What Can Industry Do to Make their Applications of Better Quality? FDA Wants to Know! But, I Have a Different Question!

I received an email yesterday from FDA (as I am sure many of you have) soliciting suggestions on ways to improve the quality of generic applications. FDA has set up a docket to receive your input and also wants to know the best way to share the suggestions with industry.  The body of the email is reproduced below:

FDA wants to hear from you!  We are establishing a public docket to receive input and suggestions from the public on ways to improve the quality of abbreviated new drug applications (ANDAs) and associated amendments and supplements.  Specifically, FDA is interested in hearing about difficulties sponsors are having developing and preparing their ANDA submissions.

FDA is also seeking input on how to best share suggestions for improving the quality of ANDAs with the generic drug industry.

Improving the quality of ANDA submissions will result in more submissions accepted for filing, fewer amendments and easily correctable deficiencies (ECDs), and ultimately, more generic drug approvals.

FDA welcomes comments at any time, but we encourage submission of electronic or written comments to http://www.regulations.gov or Division of Dockets Management, Food and Drug Administration, 5630 Fishers Lane, Rm. 1061, Rockville, MD 20852 by March 24, 2014.

While I will give the FDA credit for the outreach, I also have to speak up for the industry as well.  Over the years the requirements for ANDAs has done nothing but increase and increase, while at the same time FDA has been talking about ways to provide regulatory relief for post approval changes and risk- based review of ANDA and risk-based inspection of facilities.  I have been out of FDA for 20 years now and guess what – these were the same things we were talking about in the mid-1980s, the early 1990s and in the mid 1990s.  I do agree that quality is of utmost importance to assure safety and effectiveness of all drug products, but the real questions is, what is nice to know and what is essential to know, (or better put) what actually makes a difference  to the overall assurance of safety effectiveness and quality of the drug product.

I believe that when there is a clear showing that additional information will increase the quality of a drug product that it should be required, but when there is no clear showing of benefit and the ask is just that, an ask, then what is the real benefit and how will it be realized?

I can tell you something else that I have seen during my 40+ years in this segment of the industry (perhaps I was part of the problem for never being able to adequately address it when I was at OGD), and that is consistency of application review.  This is a problem among reviewers at all levels in the Agency, from completeness and acceptability reviews to chemistry reviews.  In the past we have provided the following advice to firms –  learn from your mistakes.  We told them to categorize the deficiencies received from OGD and, in your next submission, make sure you adequately address those issues.  Firms took our advice and did just that and the feedback we have gotten over the years has been less than promising.  Most of the firms that have adopted that approach have told us it does not really matter because they just get another set of different deficiencies either from a different reviewer or different division.  Many in the industry feel like they are trying to hit a moving target and I agree, to a certain extent.

I think it may be time for FDA to ask a different question and that is – what do we really need to assure that a product has the quality it purports to have and is safe and effective?  Perhaps if that is clearly articulated in a meaningful way to industry, then maybe industry will really know what is expected.  Just to make a point about what really impacts quality and thinking back to the generic drug scandal, when a number of firms were found to have fabricated all types of data in their applications, claimed to have performed tests that were never actually conducted, falsified records, created graphite data, among other things, I remember one thing that I never really was able to reconcile.  The FDA collected and tested 3500 samples of products from these cheaters and tested them at FDA labs and not one of the products was found to pose a safety or efficacy problem.  Now do I advocate that the FDA should look the other way when fraud is found or that standards should be lessened  for generic drugs because of this finding? Absolutely not!  But I do think that FDA should reevaluate the generic drug review process to see what is really essential to be in an application for them to make a determination that the generic drug product is safe and effective, has assurance of quality and is the “same as” the reference listed drug (RLD) upon which it is based.

The Agency should treat different products differently based on risk assessment and thus the review of a true oral solution that is qualitatively and quantitatively the same as the RLD should be different than a complex extended or delayed-release product and that should be reflected in the both the type and depth of review and data required for approval.  Isn’t after all a risk-based assessment really supposed to be based on the perceived risk?

I know that this oversimplifies the argument, but with the June 20, 2014 deadline coming up soon where the  FDA will be asking for three times the amount of data for ANDA submissions from a manufacturing and stability standpoint; with FDA asking for QbR and saying that if the QbR looks reasonable it will likely speed up review and we may not have to look as much behind the scenes into the data but are still treating all ANDAs for different dosage forms and complexity the same; with questions based on QbD  making many ANDAs seem like a Ph.D. dissertation, but it is unclear how will OGD ever get a handle on its application backlog and get applications approve  in a timely manner without clear guidance as to a consistent set of expectations and without adjusting some of those expectations based on a risk based assessment.