FDA official—Ellis Unger, the deputy director of the Cardiovascular and Renal Drugs Products Division—delivered the agency’s viewpoint on the safety and efficacy of prasugrel. (See "FDA’s Prasugrel Presentation To Advisory Committee Eases Doubts On Approvability," The Pink Sheet DAILY, February 3, 2009.)

"Adequately Presenting Their Views"

Jenkins rejected suggestions that Unger’s presentation in any way represented an effort to steamroll prasugrel through the advisory committee. The ability to present dissenting opinions relies on the willingness of reviewers to step forward. Since FDA does not always have that luxury, FDA can use the briefing documents to outline differing views, and ensure that the presentation expressing all viewpoints.

In the case of prasugrel, Unger did precisely that, Jenkins says. "There were rehearsal sessions, and all the team members were included. I can only presume that the other members of the team...decided that he was adequately presenting their views. I know that Dr. Unger went out of his way to present the worst-case scenario in the cancer findings. He tried to incorporate the concerns that were raised by some of the reviewers in his presentation."

Jenkins also pointed out that the cardio-renal review division has a history of skipping an FDA presentation altogether during advisory committees. The attitude had been that if committee members read the briefing documents thoroughly, there was no need to rehash that information during a formal presentation.

That is changing—thanks in part to a 2007 petition from Public Citizen asking that an FDA official present at every advisory committee meeting. While FDA has not yet formally responded, it looks like the petition is having an impact.

"I tend to agree with him that it is important for FDA to give its perspective," Jenkins says. "So it’s kind of a new phenomenon for Cardio-Renal to be giving a formal presentation."

No Change in Outcome?

Technically, FDA made no formal evaluation of Kaul’s eligibility from a standpoint of intellectual bias. But for what it is worth, Jenkins has concluded from his own after-the-fact review that the TRITON abstracts alone were not enough to prohibit Kaul’s participation from the meeting.

"He was critically analyzing the data, and I didn't see any evidence that he had formed in his mind an opinion of the products that would allow someone to say in advance, well, we already know how he’s going to vote," Jenkins says. "We want our advisers to come to the table with an open mind to look at the data, hear the presentations, hear the discussion, and then give us their perspective based on all of that information, and not come to the table with pre-formed judgments."

Despite the mistakes made regarding Kaul's eligibility, Woodcock stresses that his absence did not change the outcome of the meeting. The committee voted 9-0 to recommend approving prasugrel despite an increased risk of bleeding and a potential risk of cancer.

"We had nine extremely competent individuals who were able to give us advice. We believe that it was a...mistake that Dr. Kaul was not included amongst the advisors at this meeting, but we do not believe that invalidates the results of the panel," she said.

Kaul’s participation may not have changed the overall vote for approval, but it may—ironically—have made the vote more credible. After all, a 9-1 vote with a strong dissent would probably have been preferable for both the agency and Lilly to the unanimous approval recommendation followed by questions aired in the media or Congress after the fact.

Has it really reached the point where a 9-0 vote in favor of approval is a bad outcome for the sponsor? The questions raised following the prasugrel review suggest that maybe it has. No sponsor wants to risk an outright rejection, but the Lilly experience suggests that working too hard to avoid some bumps during an advisory committee may be counterproductive in the end.