Mohamed Rashed
☆    

Egypt,
2016-03-17 09:46
(2933 d 03:29 ago)

Posting: # 16109
Views: 5,322
 

 Precision in bioanalytical assays [Bioanalytics]

Dear Colleagues,

For bioequivalence/bioanalytical LC-MS/MS assays we used to calculate accuracy and precision for QC samples by carrying out three samples or five samples at three concentration levels and back calculate concentrations by taking area ratios and applying regression equation. Recently, a regulatory reviewer is insisting that we calculate precision from area ratios disregarding regression equation. We examined several validated assays and found that using area ratios only resulted in huge CV% reaching up to 45% in same assays and within limits for some others.

We need your advice on how to respond to this impasse with the regulators with references.

Thanks,
Mohamed
Ohlbe
★★★

France,
2016-03-17 11:13
(2933 d 02:02 ago)

@ Mohamed Rashed
Posting: # 16110
Views: 4,524
 

 Precision in bioanalytical assays

Dear Mohamed,

❝ For bioequivalence/bioanalytical LC-MS/MS assays we used to calculate accuracy and precision for QC samples by carrying out three samples or five samples at three concentration levels...


I would even expect 4 levels: LLOQ, LQC, MQC, HQC.

❝ ...and back calculate concentrations by taking area ratios and applying regression equation.


Yes, that's the correct way to do it.

❝ Recently, a regulatory reviewer is insisting that we calculate precision from area ratios disregarding regression equation.


From which country, and based on which guideline ?

IMHO that's not the correct way to go. The aim of the validation is to ensure that the PK data are based on reliable information i.e. reliable concentrations. AUC is the area under the curve of concentrations versus time, not of peak area ratio versus time.

The EMA and FDA guidelines don't specify whether the CV should be calculated using the concentrations or the peak area ratio. But I've never heard so far of a lab calculating the CV of the ratio, or of an authority requesting this type of calculation.

❝ We examined several validated assays and found that using area ratios only resulted in huge CV% reaching up to 45% in same assays and within limits for some others.


There I'm extremely surprised. I would expect the difference to be really minimal, and I would certainly not expect the CV of the peak area ratio to reach 45 % if the CV of the concentrations was below 15 %. But there may indeed be a difference at low concentration levels, depending on the intercept.

Regards
Ohlbe
Mohamed Rashed
☆    

Egypt,
2016-03-17 15:13
(2932 d 22:02 ago)

@ Ohlbe
Posting: # 16111
Views: 4,522
 

 Precision in bioanalytical assays

Dear Ohlbe,

Appreciate your comments and indeed such huge variability for area ratios was at the lowest concentrations especially at the pg/ml levels and was one extreme case.

I am writing from Egypt and the guidelines used are mostly EMA but this apparently came from a new member who joined the BE Evaluation committee.

I welcome other comments from members.

Mohamed Rashed, PhD
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2016-03-17 17:18
(2932 d 19:57 ago)

@ Mohamed Rashed
Posting: # 16112
Views: 4,497
 

 Training required

Hi Mohamed,

❝ I am writing from Egypt […] this apparently came from a new member who joined the BE Evaluation committee.


… who apparently has not completed his training yet. SCNR. :-D

❝ I welcome other comments from members.


I fully agree with Ohlbe. In assessing (in)accuracy/(im)precision of bioanalytical methods we are only interested in back-calculated concentrations. Haven’t seen anything else in 30+ years.

Since you wanted a reference, see how the EMA’s BMV-GL defines the LLOQ:

The lower limit of quantification (LLOQ) is the lowest concentration of analyte in a sample which can be quantified reliably, with an acceptable accuracy and precision.

See also what the GL says about the calibration curve and accuracy.
(Back calculated) concentrations. Period.

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,639 registered users;
70 visitors (0 registered, 70 guests [including 5 identified bots]).
Forum time: 13:16 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