validating HPLC method for impurity analysis

[solotallboy]

Hello Guys,
My question concerns the validation of the HPLC method for impurity analysis for degradation experiments (acid, base or thermal). I performed these studies and so far no degradation product was obtained. While this is good from a stability point of view, i don't know how to validate the hplc method for accuracy. Since the accuracy is pretty much a recovery study (how well i can recover a known amount of my API in the presence of impurity). So how do i go about validation if i have no impurity? I got the idea from my reading that I could use the API as an impurity but not sure as how much to add? Any thoughts on this would be much appreciated.

[tom jupille]

the fact that you are seeing *nothing* raises questions. Unless you're dealing with something like Freon*, it has to degrade to *something*. The issue is really one of LOQ in the presence of a huge excess of the API, so you might want to set up a "straw man" by spiking with a known potential degradant of the API (get a tame organic chemist to look at the structure for candidates).

* yes, I know there are other stable molecules out there!

[solotallboy]

Thank you for your reply. What about using my API as my impurity.

So i have a large concentration of standard say about 1000 ug/mL and then spike the sample with my limit of impurity for the standard (which should be about 0.1% which works out to be 1ug/mL). Then prepare three samples in the range of 50 (1000.5 ug/mL), 100 (1001 ug/mL), and 150% (1001.5 ug/mL) ie the 1000 ug/ mL sample was spiked with .5, 1 and 1.5 ug/mL of the API acting as the impurity.

The problem with this is the high concentration of API and the calibration curve set up would have to be in the range of 999 ug/mL to 1004 ug/mL. I would then have to use this calibration curve to determine all my stress experiments (acid, base, light and humidity). This sounds like a lot of sample. Please let me know what you think of this method, thank you.

[solotallboy]

The molecule is quite stable under acidic and basic conditions in the range of ph2-9. It degrades in the presence of peroxide solution (about 30% degradation occurs). The degradants are not identifiable at this point, but i still need to validate the method. The sample is quite stable at the different humidity (60-75%) and at 25 and 30 degree cel. The study has only been ongoing for 3 mths.

[Consumer Products Guy]

Don't you just need to show that any peaks formed during your peroxide degradation test do not interfere with your API? Or does my company QA interpret that incorrectly?

[Karen01]

the fact that you are seeing *nothing* raises questions. Unless you're dealing with something like Freon*, it has to degrade to *something*. The issue is really one of LOQ in the presence of a huge excess of the API, so you might want to set up a "straw man" by spiking with a known potential degradant of the API (get a tame organic chemist to look at the structure for candidates).

* yes, I know there are other stable molecules out there!

There was drug i worked with that under typical forced degradation conditions simply did not degrade. Sometimes you have to hit a molecule a lot harder than usual for non obvious reasons.

- karen

[Juan2011]

You try more extreme conditons of degradation by changing the concentration of the avids base peroxides etc but if the molecule is really stable , you just need to report the outcome. It is never necessary to degrade the sample always. You are just validating the method and not told you to get the deagradants forcefully. If your molecule is very stable, whats the problem:) In fact, that is really good method and molecule. You can inject the sample after one or two days also in case your system break down between run

[Murugan Saravanan]

Forced deg. studies are useful to check the ability of the method to detect the degradants which is a part of the specificity expt.
In case when no degradation occurs, pl. try different concentrations, volumes, extended times, refluxing with acid, alkali and peroxide.
You also should use the KSM's as well as intermediates of the API process(which are process related impurities) to develop and validate the method for the purposes of capability of the method with respect to LOD, LOQ, Accuracy.
Hope it helps in understanding
Regards
Murugan Saravanan

[solotallboy]

Thank you all for your responses. The problem is not trying to obtain degradants, as I said the study is still in the early phase and yes increasing the various conditions will yield decomposition. The task at hand is to validate the current samples that I have ie samples that did not yield any degradants. I read that I could use the API at the limit of impurity (0.1%) with my stock solution to act as a degradant. So the question is, did anyone ever had a situation where they had to do this?

[parsely10]

We validate our impurities methods this way all the time if we don't have impurities available. The accuracy range should be based on the spec. of impurities and ICH guideline.

[solotallboy]

parsley 10 that sounds great!! I have a couple of questions i need to ask. is that possible? Please let me know and if so please email me at solotallboy@hotmail.com please and thanks in advance.

[Rob Burgess]

At the early stage of API drug development we would say that accuracy is implied by specifcity, linearity and repatability for API. At this stage most forced degraded impurities will be unknown - so accuracy checks are not possible as you say.

Only when you get to later devlopment when you might have isolated potential degradation prodcuts and thus could do spiking/ accuracy (recivery) checks.