When developing a new drug-device combination, pharmaceutical companies must employ a long list of disciplines and sciences. Among these, design for manufacturability and scale-up (DFMSU) does not always play as prominent a role as it should. But as many in the industry have discovered, ignoring DFMSU until late in the process can have a negative impact on a product’s ultimate success. For this reason, companies must be vigilant in ensuring that DFMSU is taken into consideration from the early stages of product development. 

The Impacts of Decisions

By asking ourselves what could possibly go wrong during the development of a product, it is easy to see the ways in which DFMSU can have an impact. A product may perform poorly, resulting in a failure to meet specifications. It may be too costly in terms of either unit cost or capital cost, or may take longer than planned to reach the market. In retrospect, we can often trace the steps back to the decision from which a bad result springs, but it is rare for this bad decision to make itself obvious in the moment. For example, at the lab scale, a process with a very poor yield may not be an issue. But as the product progresses further into development, improving the yield will likely become much more important. Paying attention to DFMSU principles throughout the entire timeline of product development can help all players involved stay on the lookout for these decisions. 

In the development timeline of a new product, the number of required decisions is initially very high, while the implications of these decisions are low. However, as the timeline progresses, these situations reverse. The number of decisions decreases and the implications of decisions become more serious. Unfortunately, the impacts of early design decisions are often not apparent until later on, at which point options to address any issues are reduced. For example, the cost of a product is not necessarily an important metric, early on. At the early stage, product quality and capabilities are bigger considerations. However, once the product quality is demonstrated, cost becomes a serious factor during scale up. 

By incorporating DFMSU, companies can implement a more disciplined process to ensure a product is developed that meets all quality requirements, cost goals and timelines. With this path, undesirable or detrimental decisions can be more easily avoided from the outset. DFMSU is part of a robust development process to insure companies can live long term with the product design decisions they make at each stage in the process.

DFMSU in Early Technology Development: Finding a Happy Medium

Design controls are an important part of any discussion of DFMSU. This term refers to the formal methodology for management and documentation of product development projects that is required for medical devices and combination products. Key components of design controls include:

• a design and development plan,
• design history file,
• specifications for design requirements,
• risk management processes, and many other factors.

Each of these components is a highly complex process in its own right, requiring expertise and careful management to ensure compliance. 

If we imagine the application of design controls as a spectrum, on one end would be a process in which no one even considers DFMSU during the early development of a product, resulting in a device that is very difficult to manufacture. However, on the other end of the spectrum is “analysis paralysis,” in which excessive rigor during exploration of the early technology concept slows the development of the product. A happy medium, therefore, is a phase-appropriate application of design controls, which both recognizes the importance of DFMSU and moves efficiently in early development. 

To begin any development project on the right path, good design input is critical. It is necessary to determine the key requirements of the project early on in order to guide development, with additional detail added as the project progresses. For example, in the development of a DPI device, a device that is intended to deliver 30 doses will have different technical options than a device designed for 120 doses. Similarly, the target dose desired from the device will drive some of the technical decisions. These targets must be defined early in the process in order to create the framework in which the subsequent steps develop. 

Additionally, multifunctional participation is necessary from the early stages of development.
Expertise is needed not only in manufacturing, but also in regulatory and technical areas, as well as marketing. Input from all of these disciplines can inform technical decisions, as well as establish buy-in within the organization. For instance, if injection molding is envisioned for manufacturing, it is advisable to consult the injection molding partner in order to make decisions that are most compatible with that process. 

Maximizing the Prototype and Process Design

Prototype development also plays a major role in the early stages of product design, and should be undertaken carefully with an eye toward final manufacturing. The varieties of prototypes involved at this stage include industrial design (or form study) prototypes, subassembly breadboards and functional prototypes. Fabrication techniques used for these devices might include hand fabricating, CNC machining, stereolithography, rapid injection mold tooling, computer simulations or other techniques, all of which have their own implications for design. Companies have many choices as to how to develop these devices, but the ultimate goal should be to evaluate competing concepts and identify the most promising one. This stage also helps refine the device’s technical requirements, as well as get the product in the hands of users for testing and identification of issues. Finally, a well-designed prototype can be an important tool to persuade partners or investors in the product. It serves as a tangible marker of progress and an important tool in helping the project come to life. 

