Transparency in Action: FDA Comments on Pirfenidone

It's no secret: FDA review managers would like to make "complete response" letters public.

We've heard that directly from them, and now the idea is moving forward as one of a number of proposals from the agency's Transparency Task Force.

This idea, to put it mildly, provokes mixed feelings in industry. "Mixed" in the sense that sponsors love the idea of seeing their competitors' complete response letters, but hate the idea of having their own made public.

But a recent presentation by FDA officials during the American Thoracic Society's annual conference has us wondering what the fuss is all about.

As reported by The Pink Sheet (relying on a transcript by one of the Wall Street analyst firm, ThinkEquity, who attended the briefing), medical officer Banu Karimi-Shah discussed the recent complete response letter for Intermune's IPF therapy pirfenidone, and outlined the agency's position that the existing dataset does not meet the statutory definition of "substantial evidence," and therefore another clinical trial is required. Karimi-Shah also discussed issues around the adequacy of forced vital capacity as a surrogate endpoint, and the relevance (or lack thereof) of Japanese clinical data with a different formulation of pirfenidone.

FDA rejected the application May 4, over-ruling an FDA advisory committee that voted in favor of approval--though the committee had significant misgivings about the sufficiency of the data as a basis for approval.

This is not be the first time an FDA official has commented on a non-approval decision during a public forum, but the extensive discussion of FDA's issues with the application can hardly be described as routine. So we asked the agency whether the remarks were cleared by top FDA officials or posed any concerns regarding the appropriateness of discussing the elements of a Complete Response.

Here is what the agency told us, by email:

"The information presented by Dr. Karimi-Shah was presented and discussed at the Pulmonary-Allergy Drugs Advisory Committee meeting on March 9, 2010, so her presentation was already in the public domain. Intermune disclosed in a press statement on May 4, 2010, that the FDA issued a CR letter and the FDA requested an additional clinical trial to support the efficacy of pirfenidone, so this information was also in the public domain."

See? There is really no reason to argue about making CR letters public after all. Right?

Okay, our tongue is firmly in our cheek on that. We'll have much more on the reaction to the transparency proposals in an upcoming issue of The RPM Report.

Guest Post: At ATS, a Storm of Questions for IPF Drug Developers

Michael Gilman is the CEO of Stromedix, a Cambridge, MA biotech developing novel drugs to treat fibrotic organ failure. You can follow him on Twitter @Michael_Gilman. Interested in guest blogging for In Vivo? Drop us a line here.

At around eight on Sunday morning, just as the first sessions of the American Thoracic Society meeting got underway at the labyrinthine convention center in New Orleans, the skies opened up and unleashed ropes of rain. Thunder rumbled through the lecture halls, strobes of lightning lit the corridors. Power was lost, briefly snuffing out lights and laptops and stranding attendees on towering escalators. And it went on like that for two full hours — man, this place has some serious weather.

It was hard to miss the metaphor.

This year’s ATS was to be the moment in the sun for clinicians, scientists and drug developers working on idiopathic pulmonary fibrosis, a staggering, deadly disease for which there is no approved therapy outside of Japan. Perched prominently on the calendar just two weeks prior to opening day was the PDUFA date for InterMune’s experimental IPF drug, pirfenidone.

The relatively tiny IPF crowd is usually swamped at ATS by the hordes of folks working on asthma and COPD, but this year several significant IPF sessions were on the program. A pirfenidone approval, the first for the condition in the US, would have been a jolt of electricity to the IPF community gathered in New Orleans.

Alas, it was not to be. The FDA did not approve the drug and IPF investigators reeled. I don’t have an especially informed opinion on pirfenidone. Above all, I’m disappointed for patients, who are desperate for treatment options. But, given the bafflingly inconsistent clinical data and confused deliberations of the FDA advisory panel, approval was by no means a slam dunk.

The FDA’s action left meeting participants with a long and rather painful list of questions. What targets do we go after next? What are the right endpoints? What patients do we enroll? Do we even understand the real natural history of the disease? What does the FDA want? Will anything ever work? It also sparked remarkably strong emotions among pulmonologists, many of whom are absolutely convinced the drug will help their patients and others equally persuaded it doesn’t work.

But Monday morning in New Orleans dawned bright and sunny. And the first major IPF session of the conference packed the vast auditorium to fire-code-violation levels. The centerpiece of the session was a couple of densely-packed reports from an expert panel that had deliberated for three years on formal guidelines for diagnosing the disease and treating it.

