Some companies experience problems getting FDA Regulatory Clearance using the 510(k) Pre-market Notification process. As a Product Development company, StarFish Medical’s focus is more on regulatory compliance than strategy. We believe it is important to consider Regulatory all the way through the development process, and not try to ‘bolt on’ compliance at the end. Here are tips and advice developed over the years for efficient 510(k) submission.
Clearance is required before legally marketing a medical device product for most medium risk Class II, high risk Class III, and unclassified devices. A Low Risk, Class I device, like an Elastic Bandage (product code FQM), does not require Regulatory Submission and does not need to follow FDA Good Manufacturing Practice Guidelines.
The majority of Class II devices are cleared using the Pre-Market Notification process, commonly called 510(k) after the Federal Food, Drug and Cosmetic act reference number. Here, the submission compares their device with already legally marketed devices, and claims it is ‘Substantially Equivalent’ in terms of technology and indications for use. The device developer submits an application containing appropriate design information, testing standards and verification/validation evidence late in the device development program, and the FDA clears the device for sale based on the information provided. The cost is low (less than $5,000) and the timescales are short – the FDA promise a 90 day turnaround time, but this clock stops or resets if they need more information.
Class III, high risk devices (generally life sustaining or life critical) require Pre-Market Approval (PMA), where the FDA is involved at a very early stage in the development process, with an agreed Product Development Protocol that includes clinical validation data requirements. Class III devices also face the ongoing burden of maintaining the application, which must be re-reviewed by the FDA for even minor changes to the device. Due to the cost of PMA submissions ($248,000, waived for first submission and reduced to $62,000 for Small Business exemption) and the high regulatory burden, most companies develop Class II devices. For instance, changing a typo on the Instruction Manual (IFU) for a Class III device costs more than the entire Class II device 510(k) application fee.
Note there are some exceptions, such as certain Class I devices which still require 510(k) submission, like total Cholesterol blood test, and some Class II devices exempt from 510(k) requirement, such as pure tone audiometers. Finally, some class III devices are being downgraded, but this is not a rapid process.
Given the status quo, here are five tips for companies developing Medical or In vitro devices:-
1) Have adequate resources for Strategic Planning, both as part of development and launch. The companies which have most problems are those who just seek a ‘Rubber Stamp’ from the FDA. Companies who have a dedicated person, such as VP of Regulatory, are able to work with FDA to a much greater extent, such as understanding strategic pathways which helps resolve predicate device suitability. The FDA’s paradigm is “Safety and Effectiveness”, and they do not care if your business makes money or takes longer to get to market. This is important to consider. Saying “But that’s not fair, we are only a small company” does not affect the FDA’s decision making process.
2) Plan your Regulatory Pathway. The key here is the indication for use statement, which directly defines the required validation effort. There are several instances of companies who file a 510(k) with far too many claims – for example, diagnoses XXYYZZ, cures flat feet, bad breath, baldness etc. A more measured approach is to file a first 510(k) containing restricted indications for use, with the intent to demonstrate the device is safe at an adequate level of effectiveness for your desired, limited indication for use claim. Extra indications for use claims can be added later with a second 510(k) submission. When you have appropriate clinical data – your previously cleared 510(k) is now your predicate. An example is a new diagnostic device: the first 510(k) demonstrates safety and that the device is a tool or adjunct to be used in the clinical decision making process. The later 510(k) may say the device has diagnostic performance with Sensitivity, Specificity, Predictive values etc.
3) Follow the FDA Guidance Documents. The FDA issues “Guidance documents” for certain devices or processes. This is a euphemism: it actually means you should have an extremely good, justifiable reason for not following their guidance. For example, the “Guidance for the Content of Premarket Submissions for Software Contained in Medical Devices” causes software development clients particular angst. Generally, they have developed software using ‘regular’ techniques, such as AGILE software development, and need help in ‘backfilling’. I believe the AGILE approach is fundamentally incompatible with medical device software. AGILE development implies specs essentially change on a daily basis, whereas the FDA wants to see a whole software development plan, with formal inputs and formal outputs.
