The FDA has published a final Guidance for Industry – ANDAs: Stability Testing of Drug Substances and Drug Products, ahead with the Jan 2014 implementation without comment. Quite surprisingly, many of the questions that industry has been asking regarding the inconsistency in harmonization between ICH guidance and the ANDA requirements are still not specifically addressed in the new Guidance.
The document states:
“Over the past few years, the Office of Generic Drugs (OGD) has received numerous inquiries about what stability data FDA expects in ANDA submissions. Currently, the only published direction from OGD is contained in a 1995 letter to industry which states that OGD will accept ICH recommended long-term room temperature conditions for stability studies (i.e., 25±2°C, 60±5% RH). Although adequate in the context of other guidance existing at that time, this recommendation is no longer sufficient to serve as a basis for stability testing for ANDAs.” and refers you to a list of 5 ICH quality guidance documents (Q1A (R2), Q1B, Q1C, Q1D, and Q1E). FDA notes that, while the ICH guidance documents were developed for new drug applications (NDAs), they believe that that the recommendations contained in those guidance documents should also apply to abbreviated new drug applications (ANDAs).
The specific direction given is that partially defined at various FDA/Industry meetings; namely that:
Data from three pilot batches or from two pilot and one smaller scale batch should be provided.
At time of submission, 6 months of accelerated and room temperature data should be provided and cautions that intermediate data should be gathered consistent with ICH requirements.
Manufacture and package product that is representative of the actual commercial process.
Provide a fully packaged primary exhibit batch.
Use product from all three batches when applying bracketing or matrixing under ICH Q1D.
Provide statistical analysis when appropriate,in accordance with ICH Q1E, Appendix A.
provide justifications for any deviations from the guidance
But where are the specifics? This set of generalities leaves the industry in the same lurch as it has been in since OGD began talking of the increase in stability requirements in 2009. Unless you have personally attended one or more of the FDA/Industry meetings, the specifics are a shot in the dark and, even if you did attend one of the meetings, the targets and FDA expectations from the podium actually morph over time and appear to be a bit more dynamic than the specifics of the ICH guidance documents referenced. For instance, unless you attended the FDA/GPhA meeting last week, how would you know from reading the ICH Q1A (R2) that OGD expects 4 data points rather than the 3 data points for accelerated and room temperature stability cited in the ICH guidance? The industry needs not only clarity and consistency, they need certainty. We hear that a Q&A document is going to be issued to address some of these issues, but industry cannot wait to see the details until the 11th hour, they have equipment to buy, people to hire, containers/closures to order, additional API and inactive ingredients to purchase all before 1/14/2013.
More on the New Stability Requirements for Generic Drugs
The changes on stability requirements for abbreviated new drug application (ANDA) submissions are predicated on two primary pillars. First, that OGD wants to harmonize with the ICH stability requirements so that everything is the same for new drugs as for generics and those approved elsewhere around the world. Secondly, according to OGD, about 50% of all recalls are for ANDAs and are usually stability related.
Let’s look at the first issue of harmonization. First, it is not true that all countries require data on three batches at time of submission for generic products. I believe that the EU and some other countries require only two. Secondly, when OGD says they want to harmonize with ICH, and ICH clearly identifies a requirement for three stability data time points (0, 3, and 6 months) for accelerated and room temperature submission for the first 6 months, and OGD decides it now wants to see four time points (at 0, 3, and either 4 or 5, and 6 months) for ANDAs, that is not quite what I think of as harmonization. Looks like old habits may die hard as OGD is accustomed to seeing 4 data time points under its current 3-month accelerated stability paradigm - namely at 0, 1, 2, 3 months. Thank goodness that the generic industry escaped with only 6 months of room temperature data rather than 12 months at time of ANDA submission. [Note - with median ANDA approval times of 32+ months, there will be plenty of time to see more stability data throughout the review process.] So, is harmony really harmony or is harmony a wolf in sheep’s clothing in this instance?
