
Sign up to save your podcasts
Or


Decrease Time-to-Market by Developing Medical Device Software “In the Spirit” of IEC-62304
Getting any new medical device to market is a race against time, and it’s so tempting to jump in and start coding from the get-go. But hold your horses – you’ll save time in the long run, and potentially improve the valuation for your company, by following the spirit of IEC-62304 throughout each stage of development.
This standard impacts the entire software development lifecycle: from initial requirements and coding to release and maintenance. Understanding when and how IEC-62304 applies can potentially save you months of backtracking in the race to get your device to market.
Check out this video as Computer Engineer Jamie Kendall and Andy Rogers lay it all out and explain how “working in the spirit” of the standard can keep you on track, on time, and in compliance, particularly when developing non-regulated devices.
Need to know:
The nitty-gritty:
IEC-62304 lays out a unified process for evaluating safety in medical devices sold in the US and EU. The software safety classification, Class A, B, or C determines the safety-related processes you’ll need to use.
With non-regulated medical devices, you’re working within a highly-structured environment, without the formalized legislative pressure. Choosing to focus on specific areas of IEC-62304, even without the requirement, balances the key factors of time and risk management. And when you understand precisely how the regulations apply, you can stay in compliance without going for all-out unit testing at each stage of development.
It’s important to identify interfaces from the very beginning. Think through what goes where: for instance, what should be firmware and what should be PC-based software. You can use pre-engineered software or existing code/ libraries for things that are not unique to the system (stepper motors, pump etc.) so you’re not reinventing the wheel at each stage.
For example, using a single micro-controller on your platform gives you a good reusable framework for low/mid-level modules, especially when these are designed to IEC-62304. In addition, you can use a methodology like AGILE project management, tools like ReSharper, or static analysis techniques to help you maintain quality and compliance.
Keeping within “the spirit” of 62304 at each stage of development allows you to focus on key parts of the system and make needed changes in days versus weeks. By adding value to your device in this way, you’re building value for your entire company.
Helpful links
https://www.iso.org/standard/38421.html
Focus on The Assay to Improve Overall IVD System Development
Assay performance should be the priority when developing complex cartridge and instrument systems. The Assay is King! As assays are becoming more complex, taking an “inside-out” approach to IVD System Development is optimal for project success.
This approach starts with first defining what the assay needs to do, then figuring out how to automate that process on the cartridge, and only then focusing on what a commercially viable instrument will look like. This leads to reduced costs, shortened timelines, and superior diagnostic performance. Put another way, the assay team should be the customer when architecting and developing IVD Systems—the chemistry should drive the project.
How to implement the inside-out approach
Define the Assay First
The assay is where the magic happens and is where the intellectual property lies. That IP is what drives future acquisitions by global diagnostic players. For example, Roche recently acquired GenMark along with the ePlex system that Key Tech helped develop, to broaden its molecular testing portfolio.
At the end of the day, the instrument serves the cartridge and the cartridge serves the assay. Before starting any product design work, you need to understand whether the chemistry is locked down. Is the benchtop assay working? If not, it may make sense to wait until it is, to avoid costly rework later.
Our CEO and Co-Founder, Jenny Regan, stresses the importance of solidifying assay performance before instrument design:
“Anyone can design a beautiful instrument that doesn’t work. It’s easy to do because it is a tangible milestone for investors and developers alike. The pressure to do that and make something that looks great before doing the nitty-gritty, difficult work of making a working assay is all too easy to fall into.”
Collaborate on Menu Planning Early
It is critical to not only understand the lead assay, but also what other assays are being planned so that both cartridge and instrument can be planned accordingly. Early collaboration with the assay team will ensure that fewer changes and post-market updates need to be made. This approach also affects speed of regulatory approval, since most changes that need to be made to the cartridge or instrument, to support a new assay, will require applying for re-approval.
For example, consider reagent quantity. If your lead assay requires eight reagents, but a few others need ten onboard, you need to ensure that the planned cartridge design takes that extra requirement into account and leaves space for the additional two.
Sample type is another consideration. If the lead assay is going to work with a saliva sample, but a future panel will use a stool or blood sample, you need to ensure the cartridge is designed to be able to handle the future, additional sample types.
De-Risk and Validate Assay Automation Early
Because a working assay is the end goal, it’s critical to get working automated cartridge/instrument prototypes to the assay team early so they can validate that the chemistry is still working, and to design a workflow for their assay. It needs to be easy for the assay team to collect data and interact with these early prototypes.
Often, this means developing test fixtures for each critical step in the assay process. When translating an assay from the bench to the cartridge, there could be steps that only work part of the time on the bench. These must be confirmed to work 100% of the time on the cartridge. Or there could be steps that always work on the bench, but are extremely complex to automate, requiring validation to see if that specific step is even practical in an automated instrument.
