First of all, thanks to all who participated in our AAPS Webinar on The Role and Implications of QbD In the Topical/Transdermal Drug Development Process on February 6th. We did not have nearly enough time to address the questions that were submitted during the webinar, but your questions were collected and forwarded to me so that I could generate answers to them.
I appreciate your patience and hope that the answers are still valuable to you. If you submitted a question during the webinar, but I have not addressed it here, it means that I do not have any experience in this area and/or did not feel I could provide any meaningful response. By the same token, I invite the participants to contact me if they have additional questions or would like a more detailed response to the original question.
1: As a bigger Picture, do we need to have a DOE for all unit operations involved in the discovery to patient routine, or can only critical operations be targeted.
I don’t think there’s a simple answer. In my opinion, the amount of experimental work required for each unit operation would depend on the complexity and uniqueness of that operation.
I tend to approach these situations as a fresh skeptic asking the question “Am I convinced the filer has provided enough information to establish a sufficient understanding of the process based on first-principles or experience?”. If you answer “no” or “not sure”, then you’d be well-advised to conduct a systematic DoE to educate yourself and, in turn, demonstrate that you understand your process.
2: When developing a generic TDDS is there a need to establish the Design Space or can the Control Strategy (matching Innovator) suffice (e.g., release rate and BA/BE are pre-determined so wouldn’t only BE be needed)?
Since bioequivalent is not synonymous with identical (particularly in the realm of TDDS), there are examples of products that are bioequivalent, but of fundamentally different design than the reference product. In this context, I expect a filer would have to proceed as though the product was entirely new. The only real concession would be the pre-determined design criterion of transdermal delivery rate.
If we assume, for the sake of argument, that the new product for consideration is truly identical to the reference product; we must also assume that this assertion is based on reverse engineering (directly observable characteristics) of that reference product. However, reverse engineering does not provide a complete understanding of how a complex product is manufactured. It’s analogous to believing one can understand procreation through autopsy.
3: The issue with high amount of residual drugs in patches … is it cost of goods or environmental risk due to improper disposal?
The short answer is “yes” (both).
Minimizing residual drug for the sake of cost is a real issue, but since the pharmaceutical industry (and, in particular the generic pharmaceutical industry) is a competitive, market-driven enterprise, I wouldn’t think cost control is something a regulatory agency needs to encourage. APIs for transdermal patches tend to be (for lack of a better word) commodities that are not a significant factor in the cost of the product. Consider, for example, the overly-simple design choice of penetration enhancer concentration vs. API concentration. A developer would be most likely to minimize the concentration of the more expensive component which may or may not be the API.
Minimizing residual drug to improve safety, on the other hand, is clearly a regulatory concern. Much of the current debate and controversy regarding residual drug load in patches stems from the potential harm that residual fentanyl can cause (whether by accident or intent). But, I remember an even earlier concern with estrogens posing ecological and environmental risks in landfills and waste water effluent.
I see the initiative to minimize residual drug as part of the overall QbD vision to encourage elegant and intelligent design as well as an effort to reduce the risk to our patients and to our environment.
4: Do we need to evaluate the packaging components (tubes, bottles, etc.) in QbD for topical products?
I think so. Since the packaging (container/closure) is necessary to protect and preserve the product, I tend to think of it in the same terms as the product itself. (Although, I wouldn’t expect to have to conduct an equally detailed evaluation/design process.)
5: If the filling operation is not quite efficient, for example, it requires reworking of labels on bottles, something that is not considered an impact to safety and efficacy. How is that treated in regards to QbD?
If the process is not a routine part of filling, it would be considered ‘reworking’ as you’ve already indicated. The QbD implication would depend on whether the design process could reasonably be expected to make such rework unnecessary.
I can only speculate on your particular scenario, but if the filling of the bottles is outside the preset specification range, the batch simply fails the specification and relabeling is not an option. Although not a practice I personally endorse, it might be possible to unpackage and repackage (either under a different specification or after making modifications to the packaging process).
6: Please define both Cold Flow & Soakage. What level of both are acceptable?
