Back to the Top
Dear Group,
I would like to have the opinion of group members on following
situations:
In a bioanalysis of a clinical study samples, if a standard or
quality control sample is just outside 15% limit, say within 15.5%,
whether one need to reject such points? If this rejection point
happens to be last point in calibration curve, then assay dynamic
range will suffer. I am aware of FDA guidelines, but I want to know
how such situation is addressed.
Second point I want clarification is, regarding missing internal
standard. if internal standard addition is missing in a sample, but
the area count of the sample is matching with other two replicates,
can one externally apply avergae area count response in the study
run, and use this point in PK analysis? If not, should it be reported
as missing sample?
Thanks,
Vincent
Back to the Top
The following message was posted to: PharmPK
The 15 is actually or in most places 15.0%, percents usually being
reported to 1 decimal place. 15.5 is > than 15.0 therefore reject it.
The usual contruction of a curve allows for the LLOQ and ULOQ to have a
slightly higher variance than other points. Are you describing an LLOQ
or an LQC? If it is the LLOQ, you may want to investigate weighting the
curve 1/y or 1/y2 then reprocessing all the data including
validation-amend the validation. If it is the LQC your options are
somewhat more limited-you could open the tolerance through deviation or
amendment.
Ed O'Connor, PhD
Technical Director, Immunoanalytical
Tandem Laboratories
115 Silvia Street
West Trenton, New Jersey
609-228-0243
Back to the Top
The following message was posted to: PharmPK
Repeat the sample due to analytical errors- internal standard not
added.
Ed O'Connor, PhD
Technical Director, Immunoanalytical
Tandem Laboratories
115 Silvia Street
West Trenton, New Jersey
609-228-0243
Back to the Top
Dear Vincent,
I agree with the reply that Ed O'Connor gave regarding a standard or
quality control sample is just outside 15% limit (e.g., within 15.5%)
and whether one need to reject such points or a run based on the 15%
acceptance criteria in the FDA Guidance for Validation of
Bioanalytical Methods. You are right that for standards, this could
have significant consequences. The range could be reduced, and the
QCs will need to be within the range as well, unless a different
acceptance process is defined in an SOP for such occurrence.
I would also add the following: What does your SOP say? If it says
the limit is 15.0%, then even 15.1% will not be acceptable. If it
says 15%, then when you round, 15.4% will become 15%, and you are
within your specifications.
I would avoid accepting the value by writing a deviation. The FDA
does issue 483 citations for this type of acceptance.
-Tom
Thomas L. Tarnowski, Ph.D.
Bioanalytical Development
Elan Pharmaceuticals
700 Gateway Boulevard
South San Francisco, CA 94080
thomas.tarnowski.at.elan.com
Back to the Top
Dear Vincent,
If the ULQ of the Calibration curve fails (even if 15.2 for 15.0%)
during subject sample analysis, you may proceed with having the next
concn below your ULQ. ex. if Std J is ULQ and it fails you can take
Std I to be your quantification point provided your calibration range
is within the limit of Cmax. Also One among the entire CC set may be
risked but now 2 cosecutive concn's should fail.
For the missing internal standard the entire batch may be treated as
a failed one and should be reanalysed. Matching of concentrations to
peak response is not recommended
Hope this helps,
Santosh Tata
Bioanalytical Division, BEC,
Apotex Inc, Bangalore
Back to the Top
Dear Vincent
I firmly believe that you should not accept the batch. The batch
should be rejected. Give a chance to change the integration
parameters and see whether the value falls in the acceptable range.
For a GLP study, I think the acceptance limit is 15% until FDA sets
appropriate guidelines for rounding off 15.4% to 15%. The decision
lies with you whether to accept or reject depending on the importance
of the study results. If meant for submission to regulatory
authorities, my suggestion is reject the batch.
The missing of IS in the sample indicates that the sample is not
processed properly or subjected for neglected processing. So, go for
repeat analysis of the single sample with a fresh linearity and a set
of QC samples. The same should be documented in appropriate format.
Back to the Top
The following message was posted to: PharmPK
For the first question - QCs represent the unknown samples.
Therefore, it
is necessary to stay with the acceptance criteria. If one goes to accept
every special case then this becomes a subjective assessment.
Binodh
Back to the Top
The following message was posted to: PharmPK
For the first question - QCs represent the unknown samples.
Therefore, it
is necessary to stay with the acceptance criteria. If one goes to accept
every special case then this becomes a subjective assessment.
Binodh
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 "Acceptance criteria for bioanalysis" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)