At Key Tech, we have a proprietary software platform for hardware interaction, called KeySharp, that allows the assay team to control a prototype instrument while seeing a graphical engineering view of what the cartridge and instrument are doing. With this tool, they have a way to interact with the prototype, adjusting and testing variables, and designing a workflow for their assay.
Constant Collaboration for Real-Time Feedback
The “assay is king” approach requires an ongoing conversation between the engineering team and the assay team, and a consistent feedback loop.
The traditional outside-in approach often creates unnecessary constraints on the assay team by forcing them to operate within an engineered product that was developed without them in mind. If the assay team has tweaked the bench assay to generate a better or quicker result, the design may not be able to support it.
Pivots and updates in the bench assay are commonplace, so constant communication is the only way to anticipate and adapt the development effort in a timely way.
Parallel Development of Cartridge and Instrument
Oftentimes, three teams are working on a project – assay team, cartridge team, and instrument team – which makes efficient development exponentially more difficult. Especially if coordinating across multiple organizations.
On the other hand, parallel cartridge and instrument development by a single interdisciplinary team eliminates conflicting priorities and incentives between vendors. This integrated approach helps your IVD system get to market faster, as both instrument and cartridge are developed under one roof toward a common goal. Ultimately serving the assay, designing the cartridge and instrument in parallel reduces your overall development process time and cost.
Not all design and development firms offer both IVD cartridge and instrument design expertise to support a successful assay. Ben Lane, our Director of Engineering, shares,
“Our claim of integration is truly the case. We have cartridge designers and instrument designers who are literally sitting next to each other and talking to each other all day long.”
In Summary
Assays are becoming more and more complex, and ensuring that platform development actually serves the assay is critical. Placing your assay and engineering teams in constant collaboration and tackling de-risking, menu planning, and validation early in the process will lead to project success.
The market for medical devices for home use is projected to grow more than 20% to over $7 billion by 2027, driven by increased awareness of the importance of self-monitoring, an increase in lifestyle-driven diseases such as diabetes, and – in no small part– infectious diseases such as COVID.
But the home environment is a far cry from the usual clinical setting, and it presents unique challenges in product design. Key Tech’s Rachael Scott, Sr. Engineer, and Andy Rogers, VP of Business Development, talk about the challenges and provide valuable insights.
What you need to know:
The nitty-gritty:
Begin with a good understanding of the home environment. These considerations affect your choices in electrical safety, screen size, GUI design, portability, and durability. Expect the unexpected; untrained users, children, and pets are just a few wild cards that should figure in your design. Make sure you have a way to offload and disseminate results in a user-friendly way. Interfaces are always critical, especially where HIPAA compliance is required.. Above all, get your prototypes into a home environment early in the design stages, and give them this reality check frequently.
Helpful links:
https://www.fda.gov/media/84830/download
Understanding Risks, Tradeoffs, and Prioritization in Your Next Automated Assay Development Program:
1. Product Strategy
Working backwards from the commercial goal, the purpose is to understand key drivers of product architecture and identify technical risk. Driving decision-making throughout development: intended price target and use case.
2. Ease of Use
Identify which areas of the eventual product are most valuable to automate, and which are better left as manual user steps. Understand the risks and tradeoffs between the complexity of automation and the likelihood of human error.
3. Sensitivity Analysis
Break down assay steps and establish clear technical requirements for each step, as well as their interdependencies. Allow the engineering team to know what variables need further study to accurately assess technical risk.
4. Consumable/Durable Interface
Focus on device functionality and where features will reside in the system: what belongs on the cartridge and what belongs on the instrument to achieve performance and system requirements?
5. Reagent Handling
Consider how reagents will be introduced to the assay: through durable containers at large runs or on the consumable in small doses. Reagent handling dictates high-level architecture decisions, best addressed early for optimal product architecture.
6. Fluidic Control
Precise liquid handling is a core challenge in assay automation. Identify how liquids, and typically air, will be controlled in a reliable, repeatable way, consider automation solutions for fluidics control, and address contamination sensitivity.
7. Temperature Management
Develop methods for measuring and controlling temperature by understanding the effects on the instrument, cartridge, and assay performance.
8. Algorithm Development
The data processing algorithm is often the core value proposition and differentiator being brought to market. Two factors ensure a clinically actionable test result: understanding the limits and constraints of the algorithm and recognizing the instrument requirements that the algorithm demands.
9. Tech De-Risking Methods
Identify what risks need to be mitigated and where various de-risking methods should be deployed. Methodology options include creating testbeds, modeling and simulation, and subcomponent specialist consulting.
10. Scale
The framework bookend considers the impacts of high quantities and long-term use on the instrument and cartridge reliability, cost, and performance. Understand whether the device can work in large numbers and what strategies should be employed to improve reliability and performance while containing cost.
From the publisher's feed