While focusing on device design in the early stages, it is also important to remember process design and development. Because many process decisions are driven by device design, these two factors should be considered side by side. Early device design decisions often have the effect of locking in processes, and because process decisions can have a significant impact on the product’s unit cost, these early decisions have long-term implications. When selecting processes, care should be taken to use existing, proven technologies whenever possible in order to avoid the regulatory and design challenges of using a novel technology. Another important measure is to avoid secondary steps such as ultrasonic welding, heat staking, or heat sealing whenever possible, since these steps add cost and complexity to the product. Again, the scalability of a process should be considered early in development, and industry experts should be consulted in order to gain their insights on the processes under consideration. Finally, processes should be selected with an eye toward maximum downstream flexibility, which will allow more freedom as the project progresses. 

Developers should map out a high-level view of the overall manufacturing process, including a sub-assembly strategy and facilities requirements. Then, a failure mode analysis of the planned processes can be undertaken, by first identifying the high priority processes, then evaluating their feasibility. While resources are limited in the early stages, emphasis can be put on these high priority processes, while lower priority issues can be resolved at later stages. 

As these factors show, in early development, pharmaceutical companies must clearly develop their roadmap before starting the journey. Full consideration should be given to what is needed from the commercial process as development progresses through the early phases. Finally, experts should be leveraged throughout in order to ensure their input is “designed in” and to avoid surprises later on. 

DFMSU in Later Stages: Guiding Work Efficiently

At later product development stages, the level of rigor in applying the specific aspects of design controls increases. Exact details of this process are determined by regulatory guidelines, company policies and risk tolerance, but in general the steps include finalizing the design input and proceeding to formal design verification, design reviews and failure modes and effects analysis.  

Refinements also should continue to be made to the process development by identifying the commercial processes. A proof of concept should be performed on high priority processes before finalizing the device design, and sufficient data should be generated to justify the large capital investments required to move forward. Additionally, the process flow map can be finalized at this stage, with the processes optimized to reduce cycle time and maximize yield. 

Design Considerations

As development progresses, increasing consideration must be given to how to mold and assemble the product’s parts. When designing for injection molding, manufacturers aim to minimize the cost of molded parts and maximize their molding reliability to ensure measures such as complete shots and no flash. Companies should determine the type of tooling needed for this process based on the product’s stage in development and capacity requirements. In early technology development, a single cavity with manual tooling is typically acceptable. For pilot tooling, manufacturers should implement tooling methods that are representative of final commercial tooling, which guides high cavity design. Throughout the process, it is vital to continue consultations with industry experts and perform a critical review of every part. When possible, parts should be designed to eliminate undercuts, avoid parting lines on critical-to-function surfaces, to provide draft that enables good part ejection from mold tools, with gate locations to ensure complete filling of the parts, among other considerations.

When designing for automated assembly, typical objectives include minimizing capital costs and maximizing throughput. Additionally, measures should be taken to maximize flexibility when possible, with a design that enables low and high volume assembly options, as well as flexibility in terms of where various sub-systems are assembled. Key design features that simplify automated assembly include adding lead-ins and alignment features, enabling single axis pick-and-place assembly, and avoiding slower secondary processes. As with the injection molding process, companies should continue to utilize industry authorities and encourage three-way communication between the design group, molding vendor and automation vendor. 

Unit Cost Estimating 

After ensuring DFMSU principles are applied from early to late stage development, one of the final procedures that must be undertaken is unit cost estimating. A unit cost estimate is a critical input to many business processes including business opportunity assessments (BOA) and net present value (NPV) calculations. In terms of device and process design, a well prepared unit cost gives clear visibility to cost contributors and allows identification and prioritization of future work. The cost estimate can provide visibility on costs well before they are locked in, allowing for adjustments to be made before they result in significant implications to the project. 

By performing a sensitivity analysis as part of the unit cost estimate, developers can calculate a distribution to the unit cost, as opposed to a single value, which allows for a more realistic projection of the final results. The sensitivity analysis can also raise the visibility of factors that have the largest potential to contribute to variability. 

Leveraging the Right Partners

DFMSU is a complex process that should be applied throughout the development of a new pharmaceutical product. While the principles can be daunting for pharmaceutical manufacturers that lack the in-house capabilities and expertise to apply these guidelines, a reliable manufacturing partner can provide invaluable assistance on this front. Companies should seek a partner that offers the experience they need scaling up a project, navigating the regulatory process and manufacturing competitively at a commercial scale. By involving experienced DFMSU voices from the beginning of the development process, pharmaceutical companies can help assure a smooth path for their product. 
 

Rich Sitz is Dry Powder Inhaler Technology Platform Leader, 3M Drug Delivery Systems. He can be reached at [email protected]. Stephen Stein is a senior research specialist at the same company. He can be reached at [email protected].