Verification of Compendial methods

[michaelcarolus]

Hi all

When verifying compendial methods, for finished products, what performance charateristics would you perform for the assay? I was thinking of doing Specificity, Accuracy, Method precision and Stability of solution. Are these tests sufficient? When doing the specificity would you stress your samples in acid/base/peroxide/light etc. I was thinking of preparing a solution with my API and all degradation products/impurities to ensure that the API is sepearated from degs/imps. Is this sufficient or should I also include acid/base/peroxide/light stress samples.
thanks
Mike

[Consumer Products Guy]

I wouldn't think that you need to do the degradation part of specificity - that should've been part of the original compendial method. You wouldn't be trying here to demonstrate whether or not your API degraded anyway - just that any degradation components don't interfere with the API itself - and I believe that's already been done. I would just write that justification into the report or file.

[adeputy]

Compendial methods are not necessarily stability indicating. Both the USP and the FDA say that compendial methods should be evaluated as to whether they are suitable for the intended use.

[Consumer Products Guy]

And many compendial methods still use packed columns, wet methods, and are stone age. And it seems that everything is stated in gray when dealing with FDA, USP, etc.

[krickos]

Hi my cents on the topic

Verification of Accuracy, Method precision and Stability of solution (sample and standards) seems sound to me, SSTs also tend to give some more information.

As for Specificity, well I would not by default do all that, rather first gather what availble information is there?, of course one hopes that the compendial method for finshed product is linked to the compendial product for the API but best to check.

Look for similarities on API related/stability methods versus your method, what impurities are listed? Are those relevant for your API? Ph Eur tend to list impurities and Ph Eurs knowledge database does not require an online account (http://extranet.pheur.org/publications/ ... s_sw.shtml) here you sometimes finds spiked chromatograms.
Bug your API department for information, if you do not do the API yourself, bug the QC department for rawmaterial and the QA responsible for the API supplier, QA can and should be able to supply you with the suppliers Drug Master File (DMF) open part. The open part of the DMF lists the impurities investigated and the method, likewise stress studies is typically summarised in the stability part.

Also recall that what actually ends up on column when doing finshed product analysis versus API related substances sometime differs quite much. Just checked on of our products and the finished product injection is 15 times weaker than the API related one so the interferance from the API impurities in that case are quite low but then again it is the same chromatographic systems in this case .

Hopefully you end up with an acessesment as indicated by someone else abvove, that the method is "good enogh" with regard to Specificity.

Cheers

[aceto_81]

You should perform specificity against your other ingredients of your finished product, but I wouldn't do any stress tests.
Then accuracy, precision and stability of solutions, write a nice report (with justifications for skipping linearity, specificity against related compounds) and you're done!

Ace

[michaelcarolus]

Hi All

Thanks for all the comments/suggestions. Looks like my approach is sound.
thanks again
Mike

[Sallybeetle]

If your lab is cGMP regulated, then you should use the USP Validation of Compendial Procedures monograph <1225> and ICH Q2b. The minimum would be Accuracy, Precision, Specificity, Linearity, Range, and Robustness. You would also need to do stability indication (solution stability and forced degradation). Eventhough the USP indicates that LOD / LOQ may not be necessary, depending on your product category, I have been required to submit this data anyway.

[krickos]

If your lab is cGMP regulated, then you should use the USP Validation of Compendial Procedures monograph <1225> and ICH Q2b. The minimum would be Accuracy, Precision, Specificity, Linearity, Range, and Robustness. You would also need to do stability indication (solution stability and forced degradation). Eventhough the USP indicates that LOD / LOQ may not be necessary, depending on your product category, I have been required to submit this data anyway.

Are you seriously suggesting that we per default should do a full ICH Q2b validation on all compedial methods when introducing/verifying them?


Again, I would not recommend to generalise in that way. You really need to do it case by case depending on information at hand as indicated above.

Yes FDA indicates in the draft guidance from 2000 that "information about, specificity, intermediate precision and stability of solution" should be included when submitting compedial methods.

And yes compendial assay methods may not be stability indicated (also mentioned above) and should be considered when developing the specification (also in FDA draft).
For compendial items addition analytical procedures, such as impurities....may be requested to support the quality..these additional procedure should be validated. (also in FDA draft).

USP <1225> uses a similar approach.

So again acess the information at hand, how does the specification look, is there a separate impurity procedure and so forth. After that you should have an idea of what is needed to be done

[zlb215]

Hi guys,
I have two questions about this topic:
1.what's the difference between validation of a new method and of compendial method?
2.what will be a suitable situation for USP<1225>,and what for USP<1226>?

[zlb215]

Typical validation characteristics which should be considered are:
• accuracy,
• precision,
• specificity,
• detection limit,
• limit of quantitation,
• linearity,
• range, and
• ruggedness and robustness.
• system suitability
Question is:which characteristics should be applied in the following cases:
1.verfications of compendial method
2.validation of compendial method modifications
3.validations of new method
Are there any office regulations or personal experiences which can give me an answer?
Best regars!

[krickos]

Typical validation characteristics which should be considered are:
• accuracy,
• precision,
• specificity,
• detection limit,
• limit of quantitation,
• linearity,
• range, and
• ruggedness and robustness.
• system suitability
Question is:which characteristics should be applied in the following cases:
1.verfications of compendial method
2.validation of compendial method modifications
3.validations of new method
Are there any office regulations or personal experiences which can give me an answer?
Best regars!

1. Case by case depended I would say.
A Specificity statement, intermediate precision and stability of solutions and control of relevant SSTs is a good start.

2-3. In short no big difference, might be some detail that I do not recall immediately.
However there are at least one thing under precision: For compendial method validation you typically exchange intermediate precision (within laboratory variation) for reproducibility (between laboratory variation).

[zlb215]

I read an article written by Michael E. Swartz:<Validation, Qualification, or Verification? >, published in LCGC.
I think it may give a good conclusion to this topic.
Website link:
http://chromatographyonline.findanalyti ... 972?ref=25

[krickos]

I read an article written by Michael E. Swartz:<Validation, Qualification, or Verification? >, published in LCGC.
I think it may give a good conclusion to this topic.
Website link:
http://chromatographyonline.findanalyti ... 972?ref=25

One thing to be kept in mind, though: The article is from oct 2005, after a quick glance, most likely based on the early drafts of chapter <1226> that in some cases more or less called for a complete revalidation in some drug product cases.

The final version of <1226> came out quite different in certain aspects. For the intrested I refer to Pharmacopeial Forum Vol 31(2) Mar.-Apr. 2005 page 555-558 if you like to compare.