Dear All,
We have developed an LC-UV method for our compound of interest and its metabolite using a gradient elution mode.The compound elutes at 8min, metabolite at 14min and the Internal Standard(IS) at 16min.After the valiidation of the method in monkey plasma and pilot PK studies the column deteriorated beyond repair, resulting in excessive peak tailing and splitting of the IS. Now when we used a new column of the same material and of the same batch, under the same gradient conditions , compound elutes at 14min, metabolite at 10min and the IS at 16min. All the analytes were injected individually and as a mixture on both the columns under the same mobile phase conditions on the same day. Both the columns were equilibrated by 4 blank gradient injections following a stabilisation of the column in the initial conditions for 1/2 hr at 1ml/min.
Can anyone kindly suggest the reason for the "peak swapping" in the second column?????

This isn't exactly the answer to your question, but when my lab develops and validates a test method, we use three columns of DIFFERENT lots as detailed in cGMP; this way we know that the columns will all behave the same way, and that an outside lab could likely purchase the same column and obtain comparable separations. This step takes place before any precision/accuracy, etc., begins. "Validating" cannot take place on a single column, and maybe your experience is a good example of this.

I agree with Molever above - it sounds very much like a difference in column selectivity, however I do wonder about the MP. Have you run the same batch of MP over the two columns and gotten this result? If so, then forget what I'm about to say. If the pH of you MP is near the pK's of your compound and its metabolite (if applicable), then small batch to batch variations in pH will probably have a large effect on retention.

Also, were both columns "dedicated" to this assay from new or did the one you developed the assay on have a previous life? If it was used for something else previously, then you could have gotten something adsorbed onto the column that altered selectivity.

Hope this helps - let us know what you find, please. I'm curious!

Best of luck!

Chris

I am totally agree with juddc. From my experience, the only reson for a peak swapping is change in pH of the mobile phase. In your case, it is obvious that the Ret. time of the Int Std is still remain the same but only the test cmpd and its metabolite change their RT. So it is worth to check their pK's and the pH of the MP very carefully and you might be able to figure it out. Dont worry about the validation that you already did but it is worth to develop the method to continue your study.

Good Luck!

A.S.

Raj says that the column is from the same batch of packing. Under normal circumstances this should exclude any differences in retention times, unless something else is happening or has happened.
If such a thing would happen to me, the first thing that I would do is check all other things than the column. A reversal of the elution order is very unusual, so my very first bet is a confusion of the samples. If I can exclude all events other than the column, I would investigate, if both columns were really new, when they were used for this study for the first time. It could be that the first column was used for while, before the method was developed or during method development. Then it would not be a new column any more. It could also be that the second column was used for something else, and is not representative of the batch of packing any more. A last possiblity is a long equilibration time with an ion-pair reagent on the new column.