ANDA but No NDA – What to Rely on?
- Posted by: Ken Phelps
- Published on: July 30, 2013
Camargo participates in 2-5 PIND meetings each month and one thing we notice in the FDA minutes is that the boilerplate answer to ‘does the Agency agree this ….. is appropriate for filing under 505(b)(2)?’ keeps getting longer. Recently, the Agency (or, at least, some divisions) is informing sponsors that they cannot use a product approved under the ANDA (505j) regulations as a listed drug. In at least one instance, the same division had agreed to an ANDA for the listed drug and a couple of weeks later, for another PIND sponsor, said no to referencing an ANDA. We have not gone back to determine what will happen to the first sponsor’s drug when the NDA is filed.
What has happened to change FDA’s stance? I speculate that the basis of the change is that FDA has been painted into a corner by the recent SCOTUS ruling in Mutual v. Bartlett. For the purposes of this posting, the SCOTUS ruled that ANDA holders are prohibited by federal law from updating product labeling on their own. Supported by the decision in Wyeth V. Levine, the SCOTUS said the duty to update labeling rests with the NDA holder which the ANDA holder would then be requrieed to duplicate. Sounds neat and tidy. But wait, what if there is an ANDA without an NDA?
There are two primary ways we can have an ANDA without an NDA. One is that the NDA holder withdraws their product from the market because of generic competition or other commercial reasons (withdrawal for reasons of safety and/or efficacy causes the ANDA’s to be withdrawn too). The Orange Book is full of these ‘orphaned’ ANDA products – the RLD is an ANDA, not NDA product. “Our own analysis of FDA data indicates that out of 4,653 approved drugs with distinct ingredients, delivery routes, and strengths, more than half (2,438) are available in generic form. Of those, 1,062 are available solely in generic form; the only available versions of the drug received ANDA approval.” Brief for Marc T. Law et al. as Amici Curiae Supporting Resp’ts, Pliva v. Mensing, 131 S. Ct. 2567 (2011) (Nos. 09–993, 09–1039, 09–1501) at 18 (Brief of Pharm. Reg. Experts). Thus, FDA has not (so far, anyway) had a problem with new ANDA’s referencing an ANDA RLD, despite the RLD having no relgulatory responsibility for updating the labeling (FDA has expectations but no regulatory authority to require that these ANDA holders update their labels).
Another way we can have an ANDA without an NDA is found in the pre-Waxman-Hatch days. Those readers who follow this blog realize that the Waxman-Hatch amendment codified the practice of so-called paper NDA’s into 505(b)(2). These products were actually approved as ANDA’s prior to 1984. There are many of these products on the market today – all approved as ANDA’s but which are, in practice, original drug approvals. Thus, FDA apparently is waking up to the possibility that a 505(b)(2) application may reference an ANDA that isn’t current with safety information.
Camargo’s clients and other sponsors have successfully referenced these pre- and post-Waxman-Hatch ANDA’s for NDA approvals. Apparently FDA wishes this practice to stop. Several media reports indicate that FDA is considering ways to allow generic manufacturers to change their labeling (see, for example, http://www.medicalnewstoday.com/articles/262954.php). The FDA has given notice that it intends to propose a rule.
We have found some divisions that are, for the time being, willing to work around this issue Stay tuned for updates on this important issue.