Conclusion on the latter point: No currently available treatments were recommended, including pirfenidone. Clearly, however, the troops were undaunted. You could sense people picking themselves up, dusting themselves off and getting psyched to wade back into battle. They want to beat this disease.

Which leads me to ask the following question.

Why do we do this? We are, generally speaking, intelligent folk. We’re rational and data-driven. Yet, inexplicably, we continue to pile into an enterprise in which the odds are ridiculously stacked against us. Are we nuts? Masochistic? In denial? Or just relentlessly optimistic? Convinced that our next idea is going to be better than our last? What is it that fuels our passion to develop new medicines for patients when it so often feels like a fool’s errand?

I don’t have an answer, but whatever it is, it was on display in New Orleans this week. And it’s inspiring. --Michael Gilman

image from flickr user ray devlin used under a creative commons license

InterMune’s Pirfenidone: No + No = No, After All

InterMune’s roller coaster ride with the idiopathic pulmonary fibrosis therapy pirfenidone took a steep dive back down when FDA issued a “complete response” letter asking for another clinical trial on May 4.

That decision, in one sense, is hardly surprising: FDA already made public the conclusion of its statistical reviewer that the NDA for pirfenidone didn’t meet the definition of “substantial evidence” for efficacy. Slam dunk non-approval.

Except that conclusion was presented to an FDA advisory committee in March for its consideration—with the remarkable outcome that many committee members seemed to agree that the data wasn’t sufficient to meet the “substantial evidence” standard, yet they voted for approval anyway.

It sure looked like a case where FDA was ready to be very flexible in the interest of making a potentially useful drug available for a horrific and untreated disease. One committee member actual voted “no” on the whether there was sufficient proof of efficacy, “no” on whether there was sufficient evidence of safety—and then “yes” on approvability. (Or, as we put it at the time, “No+No=Yes.”)

Apparently “substantial evidence” isn’t as flexible as all that. According to InterMune, FDA says it needs another trial to demonstrate sufficient evidence of efficacy. It may have to be a survival trial, or perhaps a trial with forced vital capacity as an endpoint.

InterMune says it won’t know for sure what will be required to support approval until after it meets with FDA to discuss the complete response letter. Still, the pivotal trial included in the NDA took three years from start to finish; replicating it means a prolonged delay.

So much for a “flexible” standard for substantial evidence, huh?

Maybe not. There is one wildcard in all this: pirfenidone is approved in Japan and marketed there by Shionogi. FDA wanted to review the Japanese trial data, but InterMune provided only summaries; the company didn’t think it would be worthwhile investing the time and resources to make individual case-level data available to the agency.

Our impression of the advisory committee was (and is) that the agency wanted the committee to, in effect, give them permission to approve pirfenidone based on something much less robust than you would expect for, say, a COPD therapy. FDA got that permission.

Still, it isn’t hard to understand why FDA would at least want that data before taking a chance on approving this application.

It may be that the “substantial evidence” standard still turns out to be more flexible than it appears to be—but with the caveat that, no matter how “substantial” it is, FDA expects to see all the data.

image from flickr user RubyJi used under a creative commons license

Yes+Yes=No? Forest's Daxas and the Pull of Comparative Studies

That wacky Pulmonary-Allergy Drugs Advisory Committee is at it again!

Last month, we noted the remarkable outcome of the review of Intermune’s pirfenidone, where the vote in favor of approval was stronger than the vote that the drug was effective enough to approve—in part because one committee member pulled the nifty trick of voting against efficacy and against safety, but in favor of approval.

The rationale: the drug may not meet the letter of FDA’s definition of substantial evidence, but in a condition as horrible as idiopathic pulmonary fibrosis, evidence of activity is enough to allow approval.

Now, Forest Labs brings it COPD drug Daxas to the same committee and wins a narrow vote that the drug is effective (9-6), a narrow vote that it is safe (9-6)—but a fairly firm vote against a favorable risk-benefit profile (10-5). As we put it in the headline of “The Pink Sheet” DAILY here, the committee said it is safe and effective but not both.

That sounds like a real head-scratcher, though the truth is that the vote is more rational than it looks. In fact, nine committee members voted “no” to either safety or efficacy or both, so it is not surprising that a majority voted against approval. (Trust us: the math works—if you would like a complimentary copy of our analysis of the votes, email us here.)

Of course, there is that one outlier: Richard Honsinger of Los Alamos Medical Center Clinic, who voted yes on safety, yes on efficacy, but no on approvability. His rationale? That while the drug may meet a minimal standard of safe and effective, it should only be approved if it is shown to be better (either more effective or more safe) than alternative therapies prior to approval.