4) Follow Recognized Standards. The intent of standards is to irrefutably demonstrate performance, safety or both. With the emergence of “App” software, clients have been submitting ‘Devices’ based on a combined system of Tablet PCs running an App, and the FDA wants to see evidence the system is safe. A possible better strategy would be to submit a software only submission. For complete electrical based systems, the following standards are extremely important (although there are many others, particularly device specific):
- IEC60601-1 is required to demonstrate basic safety and essential performance of electrical medical equipment. If your Tablet PC is being used as a Medical Device, it must comply with all the associated Risk Management documentation required for third edition and have a test report demonstrating compliance to this standard. We have seen instances of 510(k) applications being denied by the FDA on this basis – the argument the Tablet PC is IEC60950 compliant appears insufficient. I posted tips for IEC60601-1 compliance in an earlier blog.
- IEC60601-1-2 is required to demonstrate Electro-Magnetic Compatibility (i.e. the device will not cause electrical interference, be interfered by or be affected by static discharge). Again, we have seen instances whereby the 510(k) application was denied without evidence of testing to this standard. A colleague posted tips for EMC compliance in an earlier blog.
- ISO10993 is required to demonstrate Biocompatibility, especially important for patient applied parts. The preferred route is to ensure all your materials are ISO10993 traceable with a readily available test report. Despite this, testing may still need to be repeated to confirm that material (injection mold houses) and/or sterilization processing does not introduce contamination which is non-biocompatible. However, using 10993 pre-certified materials does increase the likelihood these tests will pass. A colleague posted more on this in an earlier blog.
5) Watch out for the new, improved De-Novo process, which provides a route to market for innovative medical devices that are low to moderate risk, but do not have existing predicates. The fundamental modus-operandi of the 510(k) is to demonstrate your device is Substantially Equivalent to an already legally marketed device. This process can be in direct conflict with innovation and technology progress development.
For a while, 510(k)s were being submitted and cleared with several predicate devices referenced. But the actual predicate for Substantial Equivalence was a hypothetical mix of desired functionality from each of the predicates – the so called “Split Predicate”. This “À la carte” menu approach was specifically identified as risky in the CDRH 510(k) Working Group preliminary report published in 2010. To address this, the working group proposed streamlining the De Novo process introduced in 1997.
Presently, the De Novo process is effectively an appeal against the automatic Class III designation issued if your device ‘fails’ the 510(k) and is found not-substantially equivalent to a predicate device. (A declined 510(k) due to lack of performance data can currently be resubmitted.) So if you knew your exciting, novel device was low to medium risk but did not have a suitable predicate, you would still have to submit a 510(k) with the knowledge it would ‘fail’ and become Class III, then request De Novo reclassification to Class II.
Even the best expectation management skills would make investors nervous with the business risk of this approach, which explains why there were only about 50 De Novo clearances from 1998 to 2009. Draft guidance is available on the proposed streamlined De Novo process, published in October 2011, and an Amendment to Section 513(f) (which covers De Novo) of the Food, Drug and Cosmetic Act was made in July 2012. The FDA website states that the new De Novo Guidance is still being finalized, but it is likely that a Pre De Novo Submission (PDS) system will be put in place, which will hopefully avoid the scary “Fail a 510(k)” currently required. Stay tuned!
An often used quote, “By failing to prepare, you are preparing to fail.”, originally by Ben Franklin, is particularly relevant here. The No 1 tip on making resource available may not be practical for all companies, so discussing your case with experienced Regulatory Consultants earlier in your development process, rather than at the end, is essential. A key component to this is to talk with FDA before investing heavily in tooling or final detailed design. To that end, and another of the CDRH 510(k) Working Group’s recommendations was to facilitate high quality 510(k) submissions (especially for novel devices). The FDA is developing a Pre-Submission Program to allow easier engagement. My personal experience meeting the FDA CDRH in Baltimore was that the FDA do not give advice (sometimes known as free consultancy). However, they will agree, disagree or modify the strategy you present, hence being adequately prepared is essential.
Vincent Crabtree, PhD is a Regulatory Advisor and Project Manager at StarFish Medical. If you have other tips for Medical Device 510(k) submissions, he would be delighted to hear them.