Taking a Systems Approach to Inhalation Therapy Development

Inhalation therapies are notoriously difficult to develop, and history is full of failures, whether Pfizer’s Exubera, or generic versions of GlaxoSmithKline’s (GSK’s) blockbuster, Advair dry powder inhaler. This year, both Hikma and Mylan presented FDA with abbreviated new drug applications for generic versions of Advair, and both were sent back to the drawing board. 

“This shows the regulatory and clinical complexity of developing new inhalation devices that regulators will see as equivalent to the name-brand device,” says Steve Ellul, commercial director of Bespak, a contract development and manufacturing organization (CDMO) based in the United Kingdom. Working for both name brand and generic-drug sponsors, the company manufactures a number of different devices and components for different therapy areas, including about one third of the current 600 million/yr global output of metered dose inhalers (MDIs).

Modern manufacturing design technologies promote rapid proof of concept. “Combined with established design for manufacture tools and techniques, we can increase the probability of a successful outcome for our clients,” says Bespak’s head of product engineering, Mark Knowles. He points to MDI technology,  where significant gains have been made using these tools, along with modeling and prototyping, to improve product design.  

Improving patient compliance

But if inhalation therapies are hard to manufacture, they are often difficult for patients to take, particularly children and adolescents, who need to synchronize inhalation and device actuation in order to ensure that they receive the proper dose. Currently, significant resources go into instructing patients and their families on how to use inhalers properly. Cylindrical spacers offer a solution, but they are large and require maintenance.

As a result, between 76% and 94% of inhaler users are using their devices incorrectly (1), while 60% do not always take their medication (2). According to the Global Asthma Network’s 2014 Global Asthma Report, a significant number of deaths from asthma, particularly in the UK, might have been prevented by a combination of more, and more effective, treatment (3). 

A number of manufacturers are looking into alternative devices, including 3M, which is close to introducing the Intelligent Control Inhaler to the clinic for trials. The company expects to start clinical trials of the technology early in 2018, says Stewart Griffiths, head of product development at 3M. 

The device uses an electronic rather than mechanical actuator. As Griffiths explains, both powder MDIs and dry powder inhalers (DPI) pose challenges. With MDIs, the user must press the canister’s actuator at exactly the right time to get the right amount of drug deposited in the lungs. 

DPIs, meanwhile, are higher resistance devices that require higher inspiratory flow to break up the formulation and deliver the drug to the optimal part of the lung. “We’re working to eliminate the need for coordination and have the device manage that, so that the patient need only place his or her mouth around the mouthpiece and inhale to trigger the release of medicine,” says Griffiths. 

Ellul, Griffiths, and Knowles discussed best practices for working with contract manufacturers, operational excellence, industry trends, and the move to connected devices with Pharmaceutical Technology.  

Smart inhalers and the rise of generics

PharmTech: What trends are you seeing in the market?

Griffiths (3M): There’s an increasing emphasis on smart inhalers, which brings the question: How do we try and apply modern technology to relatively established products? Factors such as health economics and reducing total healthcare system costs are a key driver of smart inhalers. 

We need to examine how we can reduce overall healthcare costs by introducing technology to track patient activity and determine whether people are taking the right amount of medicines at the right time and adhering to prescriptions. There’s technology out there now that allows us to try and monitor these things, providing better insight into real-world situations than we’ve ever had before.

Ellul (Bespak): One of the trends we’re seeing is the rise of generic therapies, although, so far, generic-drug companies have not been able to succeed in developing versions of Advair. Over the next few years, we’re expecting to see new generic versions of branded Albuterol products in the United States, such as Proair, Proventil, and Ventolin. 

Another trend is the move to connected and smart devices.  It’s not clear how that will play out. There aren’t any imminent launches yet, but clearly, things are happening.

PharmTech: What are the greatest challenges to developing and manufacturing these treatments?

Knowles (Bespak):  The challenges comes from developing precision components and assembly that deliver to high volume and maintain the high levels of quality. We have been very successful with our customers as we customize our MDI designs to meet their needs. However, this results in many variants which creates a range of manufacturing tools and assemblies to manage and maintain. We manufacture over four million MDI’s per week, to six sigma quality levels (i.e., 3.4 failures per million devices).

By adopting robust design techniques and adhering to rigorous change-control management, we ensure that any variance in manufacturing product and process remain within specification. Any deviation is detected and we can act immediately, invoking our corrective and preventative  action (CAPA) processes. Some aspects of change are not in our control (e.g., raw material quality), so we install quality-control measures (e.g., quality inspections, quarantines, off line sampling) that provide safe guards. Continual risk assessment and impact analysis ensure these safeguards are regularly reviewed and updated.

PharmTech: When a pharmaceutical manufacturer starts an inhalation therapy development project with a contract partner, what are the key issues that have to be addressed when the project starts?

