Log in
Register|
Yasen_salsa ☆ Bulgaria, 2011-05-31 20:42 (5118 d 08:14 ago) Posting: # 7039 Views: 7,809 |
|
|
Dear All, I would like to ask if anyone had an analytical problem in determination of metformin in plasma samples when used anticoagulant is CPDA1 (Citrate Phosphate Dextrose Adenine) or for some other reason in the samples of the volunteers there are two anticoagulants (heparin and CPDA1). I would be very grateful for any ideas for potential problems with such analysis. Best regards, Yassen |
|
Ohlbe ★★★ France, 2011-06-01 01:24 (5118 d 03:32 ago) @ Yasen_salsa Posting: # 7042 Views: 6,618 |
|
|
Dear Yassen, ❝ ... when used anticoagulant is CPDA1 (Citrate Phosphate Dextrose Adenine) or for some other reason in the samples of the volunteers there are two anticoagulants (heparin and CPDA1). I can't answer specifically for metformin. But I have seen several times differences in MS response between CPDA plasma and EDTA or heparin plasma. The problem with CPDA is the dilution of your plasma. There is a large volume of buffer added to blood. As it does not go into the red blood cells, it results in a dilution of plasma of about 20 %. Consequences:
Recommendation: use the same anticoagulant for all samples for all subjects. Use the same anticoagulant for standard and QC samples as for subject samples. Use the same anticoagulant in your validation as for the subject samples. Regards Ohlbe — Regards Ohlbe |
|
Dr_Dan ★★ Germany, 2011-06-01 13:33 (5117 d 15:23 ago) @ Ohlbe Posting: # 7048 Views: 6,477 |
|
|
Dear Ohlbe Correct me if I am wrong, but as long as you compare Test and Reference in a bioequivalence study the same way, a different anticoagulant should be no problem provided that the analytical method is validated with this specific anticoagulant, right? You might have different absolute concentrations in comparison to other studies but you will be consistent within your own study. Regards Dan — Kind regards and have a nice day Dr_Dan |
|
Ohlbe ★★★ France, 2011-06-01 14:04 (5117 d 14:52 ago) @ Dr_Dan Posting: # 7050 Views: 6,576 |
|
|
Dear Dan, What I have seen or heard of several times is studies where the validation had been done with CPDA, standard/QC samples in the study were also prepared with CPDA plasma, but subject samples were taken on EDTA or heparin. Observation during the sample analysis: two-fold difference in IS response between the two anticoagulants. Investigations: difference in matrix effects, not corrected by the internal standard. Consequence: all concentrations were inaccurate. I don't think regulators would accept the fact that the concentrations are just as inaccurate for the test as for the reference as an excuse for a lousy method... The problem is also that in my experience the difference in IS response and therefore the level of inaccuracy varies from run to run. Regards Ohlbe — Regards Ohlbe |
|
Yasen_salsa ☆ Bulgaria, 2011-06-01 19:08 (5117 d 09:48 ago) @ Ohlbe Posting: # 7051 Views: 6,621 |
|
|
Dear Ohlbe, Thank you for your fast answer. I had tried to be as short as possible, but as I see I will have to explain in details our problem to clarify the picture. We validated bioanalytical method using plasma from a blood bank with a CPDA1 anticoagulant. The same anticoagulant was used in the S-Monovette for the study samples. In the stage of development and pre-study validation we didn’t encountered any problems with all parameters of the method (very good precision, accuracy, selectivity, linearity, recovery et.ec.). During study validation all quality control samples (more than 150) had very good precision and accuracy, but for our great surprise we have a big difference in the results for incurred samples. We didn’t have any problem with IS (the area of QC samples and study samples had CV~2%). The big part of the second results (ISR) had highest concentration to the primary results, like there is a difference in concentration (or pH – because we use ion-pair solid phase extraction in sample preparation) between different levels in Cryogenic Vials of the volunteers plasma samples. The samples of volunteers are thawed at room temperature (without homogenization) and centrifuge, we simulated all condition of different stages of samples preparation and storage conditions (have proof even stability for the transportation in dry ice) and didn’t get any deviation from the results of the QC’s samples. It looks very strange to have such difference in concentration levels, as Metformin is negligibly bound to plasma proteins and we didn’t meet such problem with QC’samples, or with the samples from freeze/thaw stability which are also not homogenized after thawing and are centrifuge before sample preparation. The only difference between samples for quality control and samples of volunteers for which we suppose is the existence of another anticoagulant (possibly heparin) in samples of volunteers. We assume that in the clinical center it was used to prevent blood clots in cannula and in this way it came to the samples. It looks very unlikely to have such problem from small quantities of heparin, but it is the only difference which we found from the QC samples. We will tried to homogenized study samples with vortex and without further centrifugation before taking the aliquot, to tried to eliminate this strange problem if the reason is unhomogenized study samples. We would appreciate if anyone shared his opinion if it is possible the reason for this problem to be unhomogenized samples or presence of heparin in the samples or if anyone have any other ideas about the reasons for this results. PS: In the method we used UV detection. Best regards, Yasen Edit: Full quote removed. Please delete anything from the text of the original poster which is not necessary in understanding your answer; see also this post! [Helmut] |
