On February 18, 2010, the U.S. FDA issued a complete response letter (CRL) for the application on XenoPort’s (XNPT) Horizant, a sustained-release (Transported Prodrug) gabapentin molecule, for the treatment of restless leg syndrome (RLS). In the response letter, the FDA stated that a preclinical finding of pancreatic acinar cell tumors in male rats was of sufficient concern to preclude approval for RLS at this time.

These findings were for preclinical data on XenoPort’s new chemical entity, Horizant; however, the FDA also acknowledged that similar findings were known for gabapentin at the time of its approval for refractory epilepsy, but concluded that the seriousness and severity of refractory epilepsy justified the potential risks. Therefore, we can infer that the FDA had concerns about this carcinogenicity risk signal dating back to 1993 on the original application for approval of gabapentin in epilepsy filed by Warner-Lambert.

The news clearly took XenoPort by surprise. According to management at XenoPort, preclinical data on Horizant showed a similar carcinogenicity signal to gabapentin. The FDA’s decision seems to stem more from the notion that approval in RLS simply did not justify the risk/benefit profile.

…Implications for Depomed…

We note that XenoPort’s Transported Prodrug technology differs significantly from Depomed’s (DEPO) AcuForm Technology in the way the gabapentin molecule is released in the gut. In fact, XenoPort’s Prodrug is a new chemical entity (NCE) that requires preclinical data, whereas Depomed’s AcuForm drug is a reformulation of the generic gabapentin that relies on previous data submitted by Warner-Lambert and uses the 505(2)(b) pathway for approval.

We remind investors that gabapentin has subsequently been approved by the U.S. FDA for post-herpetic neuralgia (PHN). There is 17 years of post-approval use safety data on gabapentin.

Depomed is currently has two separate controlled-release gabapentin molecules under clinical development:

  • Serada – currently in phase III trials for the treatment of hot flash / vasomotor symptoms (VMS).  Depomed conducted two phase III trials, BREEZE-1 and BREEZE-2, in 2009. The BREEZE program demonstrated safety and efficacy of the 1800mg dose of Serada (7 of 8 primary endpoints hit between the two trials). Management plans to initiate a third confirmatory phase III program, BREEZE-3 under a U.S. FDA Special Protocol Assessment (SPA) in April 2010.
  • DM1796 – currently nearing an NDA filing at Depomed’s North American partner, Abbott Labs, for the treatment of post-herpetic neuralgia (PHN). Chief data for the U.S. new drug application (NDA) will be based on the positive data from a 452-patient phase III program that offered up results in October 2009. We expect that Abbott Labs will be in position to file the NDA in March 2010. U.S. FDA acceptance of the NDA will result in a $10 million milestone payment from Abbott to Depomed, strengthening Depomed’s already solid cash position.

There are key differences between Horizant and Serada / DM1796.

1)     Depomed’s drug is not a new chemical entity (NCE). It is a reformulation of the generic molecule. Both NDA filings for Serada in hot flash and DM1796 for PHN will be 505(2)(b) pathways that rely on the preclinical safety data filed for the original approval of gabapentin by Warner-Lambert approved in 1993.

2)     The second key difference may be in how the U.S. FDA will view the risk / benefit profile of each target application. The complete response letter on Horizant noted concerns from the FDA on this preclinical data back when Warner-Lambert (now Pfizer) submitted for approval of Neurontin (gabapentin) for the treatment of epilepsy, but the agency felt as though epilepsy was a severe enough disease with a significant unmet medical need that the application justified approval. For restless leg syndrome (RLS), a non life-threatening / non-debilitating / non-degenerative disease, the risks were too great.

We would characterize hot flash in a similar light to RLS — non life-threatening / non-debilitating / non-degenerative. That being said, there are alternative medications for RLS, such as Mirapex and Requip. The hot-flash market is larger than the RLS market, and for many women hormone replacement is not an option. The U.S. FDA has stated they would like to see new non-hormonal treatment options available to women suffering from hot flash.

We believe that in the eyes of the FDA, hot flash is not as severe a disease as epilepsy. But they agency probably feels as though new treatments for hot flash are more urgent that for RLS.

Whether or not this pushes the risk / benefit profile over the hurdle rate for the FDA remains to be seen, but we would be surprised to see Serada held up at the FDA based on preclinical data relating to generic gabapentin. Neurontin was approved in 1993 and is widely available as a generic. There are an estimated 25 million prescriptions written for generic gabapentin each year. If the FDA has safety concerns on gabapentin, there are clearly bigger fish to fry than holding up Depomed’s application in hot flash.

Abbott’s application in PHN is a little more of a wildcard. Gabapentin is already approved for use in PHN. In fact, the majority of the 25 million annual prescriptions for the drug are in PHN, not epilepsy.

Depomed’s DM1796 offers advantages over the generic molecule, including a significant reduction in side effects such as dizziness and somnolence, as well as a more convenient once daily (QD) dosing profile versus the 3-4x daily for generic gabapentin. With positive phase III data in October 2009, we would tend to believe the application is strong and should be met with little resistance at the FDA.

However, if the FDA does truly have issues with the gabapentin molecule, it may simply hold up the approval of DM1796 or ask for additional pk / AUC data before they allow Abbott to commercialize it. The PHN market is very large. The disease is not usually life-threatening, but can be significantly debilitating.

We are going to assume that if Abbott does file the application in March 2010, then they must feel confident that the risk / benefit profile warrants approval. As noted above, FDA acceptance of the application results in a $10 million milestone payment from Abbott to Depomed.

Final Thoughts…

We have been bullish on Depomed in the past, and we remain enthusiastic about the development pipeline. A market capitalization of only $125 million today vastly undervalues the firm, in our view.

Depomed should have held cash on the balance sheet near $80 million at the end of 2009. Burn is declining thanks to the strong ramp in sales of Glumetza. Additional milestone payments on extended-release metformin technology, from Abbott on DM1796, from a European partner on DM1796, or technology deals similar to Covidien will all help keep the balance sheet solid.

In our view, the company is worth more than $2.40 per share today. However, the rejection of Horizant clearly creates uncertainty and the stock may be in a basing period until visibility improves. We think this perceived risk to Serada and DM1796, justified or not, will keep a lid on Depomed’s stock over the next few months. Over the long-run, we think it would be wise to average-down or establish a position on any additional weakness. Our target is $3 per share.

Read the full analyst report on “DEPO”
Read the full analyst report on “XNPT”
Zacks Investment Research