Griffiths (3M): From the CDMO’s perspective, one of the fundamentals when working as a service organization in drug delivery systems is that you really have to ensure that you’ve got a good product definition, that you’ve got quality, and a good process for gathering requirements from a partner. Both sponsor and partner must be aligned to start with, with the target product profile, for instance, and establishing the dosage form, whether MDI, solution, suspension, DPI, reservoir device, or capsule. 

Other crucial questions include: What shelf life do you need?  What cost of goods do you need? Commercial intent is needed from the start, as well. If you are looking more around the device, you must ensure that you understand the target patient population and that human factors are going to work with what you’re designing. 
Assuming that both sponsor and contract partner can get to that same starting point, the work can move into product optimization and compatibility experiments, considering long-term storage, and determining the project’s likelihood of succeeding with the right long-term stability profile.

Knowles (Bespak): By definition, a project has a beginning, a middle and an end. These types of projects typically follow a gated process. Each gate provides a check point to assess whether the project is still on track to meet the three key dimensions; commercial business case, technical feasibility and project costs  and timelines. A disciplined approach and adherence to these check points is key to a successful project.

Continual risk management flows through the life of the project. If done well, risks are driven down to levels that are acceptable, while knowledge and confidence are increased surrounding the likely project outcome.

Managing projects for success

PharmTech: How do you do that?

Knowles (Bespak): From a technical point of view, it requires establishing clear customer and clinical needs and a corresponding concept with a performance window, or design space, that will meet this need. The design space must be capable of meeting the agreed-upon level of cost and quality. Our projects are mainly high volume (i.e., over five million per year) drug delivery devices; for example, dry powder inhalers, auto injectors, and inhaler systems (valves, actuators, dose counters), but we also manufacture diagnostic devices. For these types of device, it is important to consider closely the clinical needs of the device (i.e., what capabilities will be critical to function) and prove that the technology can perform this need (critical to quality) to an agreed-upon level of confidence or sigma level. 

One example would be mapping the dose delivered performance required to the device functions that contribute to this critical need. Then, the assemblies and componentry technical requirements (e.g., dimensions, materials, tensile properties, surface finish) must be mapped or related to these functions. As a simple example, MDI dose (or shot weight) can be correlated to the chamber component volume in the valve mechanism. The volume variance of this component must produce a shot weight variance within an acceptable clinical tolerance. The design challenge is to ensure that the material and manufacturing processes are capable and, through normative use, meet this need. 

PharmTech: What modern design concepts are you using in development? 

Griffiths (3M): A lot of what we do comes back to project management. At 3M, we apply quality by design (QbD) and design controls for development of combination products and devices. A risk management process must underpin component development, where you’re looking for critical quality attributes. 

Then you must consider the manufacturing process, where you’re looking for critical process parameters. You apply risk assessment to all these factors in the product definition, to assess the likelihood that they could affect product quality. Risk management highlights the attributes that need to be explored further through product and process understanding work. Expectations are managed against the original project profile.

Knowles (Bespak): There are a number of design tools and techniques. The key issue is not to lose sight of the key goal in any design approach: to ensure the alignment of user need to design intent, and the level of capability, cost and quality you are trying to achieve. 

We use analytical (e.g. modeling) and experimental (e.g design of experiments through rapid prototyping) techniques to assess rapidly if a design concept will meet the design intent. We’ve been using 3-D printing since it came on board three years ago, and it has become an invaluable tool. It allows us to reproduce representative manufacturing parts very quickly within the lab.  

These tools promote robust designs through a marriage of analytical (e.g. modelling, 3D CAD), empirical (e.g. design of experiments) approaches, underpinned by SMART success criteria (Specific Measureable Achievable Reliable Testable) design documentation and evidence (verification and validation). Best practices for the industry include Lean Six Sigma and QbD tools. We regularly use design and process failure mode effects and analysis, which aim to design out product and manufacturing process errors.

Starting with the right information

PharmTech: Are clients using these approaches and providing you with the right information, or have you had to educate them in order to get what you need to start the work?

Knowles (Bespak): It varies considerably, as we work with many different types of clients. The more experienced companies have generally learned from past mistakes and have adopted/tuned their practices. For example, we are seeing evidence of companies designing the drug and device together, using the systems engineering or QbD approach. Convention has been to formulate the drug, and then find a device for it. If all components of a system are designed at the same time, however, the overall system can be optimized to meet the user needs; this leads to a more efficient and cost effective product.

Ideally, we are approached at the early design phase of projects, so that design for manufacturing can be done more easily and more cost effectively.  This happens with our own Innovations group in Cambridge, UK; we work with clients on their early concepts and prepare them, by design, for manufacturing.

PharmTech: Are you using analytics and process analytical technology?

Knowles (Bespak): New drugs promote new device design challenges; constant differences and changes in particle size, viscosity, density, and coagulation properties. For plume, droplet size and lung yield, our clients tend to use the electronic lung or Anderson cascade to optimize their formulation. We then use corresponding lab instrumentation, such as spray plume and particle size analysis to prove the active ingredient performance with the device design.

