Mass balance

[mpepler]

Currently performing forced degradation studies for finished product.
Customer requires mass balance ie sum of peak areas in degraded spl chromatograms vs control to be at least 95%. Has anyone heard of a similiar specification before? I need a reference to some guideline that mentions some sort of criteria for mass balance as described above and I can find absolutely nothing. Needless to say this can be a difficult criteria to meet.
Please can anyone help - I need some kind of official reference to this.

[tom jupille]

The big hole in that logic is that it assumes that the parent(s) and degradant(s) have substantially the same absorbance spectra.
Unless that assumption holds, having the areas add up to 95% of a pure standard is virtually meaningless.

[mpepler]

I totally agree but I need to convince other parties. The assumption does actually hold surprisingly often. Also, the ICH says under specificity "...to ensure that procedures performed allow an accurate statement of the content of impurities of an analyte" Hard to do if you get say 10% degradation and the degradant spectra are different.

[HW Mueller]

What could be more convincing than Tom´s statement? Have we come to the point where only "experts´" or government officials´statements count?

[philippem]

Hi,
I agree with Tom remarks, but..
A few years ago, when I was working in a contract lab I had the same question/problem .
We solved this item by determine the relative response factors (RRF) of each known product ( impurities and degradants).
Of course if you have unknown degradants/impurities it becomes diffucult. But if they are unknown, than you can assume they are not really important. The important degradants , impurities should be known, is'nt it ?


But for our products it helped a lot and we came to acceptable mass balances

I hope this will help you

Philippe

[gtma]

I agree with all of you. Without the RRF for significant related substances, mass balance is meaningless. Regulatory bodies usually require it and therefore, I do it. The approach I generally take is to use RRF for related substances where I can, otherwise assume similar RRF.

In cases where none of the RRF are known, you can calculate total related substances at different wavelengths (3) and compare the results to obtain the appropriate mass balance (some deficiency in this approach).

It is generally expected that you stress samples to about 10% degradation (potency) and obtain a mass balance within 98% to 102%.

[adam]

Gtma

Can I ask where you get the 98-102% criteria. What do you mean, exactly, when you say "it is generally accepted".

In my view it seems far too tight, given the assumption of equivalent response factors that we all acknowledge is unavoidable in most cases.

My own thinking about handling force deg studies with a UV detector would be to:
1, Use a low wavelength to make the assumption of equivalent response factors more likely.

2, Use a fairly loose criteria; such as the sum of areas must be NLT 90% of the undegraded control (even then I'm not sure I would put this in an SOP).

[DR]

It's a lot of hogwash confounded by a catch-22.

You need mass balance, but you don't know what the extinction coefficients of your impurities are until you ID and/or make, isolate and test them vs. your main peak.

The truly stupid part is that you need not ID stuff that's <0.1 (or <0.05, depending) % of your main peak (pharmaceutical regs) BY AREA!?! What if your degradant absorbs very poorly? It could be very toxic but you need not worry about it because 1) you can't see it 2) there isn't enough to hose your mass balance. We've seen some minor impurities that absorb very strongly but we'll never be able to isolate enough to do an ID or generate a good RRF because there's hardly any there even after severe abuse of the sample in forced studies. All you can do in those situations is document your efforts and conclude that the amounts involved are trivial despite the large peaks.

[adam]

True story.

[adam]

DR

It just occured to me that - since you seem pretty familiar with the process - you may be able to answer something I have always wondered about: that is the issue of qualifying impurities.

In my experience, when an impurity reaches the qualification threshold, companies will say "it's qualified since it was there in the animal safety tests, or phase I trials". But that's kind of a useless point to make. Of course it was there in the tox studies and Phase I trials. It seems like something more concrete would be needed.

How does your company handle this? Do you know of any good reading material on the topic of qualfying impurities.

Thanks Adam

[DR]

Sorry I can't help on this - we tend to focus on AA & AB rated generic opportunities. No INDs, hence no TOX work. All we have to do is match or improve upon the RLD's impurity profile over the course of our stability studies.

[ravenwork]

Greetings,

This has been one of the most intelligent and informative threads in a while. Thanks to everyone. I enjoyed following it. I don't work with this issue, but it is good to know about it.

Evan L. Cooper

[mpepler]

"It is generally expected that you stress samples to about 10% degradation (potency) and obtain a mass balance within 98% to 102%."

Generally expected by whom? Where is there a reference in a industry journal, textbook or regulatory guideline?

It seems to me the consensus is that there is little scientific basis for the requirement and this coincides with my own view, unfortunately I can't cite an online discussion thread as a reference in a technical report justifying discarding the requirement.

Thanks to everyone for their comments.

[JM]

During Stress studies it is "good to show the mass balance" but in reality it is normally not acheived due to following reasons,

1. Forced degradation is to prove the selectivity of the method in terms of interference to the main peak and major known impurities.
2. Since the conditiones used are very harse and not usually ancountered in the real stability studies. Such stress conditiones generate sencondry and tertiary degradation and you will never know whether the resultant products are UV active or not , in certain cases gases librate from the sample and end product might be a acid with no UV or just a pptation.
3. There is NO guidline or industry acceptable practice for mass balance limit for forced degradation studies but you need to demostrate the mass balance of the method , whatever is acheived if all other validation parametrs are OK.

JM

[Kostas Petritis]

I do not know if your molecules contain nitrogen or not but if they do then a CLND detector might be useful for your case (due to its equimolar response to nitrogen). I remember several years ago, in a conference, someone from Pfizer (I think) he gave an speech where he was using a UV coupled with a CLND, he was using the CLND just to calculate the response factors and then he was doing the rest of the work with the UV...

A ELSD might also be useful but it's equimass response is not as good as the equimolar response of the CLND...