The FDA delivered a second approvable letter for Neurocrine Biosciences Inc.'s insomnia drug indiplon, surprising the company and investors with requests for additional clinical and preclinical studies.

During a conference call, Neurocrine President and CEO Gary Lyons said the company had been in labeling discussions with the agency for three weeks and had received "no evidence" that it should expect "anything short of an action going forward allowing us to market and commercialize this product."

Neurocrine's unnamed potential commercialization partner for indiplon was also surprised by the news. Lyons said Neurocrine had a deal "signed and ready to execute" with a "world-leading pharmaceutical company," but the future of the collaboration will now depend on discussions with the FDA.

Investors were caught off guard as well and responded by pushing the stock down to levels not seen since 1999. Shares (NASDAQ:NBIX) closed at $5.25 on Thursday, a loss of $5 for the day, or 49 percent.

Less than two years ago, Neurocrine was trading above $60 per share and FDA approval seemed imminent for indiplon, which had hit its primary endpoint in 72 studies and was partnered with Pfizer Inc. But in May 2006, the FDA delivered an unexpected approvable letter for the 5-mg and 10-mg immediate-release capsules and a non-approvable letter for the 15-mg modified-release tablet formulation. The setback shaved 62 percent from Neurocrine's market cap and prompted Pfizer to abandon the partnership. (See BioWorld Today, May 17, 2006, and June 26, 2006.)

For the modified-release tablet, the agency requested additional studies that could take several years to address, but no additional studies were required for the immediate-release capsules. Instead, the FDA asked for a reanalysis of certain preclinical and clinical studies to support approval for sleep initiation and middle-of-the-night dosing, as well as a re-examination of the safety analysis in the elderly and a revised display of the adverse-event table. Neurocrine submitted the required information earlier this year. (See BioWorld Today, June 14, 2007.)

Lyons emphasized that Neurocrine "adequately addressed" all issues associated with the first approvable letter. But the second approvable letter raised several new concerns, many of which Neurocrine said seem to be more related to the use of sedative-hypnotics in insomnia than to indiplon specifically.

In the new letter, the FDA asked for a preclinical study of indiplon during the third trimester of pregnancy, which Neurocrine estimated could take six months to complete. Chief Medical Officer Christopher O'Brien pointed out that all preclinical pregnancy studies thus far have been normal, and that the drug's proposed label specified that it was not indicated for use during pregnancy.

The agency also asked for a clinical trial of the 5-mg capsule dosed for 12 weeks in elderly patients. Again, O'Brien noted that the company had previously completed a shorter duration trial in the elderly, which it believed was sufficient.

Yet Neurocrine seemed most miffed about the FDA's request that it conduct a safety study comparing adverse event rates for indiplon with an unspecified marketed product. Lyons called the request "unprecedented" and said it seemed to stem from recent concerns about complex sleep-related behaviors such as sleep-driving and sleep-eating. Earlier this year, the agency issued a press release requesting that all manufacturers of sedative-hypnotic insomnia drugs strengthen their labeling language concerning such side effects.

Neurocrine said that in all its clinical trials, only one incident of complex sleep-related behavior occurred with 5-mg or 10-mg immediate-release indiplon capsules. "To find a difference between a marketed product and indiplon for such low-frequency events would require a massive long-term trial," O'Brien said.

During the conference call, Cowen and Co. analyst Phil Nadeau brought up the point that indiplon has a short half life - which makes it less applicable in sleep maintenance but more applicable in sleep initiation and middle-of-the-night dosing. Sedative-hypnotics used for sleep maintenance like Lunesta (eszopiclone, Sepracor Inc.) and Ambien CR (zolpidem tartrate extended-release, Sanofi-aventis Group) have a much longer half-life and might be expected to be associated with more adverse events, he said.

In a research note, Nadeau predicted the new trials would result in at least a two year delay in indiplon's approval.

Jefferies & Co. Inc. analyst Eun Yang was even less optimistic, writing in a research note that she considers indiplon's likelihood of eventual approvability to be "very low." She added that the "FDA's cautious stance on drug safety would weigh on the approvability of the product."

San Diego-based Neurocrine will request a meeting with the FDA to gain further clarity on the approvable letter.