Several strands of physics are involved in drug delivery device design (e.g., chemistry, thermodynamics, friction, force). Knowing the relationship of these to the required system pharmaceutical performance is key. Where it is cost effective, we buy in house laboratory equipment (e.g. MDI leakage test equipment used for every release test). 

We tend to outsource infrequent analytical work such as surface and material tribology, strength analysis and material selection, using tools such as finite element analysis (FEA). For process manufacturing analysis, we take an approach that uses process failure mode and effect analysis (FMEA) with the goal of minimizing error and optimizing cycle time, yield, and output at each station (collectively known as overall equipment effectiveness [OEE] on the assembly). 

Design for assembly (DFA), which minimizes the number of parts, is a crucial precursor. Reducing the number of parts reduces the number of assembly points and overall complexity, saving significant cost. We are also looking into virtual reality tools , which allow the manufacturing process engineer to evaluate line performance in real time during the early design phase.

PharmTech: What improvements have you been making to drug delivery?

Knowles (Bespak): We are always looking for cost and performance advantages. The performance of an MDI depends on a valve and a pressurized canister, with elastomer based seal. Often, leakage can occur with the seal. Today, following the introduction of a new elastomer, static leakage is half of what it was two or three years ago. In addition, this new elastomer also has a longer shelf life. 

Patients, healthcare providers as development partners

PharmTech: What role are patients and healthcare providers playing in product design?

Ellul (Bespak): Their roles are becoming increasingly important, but now we are seeing strict expectations from regulators, particularly in the United States, that devices to be more fail safe, more intuitive, and that they require as little effort on the patient’s part to get the right dose.

In the US, MDIs have to have dose counters on them. Other countries don’t require this yet. These come as indicators that change color (e.g., from green to red), as the patient reaches the end of the supply, or an actual counting actuator that states the number of remaining doses, usually from 200 down to 0.

On the dry powder side, human factors tests are becoming much more important. The design should be intuitive and give patients all the cues they need, to use it as intended. 

PharmTech: How do you handle project management?

Knowles (Bespak): Project structure is important, and this is normally established at the very beginning of any project with project governance and tiered responsibility, involving the project review, program management, execution, and operations levels. This has to be done for all stakeholders, and defines the roles and responsibilities within the project. We adopt a ‘work package’ approach, and have a staged medical device design process called New Product Introduction (NPI). This is an example of a typical approach that is widely used for medical devices. The use of stage gates helps to enforce a high level of rigor and regular communications across sites. Work packages are communicated with a clear escalation process so that, as problems arise, they can be dealt with quickly.

Logistics as a drug development factor

PharmTech: How important are logistics considerations in respiratory therapy development?

Knowles (Bespak): They have become critical. One device we’re working with has dried reagents, and the transportation requirements are very important. You can’t design a device and then have it fail in the van on the way to the airport. This is why it’s important to capture all critical aspects of a design, including packaging and design, up front and before implementation. 

PharmTech: Where do you see smart inhalers fitting in?

Ellul (Bespak): It’s inevitable that the technologies will become more intertwined with drug delivery devices, and we have been discussing this issue with clients. At this point, it’s unclear how they will be applied and who will pay for what, whether the payers or patients. Then companies will need to access electronics and data management expertise, which will increase cost by orders of magnitude. However, the devices offer obvious and compelling benefits in compliance, and by allowing payers to monitor how well their devices have been used and to link payments to actual use of each drug, so there are benefits that could make a case for a payer paying a much higher premium for such a device. 

PharmTech: What are key failure modes for inhalers today?

Griffiths (3M):  There is a need to improve devices so that they improve patient inherence and competence, and eliminate variability in compliance. One challenge is flow control, given the high variability in the rate at which individuals breathe in. The deposition of drug depends on the rate at which you breathe in.

In our new inhaler development, we are establishing a ceiling on how fast people can breathe in, which will eliminate the need for training. The best way to ensure compliance is by incorporating ‘competence’ features directly into the device.

PharmTech: What role do you see patients and healthcare providers playing in future inhalation product development?

Griffiths (3M): Both groups need to be part of the development process. There’s an ever increasing emphasis on usability and human factors, and, in general, we need to get devices into their hands as early in the design process as possible in order to get their feedback. We have some informal interactions with patients and healthcare providers now, and we are running more formal studies looking at human factors under a clinical protocol. We are also working to ensure that we’re bringing in people from a variety of user experiences to ensure that we’ve thoroughly mapped the patient space for intended use.

References

1. F.  Lavorini et al., Respiratory  Medicine 102(4):593-604, 2008.
2. C. Restrepo et al., Int. J. Chron. Obstruct. Pulmon. Dis. 3(3):371-384, 2008
3. Global Asthma Network, 2014 Global Asthma Report, globalasthmareport.org, www.globalasthmareport.org/index.php

Article Details

Pharmaceutical Technology
Supplement: Outsourcing Resources
Vol. 41
August 2017
Pages: s28-s32

Citation

When referring to this article, please cite it as A. Shanley, "Taking a Systems Approach to Inhalation Therapy Development," Pharmaceutical Technology Outsourcing Resources Supplement (August 2017).