Back to the Top
In the spirit of the CL discussion, I wold appreciate comments on the following
1) The 2001 Guidance for Bioanalytical method validation suggests using 6 runs consisting of at least five samples at five concentrations ( LLOQ, LQC, MQC, HQC, and ULOQ) to assess assay accuracy and precision. Acceptance for each level is based on at least 4 of the 6 samples in the level meeting target acceptance criteria. Acceptance of the run is based on each level meeting acceptance.
While this is being maintained for the instrumental approaches (chromatography), ligand binding accuracy and precision runs have been reset to include only 2 samples at each level. This obviates the 4 of 6 criterion and reduces the information available as well as increases the risk of failure.
I would appreciate your opinion on this difference.
2) A separate issue is of assessing stability. The interpretation is that the mean of three samples responses at a level-timepoint is taken and used to accept or reject stability. What can happen is that two of the three may fail accuracy while the mean and precision are acceptable. By the collective mean the stability is accepted but if two of three fail should stability be accepted for that level? See the example below:
The criteria is 20%CV and +/- 20%BIAS.
Target is 10
%BIAS
VALUE 1 7.6 -24
VALUE 2 10.0 0
VALUE 3 7.5 -25
MEAN 8.37 -16.3
SD 1.42
%CV 16.9
-- Edward F. O'Connor
78 Marbern Drive
Suffield, Ct 06078-1533
Back to the Top
The following message was posted to: PharmPK
Edward,
Regarding your question on stability (based on a 20% acceptance criteria):
My understanding is that stability of quality control material is indicated if at least 60% of the replicates and the mean of all replicates are within +/-20% of nominal. Your example in #2 has two values not meeting the acceptance criteria, therefore stability is not indicated.
Stephanie Gozum, Ph.D.
Back to the Top
It's actually 67% but I agree. Collapsing the data may obscure the actually performance.
-- Edward F. O'Connor
78 Marbern Drive
Suffield, Ct 06078-1533
Want to post a follow-up message on this topic?
If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "Accuracy and precision, Stability" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)