The Data Everybody's Talking About: Proof FDA is More Conservative?

Approval letters are down, "approvable" and "non-approvable" letters are up. "Black box" warnings have seen a four-fold increase over the last five years. Biopharma executives are now drawing their own conclusions about the state of FDA regulation—based on the numbers.

By Ramsey Baghdadi

Is it tougher to get drugs through the Food and Drug Administration than ever before? That is the question executives from the smallest start-ups to the largest biopharmaceutical companies are asking themselves as the string of bad news on new drug approvals filters out of the agency at the close of a very poor 2007.

The list of recent disappointments is long: Novartis AG ’s diabetes drug vildagliptin ( Galvus); GPC Biotech AG’s satraplatin, Wyeth’s depressive disorder drug desvenlafaxine ( Pristiq), Dendreon Corp. ’s prostate cancer vaccine ( Provenge), and Merck & Co. Inc. ’s cox-2 inhibitor etoricoxib ( Arcoxia) among many others.

That doesn’t even include the drug safety actions taken against GlaxoSmithKline PLC’s diabetes therapy rosiglitazone ( Avandia), Valeant Pharmaceuticals International’s Parkinson’s Disease drug pergalide ( Permax), Sanofi-Aventis’ antibiotic telithromycin ( Ketek), Bayer AG ’s heart surgery drug aprotinin ( Trasylol), and Novartis’ irritable bowel syndrome treatment tegaserod ( Zelnorm).

In the past, claims that FDA is slowing down have stemmed from anecdotes and individual cases where a company that felt burned over FDA’s handling of a particular product.

Now drug developers say they have the data to prove it.

For example, consider GSK CEO JP Garnier speaking to investors during a third quarter conference call: "I was looking at stats on the NDA’s going through the FDA. I think four out of five get an approvable letter now."

Garnier was apparently referring to a series of slides that are making the rounds of senior drug development decision makers. The charts tabulate publicly available information and, some argue, show a clear signal that FDA is indeed growing more cautious.

"I have always thought that they were too conservative," says one senior pharmaceutical company executive. "Now it has just been taken to another level, which, in my opinion, is not good for patients or for us as an industry."

There is no denying it: the data on new drug approvals is not pretty.

For example, only 18% of standard new molecular entities (NMEs) filed in 2006 were approved in the allotted user fee cycle period of 10 months. ( See Exhibit 1.) That’s a downward trend that began in 2001 when the rate was approximately 35%. There is also an upward trend in "approvable" and "non-approvable" letters being issued by the agency, with "approvable" letters hovering at around the 60% mark and "non-approvable" letters hitting a high of 25% in 2005.

How bad is it? One drug development consultant says the trend shows the industry is past the point of a drug approval drought. Now, he says, R&D pipelines are "dying." Whether that is an overstatement or not, it is clear that the product choices made seven years ago by sponsors with planned introductions in the 2007-2008 time period are not in sync with FDA’s approval criteria.

("Straight Talk from FDA: Wyeth R&D Head Reacts to Climate Change for NDA/BLA Reviews", in this issue (Also see "Straight Talk from FDA: Wyeth R&D Head Reacts to Climate Change for NDA/BLA Reviews " - Pink Sheet, 1 Nov, 2007.).)

After a small rebound to reach 22 NME approvals in 2006, the drug industry will be lucky to get 20 in 2007 ( See, "The Approval Drought Continues," The RPM Report, February 2007 (Also see "The Approval Drought Continues" - Pink Sheet, 1 Feb, 2007.))

The mean approval times (from submission to approval) for standard NMEs and new biologics doesn’t provide much optimism for drug developers either. Mean approval times have consistently increased since 2001, when the number fell just shy of 20 months, and are now running above the 25-month threshold in 2005. ( See Exhibit 2.) The average dipped slightly in 2006.

Probably most disconcerting for drug manufacturers is the skyrocketing number of new or revised "black box" warnings over the past five years. In 2001 and 2002, the numbers hovered around just over 10 box warnings a year, with a significant bump in the two subsequent years to over 30 black box warnings. Then, in 2005, the number doubles to over 60. ( See Exhibit 3).) Merck’s rofecoxib ( Vioxx) was pulled from the market in September 2004, a likely marker for FDA’s more aggressive use of tougher warnings.

Critics of the FDA point to the "black box" warnings about a potential increase in the risk of suicidality added to antidepressants in 2004 as a potential overreach by the agency. In February, the Centers for Disease Control and Prevention released statistics showing a sharp increase in pediatric and young adult suicides in 2004 for the first time in a decade, while use of antidepressant drugs simultaneously dropped. It’s not hard for critics to make the connection.

It’s not just FDA’s strictest warnings that are rising in a pronouncedly. Lesser warnings, contraindications and precautions added to labeling have also trended upward. ( See Exhibits 4 and 5)

"I don’t think that there is any debate at all that the standards have increased," says a senior official at a top 10 pharmaceutical company. Judging from the data, he may be right.

Even FDA is starting to acknowledge it. Top FDA officials say there has not been a message sent down to drug reviewers to be more cautious when reviewing applications. Instead, medical reviewers are taking it upon themselves to be more cautious based on the tense climate, they suggest.

In any case, the data provide a grim regulatory environment for the biopharmaceutical industry. And that’s the data that everybody’s talking about.