When verification becomes validation

[jayroseville58]

When can you say that a USP method supposedly to be verified will be validated? What changes on the method can simply say that? Can a filter change made me decide that instead of verification we will do validation instead?

[tom jupille]

Unless you run a USP method exactly a written, you have either adjusted or modified the conditions. If the change falls within the limits set out in Section 621 (this is for USP 34, 2011), then it's an adjustment and you don't need to do anything so long as you meet system suitability. Any change greater than (or not mentioned), is a modification and requires verification or validation. This is where it gets to be a gray area.

I would argue that a change to a different type of filter in the sample prep would not require full revalidation because it is unlikely to affect things like robustness or repeatability. Areas of concern would be the result of the new filters adding or removing things from your sample. So:
- accuracy / recovery
- linearity
- detectability (LOD / LOQ)
- reproducibility.

[Alita]

Dear Tom, I would like to know your opinion about the following situation:
In our lab we have validated a method for the determination of b-aGonists in urine, liver and animal feed with one model of equipment (Agilent 6410 upgrated) and now we adquire a new model (6460). The problem is that the retentions times moved because the type of conections are different between both equipments, and its possible that the ramp program should change, because in the new equipment all the analytes elute earlier, in that case, we should make an entire new revalidation? We need to migrate to the new model because the old one is not sensitive enough anymore for the agonists in liver and feed samples because the limits are very low.

Thank you very much
Best regards
Alita