|
Ohlbe ★★★ France, 2011-06-02 02:10 (5117 d 02:46 ago) @ Yasen_salsa Posting: # 7053 Views: 6,486 |
|
|
Dear Yasen, ❝ ...as I see I will have to explain in details our problem to clarify the picture. Yes, it does help a lot ! ❝ PS: In the method we used UV detection. OK, so at least we can rule out ion suppression / enhancement ! ❝ We validated bioanalytical method using plasma from a blood bank with a CPDA1 anticoagulant. The same anticoagulant was used in the S-Monovette for the study samples. Good ! Though personally I would avoid CPDA due to the high plasma dilution, and possible variations in this dilution if the tube is under-filled in case of a difficult blood draw. ❝ It looks very unlikely to have such problem from small quantities of heparin, but it is the only difference which we found from the QC samples. Yes, I would also consider this to be very unlikely, though you can't totally exclude it. But that would not be my #1 hypothesis. ❝ The samples of volunteers are thawed at room temperature (without homogenization) and centrifuge ❝ the samples from freeze/thaw stability which are also not homogenized after thawing and are centrifuge before sample preparation ❝ We will tried to homogenized study samples with vortex and without further centrifugation before taking the aliquot, to tried to eliminate this strange problem if the reason is unhomogenized study samples. Do I understand it correctly, that you do not mix your samples (either vortexing, shaking, manual inversion or whatever), at any time, after complete thawing and before extracting them ?  ❝ We would appreciate if anyone shared his opinion if it is possible the reason for this problem to be unhomogenized samples Definitely, yes ! Have a look at this presentation from Tandem Labs. Also this recent paper* from Tan and colleagues from PharmaNet (look at case 3). No or insufficient mixing can lead to huge differences in concentration between the top and the middle of your tube. *By the way, that's a full issue on ISR in Bioanalysis. Some of the papers, including Tan's, are worth reading. Regards Ohlbe — Regards Ohlbe |
|
Yasen_salsa ☆ Bulgaria, 2011-06-02 17:38 (5116 d 11:17 ago) @ Ohlbe Posting: # 7057 Views: 6,452 |
|
|
Dear Ohlbe, Thank you for your answer and additional information, I reviewed the presentation of Tandem Labs and their problem looks similar to ours. Unfortunately I couldn’t see the publication of Tan, yet as I wait our information department to buy it. If it's not too much trouble to you I will be grateful if you can send it to me. ❝ Good ! Though personally I would avoid CPDA due to the high plasma dilution, and possible variations in this dilution if the tube is under-filled in case of a difficult blood draw. We knew about that problem but unfortunately the only blank blood, we can get from the blood bank is with an anticoagulant CPDA1. ❝ Do I understand it correctly, that you do not mix your samples (either vortexing, shaking, manual inversion or whatever), at any time, after complete thawing and before extracting them ? Yes, after thawing, we do not homogenize the samples, as it is necessary to centrifuge the samples (to prevent blockage of SPE cartridges) and this initial homogenizing seems pointless. We suppose that our problem is not coming from homogenization because as I mentioned before we simulated this with samples for freeze/thaw stability and long term stability. These samples were spiked with standard in the same Cryogenic vials as study samples and freeze and thawed 3 times without homogenization and there were no deviations in the results using the same procedure. However, today we'll try to homogenize the samples without subsequent centrifugation to eliminate this potential reason for failure. ❝ Yes, I would also consider this to be very unlikely, though you can't totally exclude it. But that would not be my #1 hypothesis. May be I should have to explain further why we believe that our problem may be caused by the presence of heparin or in a difference between the plasma from the blood bank and the plasma from the volunteers. In the sample preparation we used ion-pair SPE and it was noticed that by increasing the acid (from heparin in samples or any other difference in the matrix) loading the recovery decreases (probably by affecting the ionization of IPR). But it’s not our #1 hypothesis. Best regards, Yasen Edit: Standard quotes restored. @Yasen: I activated the e-mail in your profile. [Helmut] |
Ing. Helmut Schütz