Minimu resolution in HPLC

[DCL11]

I am developing a method to separate an API and 10 related impurities. The resolution is higher than 2.0 for all pairs except for one pair of impurities, which both elute right after API and the resolution is about 1.0~1.2. I tried to improve the resolution of this critical pair. However, I can only do so by sacrificing the resolution of API and the impurity right after it, which is not a favorable choice.

I read from a textbook that the resolution can be relaxed to ~1.0 if the components to be separated are more than 10. However, my method is intended to be used for a drug product. Is the above rule still applicable?Is there any official guidelines on this topic from oranizations such as ICH, FDA or USP?

Thank you in advance for your help!

[tom jupille]

Sorry, but I'm not aware of any such rule.

[DR]

I would hazard a guess that the 'rule' is about the ratio of peak areas. If pk1 is 10x pk2, a <1.5 resolution may be acceptable. Basically, if you are comfortable with the separation and think that you can come up with a defensible, reliable, repeatable integration for the peaks, go for it. If you run the chromatograms past multiple analysts and get all sorts of different integrations, you may need to continue to work on the separation. [/personal opinion]

[Rob Burgess]

Encountering these kind of resolution problems for impurities of drug products before I would adopt the following approach.

- It is essentially more repeatable, for integration puposes, to obtain good Rs between your main peak and any interfering impurities. I would say for peaks that differ in height by more than 500-1 (main peak to impurity) a Rs of > 4 is what you are after to obtain consistent integration. That way you won't over-estimate your impurities.
- Once that is achieved as long as you have good Rs of at least 1 between your impurities reasonable accurate and precise quantification can be made.
- I suppose it does depend on whether the imps you have a Rs problem with are regular occurences and do they occur at > 0.05% level on a % area/area basis.