I just received a question that looked easy at first, but the laws and guidelines are surprisingly vague on this topic.
The question is: how to choose between retest period and shelf-life for an API (or intermediate or starting/purchased materials)?
Let's do our usual exercise first: what are the definitions of these terms, let our source be ICH Q1A (Glossary section).
'Shelf life (also referred to as expiration dating period)
The time period during which a drug product is expected to remain within the approved shelf life specification, provided that it is stored under the conditions defined on the container label.'
The period of time during which the drug substance is expected to remain within its specification and, therefore, can be used in the manufacture of a given drug product, provided that the drug substance has been stored under the defined conditions. [...]'
Though, immediately after the second definition, the document adds some clarification on how to apply re-test period:
'After this period, a batch of drug substance destined for use in the manufacture of a drug product should be re-tested for compliance with the specification and then used immediately. A batch of drug substance can be re-tested multiple times and a different portion of the batch used after each re-test, as long as it continues to comply with the specification. For most biotechnological/biological substances known to be labile, it is more appropriate to establish a shelf life than a re-test period. The same may be true for certain antibiotics.'
Let's try to digest the differences, and how to choose between them.
If the shelf-life generated expiration date arrives, the product is unsalvageable, it's to be destroyed. When can we use it (above the examples given by ICH Q1)? When you exactly know when your product will get OOS.
But in case of using retest period, there is a way to somehow save the product. ICH Q1 says a way how to apply this approach.
After the retest date, you should perform the necessary tests just before the planned using of the product. Based on the test results, you issue a new CoA and you give the product a month as a retest period. As long as the product fulfills the specifications, you can use it.
I also witnessed a different approach that is also compliant in my mind. Based on the stability results AND the result performed after the retest date, we can make a statistics-based decision to add more time (1-2 years even) for the next retest. Of course, you have to have the serious knowledge that your product will healthily survive that time period.
There is also a way in the product lifecycle when you can switch between the two approaches:
first you have the 3 validation batches, and (based on either R&D data, or any privious knowledge on stability) you dare to put a 12 month retest period first.
Then after having the 6-month accelerated stability results, and the 12-month long term stability results, you can see that you can freely define the 24-month for the retest period.
And after 24 months, if you see that the product won't survive the 36-month period, you can define the 24 months for expiration date.