I am not familiar with the term ‘soakage’, but consider ‘cold flow’ to be any migration of the adhesive beyond the boundary defined by the backing film/release liner. This could be during storage or during actual use. It is important to stress that a pressure-sensitive adhesive must be able to flow or it wouldn’t be a pressure-sensitive adhesive. If the adhesive is too soft, however, it can cause problems.
Cold flow during storage means the adhesive spontaneously emerges from behind the release liner/backing and comes into contact with the packaging. In severe cases, the adhesive in contact with the packaging can inhibit the patient’s ability to remove or apply the product or (worst case) a significant amount of adhesive can actually be lost from the patch rendering it sub-potent.
Cold flow during use means that the adhesive spontaneously emerges from behind the backing film while the patch is on the patient’s skin. This is very common and tends to depend on (among other things) the duration of wear. Often, this is merely a nuisance and the exposed adhesive collects dirt and fibers forming a dark ring around the patch. Hypothetically, excessive cold flow during wear can either increase the dose received by the patient (by effectively increasing the patch size) or decrease the dose (if the adhesive is lost or if delivery is negligible without the occlusion provided by the backing film).
7: Residual guidance is for future products or can be implemented for generics also?
I believe the Guidance is equally applicable for generic and branded patches.
8: With no good in vitro tests to predict in vivo adhesion and wear, how can QbD be used to enhance the design process early?
There are, of course, other aspects to patch performance and many of these ARE amenable to in vitro evaluation. To your point, however, one can do limited clinical screening for wear on prototype formulations and correlate measurable visco-elastic properties of those clinically tested adhesives to guide the design process for similar formulations.
I think of this approach like chromatography where a standard is necessary in order to interpret the test results.
9: Why do we need so much residual drug in a patch?
To a first approximation, the rate of delivery from the patch is proportional to the degree of saturation in the matrix (diffusion is a first-order kinetic process). In other words, the rate of diffusion will only be half its initial rate once its dissolved concentration decreases by 50%.
In order to achieve a consistent (pseudo-zero-order) delivery rate, the driving force for diffusion needs to be constant. One way to do this is have an abundance of drug loaded into the patch so that the driving force during the wear period changes very little. The consequence is that most of the drug never leaves the patch.
There are other ways to design patches to minimize this concentration-dependent reduction in delivery rate, but are not universally applicable for transdermals.
10: Can we design a patch having lower amount of residual drug than that of RLD? Is it approvable?
Absolutely. Multiple approved designs for bioequivalent fentanyl transdermal systems with different drug loads already exist.
11: How is the guidance to reduce residual drug applied in generic product development, such as a copy-cat formulation?
A true copy-cat formulation would have the same drug load as the branded product, however, that does not necessarily mean that the generic filer would be exempt from the requirement to justify the drug load.
It is hard to imagine an approvable generic product with a greater drug load than the branded product in the wake of the guidance, but there might be circumstances where the benefits outweigh the risks. It is more likely that generic developers would be encouraged to develop a novel, but bioequivalent system with a lower overall drug load than the branded product.
12: Is there any concern for shrinkage of the backing film during storage (which may also be confused with cold flow)?
This is theoretically possible, but I have never seen it. The backing will be under tension during manufacture and can relax when that tension is released. I have seen this many times, but the backing doesn’t relax uniformly. There is usually more tension in the web direction than across the web and the backing shrinks (recovers) more in one dimension than in another. This is a more immediate problem than exposure of the adhesive around the periphery because the patches end up looking like potato chips.
13: Can we use the specific polymers (tackifier)? How can we prepare this particular toxicology study?
Qualifying any new material for human use is a complicated process and depends on the nature of the material, the potential routes of exposure during therapy, the duration a patient might be in contact with the material and many other factors. A polymeric tackifier would have to be tested for its general toxicity as well as its potential for skin irritation and sensitization. In addition, it would have to be tested to understand with a reasonable degree of confidence how it will interact with the other materials in the proposed product.
Unfortunately, there is no way for me to provide any kind of meaningful answer without more information so I invite you to contact me directly and I might be able to be more specific.