Non-clinical bridging – Most 505(b)(2)’s Don’t Require Full Tox Package
- Posted by: Ken Phelps
- Published on: April 18, 2013
I have just finished a webinar under the sponsorship of the DIA dealing with non-clinical bridging. In this post, I’d like to share one of the case studies from my presentation to illustrate what should be considered during development.
The example is the development of gabapentin enacarbil by XenoPort/GSK for restless leg syndrome. Gabapentin enacarbil is a prodrug of Pfizer’s gabapentin, approved in 1993, indicated for the management of postherpetic neuralgia in adults. Virtually all of this prodrug, gabapentin enacarbil, is coverted to gabapentin via first-pass hydrolysis. How fast and where the prodrug converts to the active moiety drives the determination of the drug development program, or should. In this case, apparently it didn’t, since a full nonclinical program was conducted.
The FDA will regulate a prodrug that rapidly converts to the active moiety as if the active moiety was the administered drug. The sponsor will need to support the ‘pro-‘ part – enacarbil, in this case.
Thus, we can bridge to the approval package for gabapentin (the reference product) and then determine if there are any additional requirements for non-clinical based on a) FDA nonclinical requirements since the time the reference product was approved b) nonclinical findings in the public domain and c) studies needed due to a change in dose and/or duration, route of administration or indication-related.
In this case, there was no need to conduct any nonclinical studies except for studies to support the use of enacarbil. Yet, the sponsor conducted the full nonclinical program and the required clinical studies needed to support the new indication and filed a 505(b)(1) application. The original NDA did not reference the NDA for gabapentin (!). Upon review, the Agency issued a complete response letter (CRL), based in large part on the findings of pancreatic acinar cell tumors in a 2-year carcinogenicity study in rats.
To resolve the CRL, the sponsor revised the filing to a 505(b)(2) to reference gabapentin (yes!!). In the gabapentin NDA, the 2-year lifetime carcinogenicity study of gabapentin in rat showed the same tumor finding, and 1000 mg/kg/day was determined to be a “no-effect” dose. The sponsor provided toxicokinetic bridging data obtained under conditions similar to those used in the original gabapentin rat carcinogenicity study, at doses of 1000 and 2000 mg/kg/day. The gabapentin exposure (AUC) in the rat at the NOAEL (1000 mg/kg) was approximately 25-times higher than the human exposure from gabapentin enacarbil at the to-be- labeled maximum recommended daily dose of 600 mg. Additionally, the sponsor was able to use gabapentin market experience and FDA was able to confirm that no related tumors have been found despite millions of patients. See the FDA review document for a complete discussion.
Yes, the bridging was a calculation that should have been done to begin with. As we say, save time, money and lower your risk with 505(b)(2) development.
By the way, since the gabapentin enacarbil is really just gabapentin, why don’t physicians just use generic gabapentin? Maybe this has contributed to last year’s announcement that GSK/XenoPort would end their relationship on this drug effective the end of this month (April, 2013), with the commercialization reverting to XenoPort? The prodrug was awarded several patents and 5-year exclusivity.