Next, let’s look at the claim that 50% of all recalls are related to generic products and the reality of the marketplace. To examine this proposition, let's look at some numbers first. OGD receives about 1000+ ANDAs a year, whereas the Office of New Drugs received in the order of 1/10 of that number. The most current cumulative supplement of the Orange Book (see below) lists the number of single source drugs (which is roughly translated into innovator products) as 2467 and the number of multisource drugs as 12,825. Even if we assume that another 500 of those multisource drugs may also be innovator products, that would bring the total number of NDA innovator product to about 3000 and the number of approved generic products to a number greater than 12,000. Granted, these are not precise numbers but I think you may be beginning to get my drift. There are a lot more generic drugs than innovator drugs approved and being marketed by about a 4 or 5 to 1 margin. Remember that about 82% of all prescriptions filled in the US are for generic products. If that is indeed the case and only 50% of the recalls are for generic products, then maybe we should be focusing our attention on why 50% of the recalls are for innovator products that currently provide the ICH data compliment (3 batches with 6 months accelerated stability and 12 months room temperature at the time of submission) as a condition of NDA acceptance.
While I firmly believe in additional requirements when there is an identified problem to fix, I do not believe in expanding requirements to fix a problem that may not actually exist. The fact that the generic industry has been so quiet on this issue is puzzling. At this point in time, it looks like the train has left the station, and firms had better be shopping for bigger stability chambers and more analysts!
At its June 4, 2013 GPhA/FDA CMC Workshop, Office of Generic Drug (OGD) officials began to peel back the onion on the revised stability program for abbreviated new drug applications (ANDAs). While we have been hearing the about the need for submission of stability data in ANDAs consistent with ICH stability requirements (a Draft Guidance document issued in September 2012 that provided few details) for the last two and a half years, and with the January 1, 2014 implementation date around the corner, the details are now beginning to flow to industry. Team Leader Suhas Patankar, Ph.D. of Division of Chemistry III outlined OGD expectations at the meeting.
Dr. Patankar noted that the new stability requirements will be applicable to all new ANDAs and DMFs, but not for post-approval changes, which I am sure was met by a collective sigh of relief from industry. He went on to indicate that OGD expectations are for 6 months of accelerated and room temperature stability data at time of ANDA submission on three pilot scale batches or two pilot scale batches and one small scale batch. Industry has asked for definitions of what those terms mean and here is what the Agency is thinking:
For solid oral dosage forms - A pilot batch is considered to be a minimum of 100,000 dosage units and a small scale batch can be less than the 10% of the proposed production size batch, but cannot be smaller than 25% of the other two pilot scale batches.
For parenteral products - Two pilot scale batches of at least 10% of the proposed production batch size, but not less than 50 liters [which is a change from the previous minimum requirement of 100 liters]. The third batch can be smaller than 10% of the proposed production batch, but not less than 25% of the pilot scale batches.
Split filling of bulk solution does not constitute discrete batches.
All batches must be of the same formulation, manufactured under cGMPs, have the same specifications, and must be manufactured at the commercial site.
OGD has also outlined its expectations for the use of active pharmaceutical ingredients (API) in ANDA batches. They expect a minimum of two different lots of API to be used in the pilot batches (however, for metered dose inhalers [MDI] and metered dose spray pumps, three different lots of API should be used). However, when two or more sources of API are proposed, there must be data on three exhibit batches to qualify the first source and one exhibit batch (on the bio strength) and comparative dissolution to qualify the second source with full 6-month complement of accelerated, long term and intermediate conditions studies. If first source is withdrawn, then additional data will be required to qualify the second source.
If 6 months of accelerated data shows significant change or fails any attribute, then OGD expects to see 6 months of data at intermediate conditions. This is something that the generic industry has not been doing. And of course, the applicant must be updated with real time stability data and (where applicable) intermediate data. OGD refers applicants for other dosage forms to a variety of guidance documents.
One inconsistency we have seen for storage conditions for liquid products (including parenteral products, suspensions, and semi-solids) has been clarified by OGD. Expectations for package orientation during stability is that samples be stored both upright and either inverted or horizontal for pilot batches and in the worst-case orientation (inverted or horizontal) for routine stability studies. For one of the primary stability batches, preservative effectiveness and preservative content should be conducted, where appropriate, at the end of the proposed expiration date. Also, extractables and leachables should be studied, where appropriate, as a one-time study. Another requirement will be the inclusion of data on reconstitution and dilution and in-use stability studies where the innovator product is so labeled. Firms should follow ICH recommendation for these studies.
In another presentation at the conference, Dr. Radhika Rajagopalan outlined the specific expectations for time points in the stability studies, which actually add a 4th time accelerated point, not usually considered for new drug application products. OGD expects at least 4 time points for accelerated conditions to be 0, 3 and 6 months and one point between 3 and 6 months (either 4 or 5 months). She went on to discuss bracketing and matrixing for ANDA products.
Additional details will likely come in the form of a guidance document, so keep your eyes peeled. But that is all (or even too much) for today!
