In Reply to Parent Post by Arie0712
(...) SOP/WI to define the shelf life for material/products in our storage (...)
I don't know the particular products involved, of course. I assume there was some logical basis for your customer's concern. Usually where there's smoke, there's at least a little fire.
The FDA has guidances specific to a few product types, but in the general case for shelf life or product expiration based on some factor other than sterilization packaging stability, there is no stated guidance other than the general requirements that any claim be proven and that no device be sold without proof that it is safe and effective.
For FDA Class II and III products, of course, such proof must be provided to the FDA with the device's 510(k) or PMA, so those are pretty obvious. For devices not requiring a 510(k) or PMA, proof must still exist in the device's DMR.
Based on review of what the FDA has to say about this across various guidances and other documents, my summarization of the general rule would be:
1. If you're going to state an expiration date or shelf life for a newly developed product, it should be based on already-competed accelerated aging, and have real-time aging in progress to verify the data ASAP.
2. If the product has some mechanism or issue that gives rise to concern about shelf life, such as a material that is known to have a life limit, it cannot be sold at all without data to show that it is safe and effective at that length of time after manufacture. It's not necessary to prove when it becomes unsafe or ineffective; it's only necessary to prove that it is safe and effective throughout the period of time when it is permitted to be used.
I regard it as likely that, for a relatively simple product that contains a material for which a substantial "track record" exists and for which the risks are completely understood and clinically acceptable, the FDA will accept well-reasoned analytical application of already-existing aging data or retrospective market performance data for that material in place of newly acquired aging data. However, as far as I know no guidance exists, so any such instance would be individually judged. The more complicated the product and the greater the risk involved, the less likely would be such acceptance.
You would write your procedure for defining shelf life based on your knowledge of what material, process or other device characteristic gives rise to the concern. Similarly, you would categorize your products as to whether that procedure applied to them based on that same knowledge.
I think the same general approach is likely to be considered reasonable and acceptable in an MDD or Health Canada context, notwithstanding the different applicable rules.