Shelf life automatically set to 6 months when no device performance shelf life data provide

[zx714]

I am preparing the design verification plan. The device is for first-in-human submission, and someone tells me that we don't need to do performance shelf life test if we claim the shelf life is 6 months. He says that for the FDA if no performance shelf life data is provided, the shelf life is automatically set to 6 months. That sounds strange to me. Has anyone had previous experience to prove this "practice" to be right or wrong?

 

[planB]

Never heard of that (yet). Would you have a source for this "practice"? How would you define a "performance shelf life test"?

 

[zx714]

Never heard of that (yet). Would you have a source for this "practice"? How would you define a "performance shelf life test"?

Thank you for the reply. It was said that it is from a consultant for FDA submission.
For performance shelf life test, this is the tests to demonstrate that the device still performs as intended after aging. We are going to look at the design verifications related to the product specifications and see if the performance is likely to be impacted by aging (for example, tensil strength of polymer material). If yes, we'll do the accelerated and real-time aging for that test. Is that a correct way to identify performance shelf life test? There is another idea, is to look at the design verification tests and only choose the "worst case tests" to do the aging test. However, I am not sure how we can say one test is a worst case while another is not.

 

[planB]

There is an old, still effective FDA guidance saying in the introduction:

To determine if a particular device requires a shelf life and assign an expiration date, there are a number of different parameters that must be considered. The device must be analyzed to determine if it is susceptible to degradation that would lead to functional failure and the level of risk that the failure would present.

Thus, you would perform shelf life testing only if you had indication that critical device characteristics were prone to degradation. In case you had enough (literature) data at hand that your device performance is likely not adversely affected over time and actually stable, you may justify to not specifically test for aging effects.

HTH,

 

[zx714]

There is an old, still effective saying in the introduction:

Thus, you would perform shelf life testing only if you had indication that critical device characteristics were prone to degradation. In case you had enough (literature) data at hand that your device performance is likely not adversely affected over time and actually stable, you may justify to not specifically test for aging effects.

HTH,

That is helpful. Thank you.

 

[Ed Panek]

Your claim is 6 months? Can you support that claim? You arent claiming it but if you put it on the label, its a claim