Stability Community

 Hi Everyone,

Hope you can share your inputs/practice regarding end of shelf-life time point.

For small molecule stability, the study initiation date (T0) is when samples are loaded into stability chamber. The initiation date can be 2 months after manufacturing date. For a product with 36 months shelf-life, the 36-month stability time point on protocol is actually 38 months from manufacturing date. In this case, do you add a 34-month time point which corresponds to the expiration date of the lot? The concern is that without the 34-month time point, if anything is wrong for the lot at end of shelf-life, you would not know until two months later.

Thanks in advance!

NOTE:  This question was initially submitted to the AAPS Stability Community's email distribution list and shared anonymously here in the Stability Discussion Topics with the permission of the questioner.

One approach I have used, assuming that the stability samples are chambered in time, is to set the base date of the long-term stability study as the date of manufacture of the source batch / lot.  This eliminates discrepancies between the timepoint and the material's age.  This approach requires that samples have been stored at the long-term storage condition prior to being chambered on stability.  This also requires that stability studies are initiated prior to the first scheduled stability timepoint and release testing.

------------------------------
Dan Willingmyre
Vir Biotechnology, Inc.
Associate Director, Stability & Data Analytics

Disclaimer: Opinions expressed are solely my own and do not express the views or opinions of my employer.
------------------------------

Expiration Date is calculated from Manufacturing Date, defined by GMP.

Study timepoints are calculated from Study Start Date, defined in WHO Stability guideline and some others.

Therefore, for Study End Date, If you start the study 2 months after the expiration date, you have 2 choices:  

      1. End it at expiry (i.e. 34-month timepoint) - GMP requires meet specs at expiry.

      2. End it at 36-month timepoints (which is 38-mo old sample) -  Easy to batch testing and look good on the stability report or graph.

The risk of failing Option 1 or 2 would be the same, you still need to investigate it anyway, as a requirement of 21 CFR 211. 

For small molecules, if our product fails in 2 months at the end of shelf life (between 34- to 36- month timepoints), and we don't know it well enough to anticipate it, then I think we have a bigger problem than just failing stability at expiry!  You may need to redesign the package. However, for these products living at the edge, I'd add a timepoint at expiry.  (My protocol would list End of Study or Expiry.) 

There is no regulatory requirement that we must start within 30 days with regards to ICH; it has been typical 'industry practices' that release can be used for TZ if study starts within 30 calendar days.  However, this should depend on several factors (i.e., the product stability - small molecules vs biologic vs devices, the dosage forms - solids, injectable, liquid, or if samples are shipped around the world during manufacturing, etc.)

This question has been asked many times, and I have written it in Chapter 15 (Stability Operations) of the Stability Handbook and Chapter 9 (Stability Program) in the GMP Laboratory book with figure to illustrate.  You may be interested. After all, your SOP rules.  

Hope this helps, 

Kim.

NOTE:  This response was initially provided by Kim Huynhba to the AAPS Stability Community's email distribution list and shared  here in the Stability Discussion Topics with the permission of the responder.

In my experience end of shelf life is based on the stability study. In some cases 36 months may be real (i.e., the product degrades) but in many cases, that's the time the stability study was run for.  In some of my previous work, we extended shelf life of several products by opting to perform 5 year stability study and extending the shelf life.

Maybe a starting point would be to look up the product history…..

Our company includes an 'ad hoc' time point at the true expiry (in addition to the full testing plan based on date in chamber). We tried to avoid the 'ad hoc' time point and ran into difficulties with regional inspectors. Scientifically, we argue that the 'ad hoc' is not necessary because we evaluate trending after each time point and should know if a stability lot is at risk of being OOS prior to expiry. However, regulators disagreed so we take the conservative approach.

NOTE:  This response was initially provided to the AAPS Stability Community's email distribution list and shared anonymously here in the Stability Discussion Topics with the permission of the responder.