When does design transfer take place?

J

Julie O

The FDA allows use of equivalents for purposes such as design validation, suggesting that the untis don't have to be identical to the "final product".

I think the essence of the situation is reflected in the fact that we all have to keep putting things in quotations.

The only objective definitions of "prototype" and "final product" I know are that a prototype is not made with the same process that will be used to make the final product. The fact that the FDA will allow you to do something doesn't mean it's a good idea. In its long history, the FDA has never marketed a single medical device. It has never been responsible for complying with any of its own regulations. It has never had to turn a profit. It has never been litigated into bankruptcy.

The extent to which the prototype and the processes use to make it differ from the product that you put on the market and the processes used to manufacture it is the extent to which the data you collected using a prototype may not be applicable to the marketed product. The FDA will not have to do the work that needs to be done to show that the data are still applicable. If this work is not done, or not done to their satisfaction, they just get to ding you and walk away. If it turns out that...oops!...you can't show that the data collected with the prototype are applicable to the final product, the FDA will not have to bear the cost of re-doing the testing with the final product. If there is a post-market problem with the product because the data you relied on to validate its safety and effectiveness were applicable to the prototype, but not to the marketed product, the FDA can just ding you for that and walk away. So, easy for them to say.
 
Last edited by a moderator:

Ronen E

Problem Solver
Moderator
The only objective definitions of "prototype" and "final product" I know are that a prototype is not made with the same process that will be used to make the final product.

The tricky question is where exactly the process stops being "the same as". If the process includes molding a plastic part, and it's done on a different injection molding machine of the exact same model and same age etc., using the exact same process settings and the same tool, is it "a different process"...? Maybe it is, maybe it isn't. What about when the machine is a bit older, or not exactly the same model? You see, it's not a clean cut. We're definitely in the grey zone here. That's why equivalence is allowed and most times it has to be adequately reasoned.

If there is a post-market problem with the product because the data you relied on to validate its safety and effectiveness were applicable to the prototype, but not to the marketed product, the FDA can just ding you for that and walk away.

While this is true, postmarket surveillance and premarket clearance are two different things. A FDA determination that the test-units are equivalent to the commercial production ones might not save a manufacturer who later has a failure on market, but it can make a big difference in the premarket stage (and that might make a big difference for a small start-up company, for example). A wise manufacturer would weigh the significance of any difference and might repeat some of the testing once production has stabilized in its commercial form, but that doesn't diminish the importance and benefit of being able to progress quicker on earlier stages.

:topic:

The fact that the FDA will allow you to do something doesn't mean it's a good idea.

The thing with regulatory compliance is that it's not about whether a certain requirement / concession is "a good idea" or not. If we want to be in this game we have to accept the authorities' power and discretion. Of course, if we believe that a certain requirement / concession is a "bad idea", we can try to debate, appeal and use any available route to change the regulation or its interpretation / implementation.

In its long history, the FDA has never marketed a single medical device. It has never been responsible for complying with any of its own regulations. It has never had to turn a profit. It has never been litigated into bankruptcy.

This is irrelevant. Regulatory systems derive their governance from the legal system. Besides, the FDA takes input from industry in many ways; for instance, some FDA staff are industry veterans.
 
J

Julie O

The tricky question is where exactly the process stops being "the same as".

Not really my area of expertise, but isn't this why you revalidate the process when you make these types of changes to it?
 
J

Julie O

A FDA determination that the test-units are equivalent to the commercial production ones might not save a manufacturer who later has a failure on market, but it can make a big difference in the premarket stage.

An FDA determination that the test-units are NOT equivalent to the commercial ones can make a big difference in the pre-market stage too. And not in a good way.

So can the burning of resources documenting and justifying all the changes made from prototype to marketed product, in order to convince the FDA that they are equivalent.

Use of prototypes for testing does not get a startup to market faster, in my experience. You must have either data collected with the marketed product, or data collected on prototypes, plus a stack of documentation intended to connect the two. Time and money, at best six of one half dozen of the other, only higher risk with the prototype.
 
J

Julie O

The thing with regulatory compliance is that it's not about whether a certain requirement / concession is "a good idea" or not.

The thing about regulatory compliance is that it's not about what FDA allows, but what FDA requires. You are confusing the two. The fact that FDA allows something, such as testing with prototypes instead of production units, does not mean that FDA requires it. FDA also allows testing with production units instead of prototypes.
 
J

Julie O

This is irrelevant.

From a regulatory compliance perspective, perhaps. From a regulated business perspective, it is the most relevant thing there is.

The fact that some individuals employed by FDA have prior industry experience is irrelevant. FDA is a government agency that has no skin in the profit game, nor is it FDA's job to provide guidance to companies to help them make profitable business decisions.
 

Ronen E

Problem Solver
Moderator
Not really my area of expertise, but isn't this why you revalidate the process when you make these types of changes to it?

It all boils down to the same thing. You revalidate when the change is such that the process can't be considered "the same" any more. In many cases it's possible to successfully argue that when a tool is moved from one IM machine to another, the process stays essentially the same. But not always...
 

Ronen E

Problem Solver
Moderator
An FDA determination that the test-units are NOT equivalent to the commercial ones can make a big difference in the pre-market stage too. And not in a good way.

So can the burning of resources documenting and justifying all the changes made from prototype to marketed product, in order to convince the FDA that they are equivalent.

Use of prototypes for testing does not get a startup to market faster, in my experience. You must have either data collected with the marketed product, or data collected on prototypes, plus a stack of documentation intended to connect the two. Time and money, at best six of one half dozen of the other, only higher risk with the prototype.

I wasn't referring to prototype testing. I was referring to units that are mostly similar, but not identical, to the full-scale commercial production. It's not the same thing. In some cases, e.g. with mass-produced disposable plastic parts that end up being made in millions, it's pretty common to test run and validate the device on a small-scale pilot line. In some companies this is still under the domain of R&D / design engineering.

When the gap is relatively small, it doesn't take such a big effort to justify and document, or get the FDA to agree about the equivalence.
 

Ronen E

Problem Solver
Moderator
The thing about regulatory compliance is that it's not about what FDA allows, but what FDA requires. You are confusing the two. The fact that FDA allows something, such as testing with prototypes instead of production units, does not mean that FDA requires it. FDA also allows testing with production units instead of prototypes.

I am assuming that when the FDA allows something easier, most manufacturers would take that option. This has been my experience in 99.9% of the cases I've seen. Of course, there's no obligation to take the easy way.
 

Ronen E

Problem Solver
Moderator
From a regulatory compliance perspective, perhaps. From a regulated business perspective, it is the most relevant thing there is.

My comments only relate to the regulatory perspective. I'm not a business expert.

The fact that some individuals employed by FDA have prior industry experience is irrelevant. FDA is a government agency that has no skin in the profit game, nor is it FDA's job to provide guidance to companies to help them make profitable business decisions.

In principle it's true, but if someone has been in the industry for 20 years and then crossed the lines, those 20 years don't just vanish overnight. More likely, that experience plays a very important role in everything the person does later on (especially if they started fresh in industry and "grew up" there). No matter how impartial and objective the person is, that perspective will always be there, and while it's not their role to guide businesses towards profit, they will be aware of the hardships and consequences of burdensome, unfair regulatory decisions. I don't think that's a bad thing.

People's backgrounds, personalities and biases are THE MOST relevant thing in any aspect of the industry, IMO. This is something I've only come to realize over the years.
 
Top Bottom