USP <621> Allowable changes

[michaelcarolus]

Hi All
We have an in-house HPLC method that has been validated as per ICH/USP. Our QC lab believes that changes to any HPLC method can be made as per USP<621>. My understanding is the USP<621> is only applicable to USP monographs. Therefore, the allowable changes can only be done to USP methods. The “design space” of an in-house method will be determined by the method validation e.g. Robustness testing.

Is my understanding of USP <621> correct or can we adjust all HPLC methods as per USP allowable changes.

Regards

Mike

[tom jupille]

Depends on your SOPs. The US FDA has a similar set of default adjustment limits in theit document ORA LAB 5.4.5, but strictly speaking that only applies to FDA labs. My suggestion would be to write an SOP based on ORA LAB 5.4.5 (in fact, copy it verbatim).

Any adjustment limits you specify in a particular method would, of course, take precedence.

[KM-USA]

Depends on your SOPs. The US FDA has a similar set of default adjustment limits in theit document ORA LAB 5.4.5, but strictly speaking that only applies to FDA labs. My suggestion would be to write an SOP based on ORA LAB 5.4.5 (in fact, copy it verbatim).

Any adjustment limits you specify in a particular method would, of course, take precedence.

Tom - I've tried to convince my supervisor that FDA ORA LAB 5.4.5 would also apply to our own in-house validated assays. But he isn't so sure....

[tom jupille]

Tom - I've tried to convince my supervisor that FDA ORA LAB 5.4.5 would also apply to our own in-house validated assays. But he isn't so sure....

The actual person who must be convinced would be an FDA auditor! That's why I would write an SOP using the same verbiage as ORA LAB 5.4.5. If any one questions the SOP, you reference the FDA document. In the absence of an SOP, though, it *is* a gray area, which I suspect is what worries your boss.

[KM-USA]

Perhaps someone here has had experience with an auditor on this topic...and could share his/her experience? I'll see if supervisor will go for the SOP idea.

Gerhardt? Consumer Products Guy?

Thanks

[aceto_81]

This is always such a grey area.

If this is written in a SOP, and the auditor confirms the SOP, then you can do whatever is in your SOP.
Else, you could write a SOP, and hope if this would by confirmed by the auditor.
But then: what to write in your SOP?

We never allow changes outside of the robustness, I think that's maybe overkill sometimes, but that way, you are sure it is within the validation scope.
Maybe other guidelines are applicable, but are you willing to take the risk?


HTH

Ace

[Peter Apps]

Hi Michael

Since the robustness of methods is highly variable I would be very cautious of allowing any set of generic changes. Some methods can tolerate all kinds of abuse, others get cranky if you feed them a different brand of solvent. Whether they are tolerant or cranky can only be shown by robustness testing, and if you have the robustness results you do not need to fall back on generic policies.

Peter

[tom jupille]

We never allow changes outside of the robustness, I think that's maybe overkill sometimes, but that way, you are sure it is within the validation scope.

I don't think that's overkill at all. Any adjustment limits specified in the method will take precedence over the default limits. The default.limits apply only in the absence of specific information. And, of course, you still must meet the method's system suitability test(s).

[michaelcarolus]

Hi

Thanks for all the responses. Unfortunately we don’t have a company SOP dealing with allowable changes for HPLC methods. However, I think we may include a table of allowable changes in the method of analysis. This will guide our QC dept on what changes can be made e.g.


Parameter Allowable change
Mobile Phase - ± 5 % MeoH
Flow Rate - ± 0.1 ml /min
Colum Temp - 20 - 30 C
pH of Buffer - ± 0.2

Regards
Mike

[krickos]

Hi

Yes, the question if USP/Ph Eur allowed changes can be used is a common one, but as mentioned it tends to become an issue within the labs, even worse if you involve QA and regulatory functions (poorly written regulatory procedures even tend to "flip" regulatory people even if you "just want to replace a no longer avaible guard column dimension").

I agreed that if you put some effort into the robustness testing during validation it will reward you sooner or later especially if it contains a clear statement that if you operate within X-Y the performance will be satisfactory.
Operating outside robustness but within USP/Ph Eur is a grey area and as stated some procedures can handle more than others.

Another aspect is to actually submitt "wisely" written company specific procedure with flexible descriptions using recommendations rather than stating that you must use this brand of XXX etcetera.

To take things even further, there are a few companies that nowdays try to submit more "Quality by design" type of procedures i.e. the procedure contains ranges for mobile phase, injection volume etcetra, still there is a recommended setting included.
This brings robustness even further and often is helped with DoEs/Chemometrics.
Unfourtunately to this day I have not heard how succesful companies has been in these efforts.