That would obviously be tough for Forest if FDA agrees.

However, we suspect there will ultimately be a different outcome. As we noted in The RPM Report here, Forest made some important changes to the NDA after submission—and FDA basically said it was too late to talk about those before the committee. But we suspect the path forward for Daxas will involve putting brackets around the patient population and applying a robust post-marketing program—which may very well include comparative trials.

Still, let’s hear it for the Pulmonary-Allergy Drugs committee for once again tapping into the important themes of the regulatory process these days.

With pirfenidone, it was a perfect marker for one theme: the way the new regulatory process can make it easier for products to treat unmet medical needs (especially in relatively small patient populations) to reach the market.

With Daxas, the vote is a perfect marker for what happens to products where you cannot find such a population: there will be a strong desire for a de facto superiority standard for approval.

We look forward to the next meeting of this committee—no matter what is on the agenda.

image from flickr user RubyJi used under a creative commons license

No+No=Yes: Pirfenidone and the Power of Orphan Drugs

Two wrongs may not make a right, but when it came to Pulmonary-Allergy Drugs Advisory Committee member Les Hendeles' views on the approvability of InterMune's idiopathic pulmonary fibrosis therapy pirfenidone, two "no's" made a "yes."

Hendeles, whose day job is professor of pharmacy and pediatrics at the University of Florida Health Science Center, first voted "no" to the question of whether InterMune's pivotal trials showed "substantial evidence" of efficacy in IPF, as measured by changes in forced vital capacity.

Hendeles was one of five committee members to vote "no" on that question; there were seven "yes" votes. Hendeles noted FDA's definition of "substantial evidence" and his view that the dataset--one study showing improvement in FVC, one failing to show a change, and no unequivocal indication of an overall survival benefit--didn't make the cut.

Then he voted "no" on whether InterMune had provided sufficient evidence of safety; there he was one of three "no's," vs. seven "yes" votes. Hendeles described pirfenidone as a "theophylline-like drug" and explained his vote by asking, "have you ever heard of Vioxx?"

Then he voted "yes" on approvability. That made him one of two committee members who ended up taking the formal position that InterMune failed to demonstrate "substantial evidence" of efficacy but that the drug should be approved.

His explanation of the apparent contradiction (as we note in "The Pink Sheet DAILY") was simple. Based on the indication of a mortality benefit and the complete lack of effective alternatives, he "would be on the first Delta flight to Japan," where pirfenidone is approved already, if faced with a diagnosis of IPF.

That, in a nutshell, explains the committee vote (and what is likely to be FDA's final action on the appication) perfectly.

Bear in mind that Hendeles is not some naif from academia with no idea how the regulatory process works. He has been involved in FDA advisory committees for three decades and has weighed in on many different aspects of the regulation of pulmonary drugs in that time. He knows the regulatory process pretty well.

The fact is that InterMune's dataset doesn't meet the letter of FDA's definition of substantial evidence. That's not just our opinion: FDA's own statistician says just that in her written summary provided for the committee.

But FDA is prepared to approve this application anyway, because the concept of substantial evidence is and always has been more flexible than that.

Plenty of sponsors have learned that the bar is often higher than the definition makes it sound. But this is an important case of where and when it can be lower.

IPF is a disease that all parties agree is horrific, involving a steady decline in lung function towards a certain and unpleasant death, with absolutely no treatement options that seem to do anything to help. It also has a relatively small (approximately 100,000 patients in the US) and reasonably well-defined patient population.

And these days FDA has the tools to approve drugs like this with more confidence. InterMune is proposing a Risk Evaluation & Mitigation Strategy to highlight the need for liver monitoring and limit the risk of phototoxicity. The company also plans to sell the drug via a centralized distribution system, though it is not proposing that as part of the REMS per se. (That seems like an increasingly common strategy, by the way--read more here.)

Committee members urged a registry, as well as deeper dives into potential markers of response. All that seems very easy for InterMune to do--and for FDA to accept as a reassurance that it will ultimately gather sufficient data to prove (or disprove) a mortality benefit from therapy.

Hendeles put it perfectly: Who wouldn't take pirfenidone, given its success in improving lung function scores in one trial, and consistent albeit not statistical signficiant sign of an increase in survival?

That may not be "substantial evidence" by the letter of the definition. But these days and for this type of therapy, it is probably sufficient.

image from flickr user RubyJi used under a creative commons license