QbD for Drug Device Combination Products – Part 2

Posted On

This is the part 2 of the interview from Manfred on the drug device combination product. If you haven’t yet, please listen to the first part of QbD for Drug Device Combination Products.

Today, we cover:

• Design Controls versus Quality by Design (QbD) – how is it different?

• When to plan for drug delivery design – Timing for project management

• Design review vs CQA (Critical Quality Attributes)

• How the Drug and the Device teams should collaborate

• QTPP (Quality Target Product Profile) for the drug device combination products

• Design inputs for drug device combination products

• Design space for drug device combination products

• Validation and verification of drug device combination products

• Design history file for combination products

Don’t forget to sign up for the next part of the interview!

Here’s the interview!

Manfred: Okay now, if we do look at the comparison, look at the design controls versus quality by design, which is very interesting and very important and there we’re looking at QbD and this QbD concept is out there since a couple of years.

And now for pharma product, this is the risk I mentioned for acceptance and per product and the design controls and the idea of design controls does exist and is inherent of the design control process for any devices therefore also for the device parts of the combination product. And also as listed in CFR Part 4, which came out in 2013, which is saying if you develop a device constituent part, you have to follow CFR 20, which is the design controls for the medical devices, which is out there since more than 20 years.

That means the design control process is existing since quite a while and QbD is fairly new compared to that. But… And I’m allowed to say that because by training I’m a pharmacist so that means by a pharmacist point of view that we are well behind and what we will behind and what is mandatory for devices is a little bit newer for drug product development.

And basically the parallel and QbD we look at the target product profile and we look at quality safety and efficacy and this is described in the design input. That means we have user requirements, we look at… we have risk management and also we do human factors study that means we evaluate – is the patient able following… to follow our descriptions, to follow our the instructions for use. If he’s capable to follow that and this is also demonstrated in tests with patients and if they can perform these tasks we expect them to do.

And side remark this is now… we also starting to do that for vials. It became mandatory last year from the FDA that means if we look at vials and syringes we are also performing human factors evaluations and also we think and we’re starting through that it make sense if we develop product for tablets for example. I have heard of examples from our competitors that develop a child-safe blister, unfortunately seniors were not able to open that one. And… so these are examples for human factors and to put the patient in the center and considering is the patient and also patients with difficulties – arthritis or whatsoever. Are they able to handle these products?

Sun: I have a question. You know in design controls, design development plan, a lot are focused around usability of the patient, how the drug is administered and so forth.

Manfred: Yes.

Sun: In QbD, when you’re talking on a typical list of QTPP (Quality Target Product Profile), these requirements are not usually listed. They are more focused on safety and efficacy, for example, purity and sterility and so forth.

Manfred: Yes.

Sun: Now when you develop these 2 combination products or combination products, do you segregate these requirements separately, how do you harmonize them at the end?  I know you submit them do different entities and so forth, regulatory agencies. So how do you manage that work process?

Manfred: Okay. We, of course, the QTPP is defined on correct product basis first and then we are using that product information as a design input for the combination product. Also this is described, I think it was mid last year, in the drug guideline of the FDA, that the drug product part should be used as input for the design controls of the combination product. So that means we have to phase what is available from the drug product. That means we have to consider viscosity, we have to consider that maybe protein cannot tolerate very strong shear forces. It could push into a needle, for example.

So with all this information, what is available from a drug product, we have to consider in the design controls for the combination product in the end. And this is also how we tackle it. That means we get the information from our colleagues, from the drug product development and also we give feedback. Okay, well if you want to have a very safe needle, maybe you cannot develop a drug product with a very high viscosity to use a very simple example. And this is also what we provide as feedback and that way we have that exchange. But in a documented way, in a workflow, we use the information QTPP and the description and the properties of the drug product, we use as design input. It’s one component of the combination product.

Sun: Very helpful. One follow-up question. What stage typically do this design inputs come in regarding the timeline of the pharmaceutical development? Is it around Phase 1 product development? Phase 2?

Manfred: Yes. We are involved earlier and earlier in the past. That means 2 to 3 years ago we are involved. Maybe during or at the end of Phase 2 and we received information by the way for Phase 3 – we need an auto-injector. So this doesn’t work of course. So for that reason, we have a process established now that means we are involved quite early. Quite early means definitely prior to Phase 1 even if we know we need a deliberate system in Phase 3 only.

