USP Based Validation - Coeluter with RID

Discussions about HPLC, CE, TLC, SFC, and other "liquid phase" separation techniques.

5 posts Page 1 of 1
Hello

We are in the process of validating a method for inositol within a powdered nutrient blend. The method is based on the USP Monograph for Inositol utilizing a Ca++ Ion exclusion column - unfortunately we have a coeluter that does not look like it will resolve appropriately with allowed levels of modification of the method parameters outlined in the monograph.

My question is not regarding finding a way to resolve these two - I am pretty sure I could figure that out, but probably not without major modifications to the method (changing column type, mobile phase, detector etc), and we actually have a GCMS method that solves this problem. However switching instruments or substantial swaying from the monograph will cause issues with our in house auditor and if that can be avoided I would like to do so.

My question is with regard to dealing with the co-eluter in our calculations. Our preliminary and possibly inappropriate way to do so is to run an inositol free matrix blank and do a manual area subtraction of the coeluter peak from our analyte peaks.

At this point we have all the data and I am going to proceed with my protocol and report, if at least for the opportunity to practice and learn from the process even if it means getting laughed at and kicked in the ass, but would appreciate comments on whether or not others have proceeded in such a fashion with regard to an unresolvable co-eluter (utilizing a mandated method) and just explained the situation and process for dealing with it in their validation protocol and report.

Thanks
If I were reviewing this, I would say that doing a baseline subtraction, in and of itself, constitutes a modification of the compendial method and thus requires validation of the modified method.

In general, I don't like baseline subtraction because it adds an additional step, and thus additional variability. That said, I will accept it if the validation data show acceptable reproducibility.
-- Tom Jupille
LC Resources / Separation Science Associates
tjupille@lcresources.com
+ 1 (925) 297-5374
1. You don't need to validate a USP method for a USP active.

2. You're developing/validating a method for such active in a finished blend, so it's not the USP method, it's YOUR method, so you need to validate. So I'd work on the separation, not do a subtraction.

We used USP actives at low levels in finished products, so we always had to develop and validate our own methods for those products. For the USP actives, typically we just used the USP procedures.
Thank you for the input - I have submitted the protocol and report for review that had the subtraction we shall see how that flies.

In the meantime I have begun exploring an alternative procedure for separating these two lovebirds (inositol (mw 180) and glucuronolactone (mw 176)) via the use of HPLC and an RI detector. I see that this HILIC method has been used for glucuronolactone:

https://www.google.com/url?sa=t&rct=j&q ... hT86ycZVV-

and there have also been examples of similar setups being used for inositol. However with the glucuronolactone example the detector was the quite selective MSD. I am not familiar enough with the amide columns used here to know if an isocratic run would ever be able to separate these two (and can't find an example where the two were both analytes of interest in the same run) using this HILIC method and an RID.

So I thought I might ask for some insight before I go through the time and expense of setting it up if it comes to it - do these two chemicals look like they can be pulled apart on an amide column (HILIC) with an isocratic method to accomodate the RID as I don't see it happening to any useful degree with my current ion exclusion set up. I'm not expecting someone to know directly if this can be answered confidently in the affirmative - but it would be nice to know if anyone can confirm I'd just be wasting my time going that route.

Thanks
Gizmo wrote:
Thank you for the input - I have submitted the protocol and report for review that had the subtraction we shall see how that flies.


I assume you mean submitted to your QA department.

Even if QA OKs this, you may want to also keep working on a separation method. Even with RID, first thing I'd try would be 10 C higher, 20 C higher, 10 C lower, 10 C lower, and see if the separation gets better, worse, or stays the same, and maybe explore if any of those gives an improvement.

Second, I might try adding a little organic/more organic into the mobile phase and see if the separation gets better, worse, or stays the same, and maybe explore if any of those gives an improvement. Try 5%, 10%, etc. try methanol, ACN, even HPLC ethanol, THF. One sunscreen mixture we had (5 actives) turned out to work best with HPLC ethanol (denatured with methanol and IPA) which gave us separations we could not get otherwise, and we validated that procedure for FDA product.

Third - because there is a cost, unless a supplier will send you a column to try - would be to try a column of same or similar phase from vendor or from a different vendor. Even a small difference in phase (such as endcapping) could help (or hurt). I assume that you've tried a smaller particle size too, wouldn't other with longer columns as any improvement that way is typically minimal.
We always used new columns for method development ! Anything else is typically a waste of time.
5 posts Page 1 of 1

Who is online

In total there are 20 users online :: 0 registered, 0 hidden and 20 guests (based on users active over the past 5 minutes)
Most users ever online was 234 on Tue Feb 06, 2018 7:33 am

Users browsing this forum: No registered users and 20 guests