Thi Nguyen
☆    

Australia,
2017-09-18 16:43
(2383 d 03:45 ago)

Posting: # 17814
Views: 5,313
 

 Odd QC values but not outliers [Bioanalytics]

Dear all,

I tried to look through all related topics in the forum and found some questions very closed with my issue, however, I could not get the final answer. Could you please advise what I should do in my case: My lab is validating a method for a compound in human plasma. Two core runs have been done (by 2 different analysts), QCs in 3 levels low, medium and high (6 replicates at each level) past with flying colours (all around 100% for accuracy). No carry-over observed. BUT: core run 1 has one LLOQ 498% and one LLOQ 727%, the other four LLOQs have good accuracy 〜100%. Core run 2 has five LLOQs with accuracy 〜100% and one LLOQ 3000%. For core run 2, 3000% can be regarded as outlier (from Grubb's test - calculated by Prism) but for core run 1, both values 498% and 727% are not outliers.

We couldn't figure it out whether there are any sample processing errors so we could't exclude those values from accuracy and precision calculation for LLOQ, it means the core run 1 failed.

We stop validation at this point to have investigation. Much appreciated if you can help with any clues for this situation.

Thanks and best regards,
Thi Nguyen
Ohlbe
★★★

France,
2017-09-18 17:53
(2383 d 02:34 ago)

@ Thi Nguyen
Posting: # 17815
Views: 4,619
 

 Odd QC values but not outliers

Dear Thi Nguyen,

❝ BUT: core run 1 has one LLOQ 498% and one LLOQ 727%, the other four LLOQs have good accuracy 〜100%. Core run 2 has five LLOQs with accuracy 〜100% and one LLOQ 3000%. For core run 2, 3000% can be regarded as outlier (from Grubb's test - calculated by Prism) but for core run 1, both values 498% and 727% are not outliers.


Forget about outlier tests in such situation. The validation is there to demonstrate that your method works. If you exclude those values that may indicate it doesn't, you're losing the plot, whatever the result of the outlier test.

Now there are several possible explanations, including contamination, sample spiking error and carry over. It is rather difficult to comment further without more information. I would go back to the data and look at, for instance:
  • how were the samples spiked practically ? Risk of contamination of LLOQ samples ?
  • do the concentrations measured match with the nominal concentration of LQC, MQC or HQC samples, or of a calibration sample ? Could be just the use of the wrong spiking solution.
  • were these samples injected right after a sample with a high concentration (carry-over) ?
  • are there "strange" results at other QC levels too, even if "less" abnormal ?
Plenty of possible reasons and aspects to look at. But don't exclude any result unless you have identified what happened... And "we must have made a mistake somewhere" is not considered an "identifiable cause".

Regards
Ohlbe
Thi Nguyen
☆    

Australia,
2017-09-19 15:14
(2382 d 05:13 ago)

@ Ohlbe
Posting: # 17816
Views: 4,567
 

 Odd QC values but not outliers

Thank you so much Ohlbe for your reply!

We are trying to identify the cause of the issue and focusing on the contamination. Few tests have been done such as extracting LLOQ at the same time with ULOQ by different analysts. The result of first analyst showed perfect clean, no-contaminated, consistent LLOQs (6 replicates). The second batch from another analyst is running right now.

LLOQ and all other QCs (LQC, MQC, HQC) were prepared in bulk by spiking solutions into 10 ml plasma then aliquoted into small vials (enough for 2 replicates in assay) so LLOQ are homogenous, it shouldn't cause one very high, another from the same vial normal.

And yes, the concentrations measured match with the nominal concentration (accuracy were quite good, most of them around 100%). The abnormal only happened with LLOQ, but not with any other QCs.

No carry-over was observed at all, and the LLOQ with high values were not following any ULOQ in the run.

Agree with you that there are plenty of possible reasons and aspects to look at. That is my concern: If we can't identify any causes, what should we conclude about the method?
Could you please advise how you do in this case in your lab?

Thanks and best regards,
Thi Nguyen
ElMaestro
★★★

Denmark,
2017-09-19 15:38
(2382 d 04:49 ago)

@ Thi Nguyen
Posting: # 17817
Views: 4,528
 

 Odd QC values but not outliers

Hello Thi Nguyen,

can you:

1. Indicate if the same sample no from run 2 was also one of the affected samples in run 1? (I mean if the affected sample in run 2 was "sample 17" then check if "sample 17" was also funky in run 1 etc)

2. Post bits of the result table, in particular indicate if the aberrant result for certain sample arose out of a fluke in the area of the analyte itself, or due to a low IS response, or possibly both?

3. Tell if the RTs in analyte and IS was more or less the same for affected samples and unaffected samples?

4. Eyeball the integration and check of you feel it was correct? Sometimes, at LLOQs depending on noise and bunching and a ton of factors that I know very little about, the actual interpolated baseline under the analyte chromatogram is visibly off and this in itself can give rise to some seriously deviating areas.

5. Check if the chromatograms are rugged, peaks-on-peaks etc.

Pass or fail!
ElMaestro
Ohlbe
★★★

France,
2017-09-19 15:49
(2382 d 04:38 ago)

@ Thi Nguyen
Posting: # 17818
Views: 4,557
 

 Odd QC values but not outliers

Dear Thi Nguyen,

In addition to ElMaestro's comments and questions:

❝ LLOQ and all other QCs (LQC, MQC, HQC) were prepared in bulk by spiking solutions into 10 ml plasma then aliquoted into small vials (enough for 2 replicates in assay) so LLOQ are homogenous, it shouldn't cause one very high, another from the same vial normal.


OK, so it doesn't look like a spiking error.

❝ And yes, the concentrations measured match with the nominal concentration (accuracy were quite good, most of them around 100%). The abnormal only happened with LLOQ, but not with any other QCs.


Right. My question was also: does the measured concentration of these problematic LLOQ samples match with the nominal concentration of e.g. HQC or CC7 or whichever other spiked sample ? It could be simply a labelling error after aliquoting, or taking the wrong sample out of the freezer.

❝ No carry-over was observed at all, and the LLOQ with high values were not following any ULOQ in the run.


OK. Could it be a contamination at some point ? How do you process the samples: manually, or in 96-well plates with robotic equipment, or what ?

Regards
Ohlbe
UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,636 registered users;
91 visitors (0 registered, 91 guests [including 6 identified bots]).
Forum time: 19:28 CET (Europe/Vienna)

Nothing shows a lack of mathematical education more
than an overly precise calculation.    Carl Friedrich Gauß

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5