Yet, involved very early in the process. In several cases, in very innovative systems where the device enables the administration. That means we are even involved pre-clinical phase. That means well prior to Phase 1 because if you have an example, for example, you have a vial syringe system you can use vial-syringe in Phase 1 and 2.

But if you cannot use vial-syringe to administer the drug product because it’s an intra-ocular treatment for example or very complicated, very complex treatment where you need the device to enable the administration. We are even involved for a proof of concept or definitely before Phase 1.

Sun: Excellent!

Manfred: And also our process we also, within my team, we also have a drug product expert, which make sure we are consulted and we have that exchange of information as early as possible. That the device aspect can be considered during the drug product formulation development and vice-versa.

Sun: I think that is very important, practical consideration when trying to harmonize devices with drug substance and drug product.

Manfred: Yes, it has to be a 2-way road. The one-way road is not good. That means if we have a formulation and the formulation is fixed, then we have to adjust the device accordingly.

Sun: Right.

Manfred: But maybe, it’s a balance. It’s a balance of maybe a very viscous formulation and a larger volume, for example. And they need… maybe it’s more convenient for the patient who administers a 2 ml preparation with a very thin needle than one 1 or .5 ml presentation which could be quite painful to inject. So this is a balance. And so for that reason, that exchange is very, very important.

Sun: Yes, let’s move on to the design reviews and the CQA which is very interesting as well.

Manfred: Yes, absolutely. Yes, okay. The CQAs we look at the mature properties and also we look at the limits from the drug product effective to have in the end the desired product quality. And in the design reviews, we look at the design output that means what we initially define in the user requirements

– it should be an injection, it should be in less than 10 seconds, should be convenient for the patient – whatever we design and we define there and the design output we are checking there. Did we really develop what we intended to develop and if that is the desired outcome. And of course, this is also always in combination or in alignment with the drug product properties.

Sun: The way I like to differentiate QTPP versus CQA is QTPP is what matters to the actual user including the doctors and the patients; clinical studies, and CQAs is what matters to the manufacturers; what we test at the end before we ship the product out. So is that the same parallel thinking in drug-device combination products?

Manfred: Yes! It is in practice. Okay. It’s pretty much in parallel. I equally agree with that. Pretty much in parallel. And what may be done in the design output is being checked and what did we defined in the design input is really valid. Did we really develop what we intended to develop? This is being checked in the design output. Also, of course, taking in consideration what you just mentioned. Absolutely.

Sun: Yes. So validation and verification throughout the process.

Manfred: Yes, of course. We also have the next step looking at the design history file. This is basically the reflecting documentation, what you just mentioned. These are verification and validation of causes documented there – I need a transfer, I need changes, it’s documented there. And the design history file is basically the design space within the QbD document. That means if you stay within your design history file, within the specifications which we try to somewhat broad and of course, it cannot be too broad because then the device might not be reliable or not be robust enough. But at the design space you can compare to the design history file – design history file is describing the area you are allowed to move within – and the same is true for the design space for the drug product, for example. That means as long as you stay within your design space or within your design history file, there are no regulatory activities necessary also. And one component in QbD, an important component also is the control strategy and also there… during the last one and a half two years, we made sure the control strategy is considering a cost drug product properties as well as device properties. That means if you look a text in time, if you look at other activities, it’s always taking into consideration the combination product. That means the device part including the drug product part for these activities.

Sun: And it’s interesting where the risk and assessment falls in place or QbD is mattered in identification of QTPP and CQAs but in design controls you put it as design input.

Manfred: Yes. It’s very early in the process but it’s regularly updated. And also this is also maybe a very interesting point. Risk management starts very early in the device development process. And also based on previous information from other devices – maybe from devices from competitors which are on the market.

That means any public information needs to be looked at and considered if you start your development already. And then risk management file is updated throughout the development process and also it’s mandatory to have an annual update of your risk file – not an update but an annual evaluation of your risk file. And this is very important to keep your design history file updated all the time.

Sun: Absolutely.

Manfred: And this is mandatory for devices and/or combination product. But it’s not at this point of time, to my knowledge, mandatory for drug product.

Sun: The risk assessment piece is actually considered as an executive summary of the dossier these days. It’s becoming more and more of importance to the regulatory agencies.

Manfred: Yes, absolutely.  I equally agree. Yes, basically.

