Structure of internal standard in MRM method

[MrGC]

I am going to quantify a carboxylic acid derivative in a biological sample in MRM mode. The sample preparation only consists of adding ACN (spiked with IS) followed by centrifugation. How important is the use of an isotopically labelled internal standard in this case? Isn't it sufficient to have a standard that is structurally more or less similiar (i.e. having the same functional groups) since no structure dependent sample loss is to be expected?
What do you guys think?

To not start another thread:
How important is the observation of a qualifier ion in addition to a quantifier ion in target analysis?

Best regards

[Steve Reimer]

An internal standard that elutes with the target analyte will more accurately reflect any suppression or enhancement caused by matrix effects. With a similar compound the retention time will likely be different and so won't reflect what is happening at the target's retention time.

[MrGC]

Thank you for responding! That is obviously true but are matrix effects of particular note if I make several calibration curves in a given reference medium before I make my quantitation? Aren't matrix effects in this case already factored in?

[Rndirk]

I agree with Steve. Recovery correction is just one function of an internal standard.

If there's an isotope labelled analogue (commercially) available, and you have the budget for it, I advice to use it.

To not start another thread:
How important is the observation of a qualifier ion in addition to a quantifier ion in target analysis?

We always record at least one qualifier ion. Recording it doesn't mean you have to check the ion ratios every time (for instance in calibration and spiked samples). It doesn't take time to record it, so better to be safe than sorry.

[James_Ball]

Sometimes you just can't find an isotopically labeled version of the analyte of interest and it would be very cost prohibitive to have one custom made, therefore another functionally similar analyte that you do not expect to find in the sample matrix will work, and the closer it is in retention time to the analyte of interest the better, for evaluating matrix effects in the MS source.

Qualifiers are not always required but the more you have, the higher confidence you can have that you do not have a false positive result.

[Kenn]

Sometimes you just can't find an isotopically labeled version of the analyte of interest and it would be very cost prohibitive to have one custom made, therefore another functionally similar analyte that you do not expect to find in the sample matrix will work, and the closer it is in retention time to the analyte of interest the better, for evaluating matrix effects in the MS source.

I agree wholeheartedly and had this situation just the other day. A recent aminoglycoside analysis in wastewater was causing significant suppression at the instrument. I altered the gradient and pushed the analytes of interest out to a retention time of about 12 minutes (instead of 3-4) and fortified all samples and standards with another aminoglycoside that was not in samples. The fortification of samples was done AFTER all processing so it wasn't going to correct for extraction efficiency only instrument suppression. Pushing that retention out so my analytes were not eluting with a bunch of polars helped but I was still looking at 60-75% recoveries. Employing that other antibiotic as internal standard and those recoveries were all 90+%. I don't think the question is can you do it but does it work. Sometimes 20-30k is just not in the budget for a deuterated analogue.

Qualifiers are not always required but the more you have, the higher confidence you can have that you do not have a false positive result.

Some of our European clients are now asking for quantitation to be performed and reported for both the primary and secondary mass transitions in method validations. A new SANCO guideline that I believe they are misinterpreting but what are you to do, they are the clients. For the ensuing toxicological studies that validation was done to support, that confirmation ion need not be reported other than it was there and confirmed the identity of the analyte. There have been a couple of occasions where there is no confirmation ion or it's only 10% of the primary and you can't see it in you bottom 2 standards which poses a problem. When they start talking about changing the method from a simple direct aqueous analysis to an extraction just to see that qualifier is when I just put my face in my hands.

[Rndirk]

Some of our European clients are now asking for quantitation to be performed and reported for both the primary and secondary mass transitions in method validations. A new SANCO guideline that I believe they are misinterpreting but what are you to do, they are the clients.

So you have to set up your processing method to calibrate and calculate on ion 1, and again on ion 2 giving you twice the amount of data/QC/...

This is where I would personally draw the line

About the original post. If an internal standard is not an option, and if a structural analogue is not available/suitable, you can still do a very good LCMS quantification by means of standard addition of the analyte - which corrects for basically everything an internal standard does in LCMS. The drawback is that you have to prepare a series for each different sample. You should also know at least the order of magnitude of concentration of the analyte present in the original sample.

I like using standard addition for those kind of one-time analyses where our customer service team promises a (new and interesting) client to perform a quantification of a special kind of analyte in a special kind of sample - completely out of our scope.

[Kenn]

So you have to set up your processing method to calibrate and calculate on ion 1, and again on ion 2 giving you twice the amount of data/QC/...

That is precisely what you must do. It's really no different than doing a multi-component analysis and yes you generate quite a bit of data with half of it being pointless.

This is where I would personally draw the line

Except I'm not the one in charge of drawing lines.