If we move on to the next slide… if you look at device development and you have the device component, you have the drug product component or constituent part.

And also secondary packaging is extremely or became extremely important. If you put a tablet into a blister that was fairly simple – I’m exaggerating now – but if you deliver device and then you transport the device and then it arrives at the customer and it has been activated with a needlestick prevention device and it has been activated so it cannot be used by the customer or by the patient anymore that would be worse case that means the secondary packaging became very, very important for the overall quality of the device.

And for that reason, the FDA I heard comments, they are very strong on that and they have the opinion that secondary packaging is important. For that reason, it has to be part of the design history file because it is important for the overall quality of the device.

Sun: Absolutely.

Manfred: And the…so this is… that means this viewpoint… okay, well this is packaging only, that viewpoint will change significantly especially for these more complicated, more complex combination product.

Sun: Yes, I’m very familiar with this because one of the product I worked in the past was with the hemophiliac, where they have to reconstitute the drug end product to be administered by the patient. So this is of absolute importance.

Manfred: Yes, and this is what needs to be considered and also this is also the opinion of the authorities, of course. And also we do not want to have complaints because we did not consider that and also it would be very unfortunate to say that it did not work for the patient and this is completely unacceptable on that reason. So that means all of this information and this aspect has to be considered during development.

Sun: So on the slide, I see different requirements or QTPPs or could be CQAs – depending on how you look at it – of devices and product environment including the packaging. For example, devices in paramedic experience, feedback signals, sizes, and shape, so forth, the product typical of pH composition. But there are overlapping ones such as viscosity injection types, shear forces… which do these attributes go into both the device side and the drug side? How do you deal with the overlapping requirements or the parameters?

Manfred: Yes and this is the overlapping piece and this is exactly what we previously discussed and this is exactly where the interaction between the formulation and development colleagues and the device development colleagues. This is the… exactly the important piece of the interaction.

For example, if you have a high viscosity you will increase your injection time. If you have an injection time more than 10 seconds for 1 ml it becomes more and more unacceptable for the patient. That means they most likely will remove it prematurely prior to a complete injection. And that means we have to go down with viscosity or we have to make a decision – okay we cannot go down with viscosity because then the volume gets too large; we need to use a bigger needle or a different needle dimension.

And so that means, you have to balance these properties – they have to look at that – maybe it’s possible to modify the formulation, if not possible then we have to modify the device. But this is the point of discussion between device and formulation developer. And for that reason, it is very, very important to have this discussion as early as possible in the process.

Sun: And for QbD, the design space becomes much larger now. I mean by just looking at the physical and biological characteristics or now you’re looking at human behavior – it’s part of the design space and…

Manfred: Yes. Unfortunately, yes. This is one of the difficulties. Yes, I equally agree with you. And also size and shape and all that extremely become more and more important. How you hold it – that the shape forces you to hold it, your device in a specific way already. So, also that… all that can be controlled to some extent but not completely. So… sometimes users are very intuitive of how to use it or not to use it.

Sun: So in a practical sense, when you interact with the regulatory agencies as you’re developing a new combination product, of course, there could be infinite number of requirements, scenarios that a patient can go through – whether injecting drug. How much is expected from regulatory agencies from you, practically speaking?

Manfred: Yes, okay well to explain it in very simple way, there’s clear guidance for human factors. Prior to submission, we have to submit our human factors protocol. What and how we intend to test these devices, including the instructions for use. This is being submitted to the branch being responsible for human factors studies at the FDA, at EMA.

And then typically within 90 days – they make it also most of the time within 90 days – we get feedback if they would accept our approach to do this human factors testing, which includes the look and feel for the patient, the capabilities.

They also would consider if we pick the right patient population, if we pick the right statistical numbers for the patient and if we pick the right numbers for trained and untrained population and then we get the feedback then we execute the human factors studies and then this protocol and report is part of our submission package then. So that means prior to submission, there’s also… already a mandatory feedback loop with the FDA.

While it’s kind of mandatory you can do it without consulting the FDA but then you have a very significant risk that they are not accepting your submission because they disagree with one small step.

Sun: Thanks for the explanation.

Manfred:    Okay, now… Yes if I can move on.

Sun: Yes! Let’s move on

 Don’t forget to sign up for the next part of the interview where Manfred and I discuss the CPP (Critical Process Parameters) of drug-device combination products and the exciting trends of combination products!