Back to the Top
The following message was posted to: PharmPK
Hi All,
I am seeking some opinions about calibration as we are having a bit of a
debate where I work about the extrapolation of values above and below
calibration lines.
I work for a company where we do a number of PK and tox studies and
occasionally get values in an analysis that although falling within the
nominal calibration range lie outside of the calibration
line for that particular run. For example, taking an assay where the
validated nominal range is say 0.1 - 100 ng/ml and allowing a 15.0%
deviation for each standard , you may have a calibration line where the
highest acceptable standard is perhaps 93 ng/ml and the lowest is 0.11
ng/ml
(each analysis always has standard samples run to establish a
calibration
line).
To my mind, any unknown samples run as part of the same analysis, giving
concentrations between 93.0-100 ng/ml and 0.10-0.11 ng/ml are therefore
off
the calibration line and must be considered above/below the limit of
quantitation. However, company policy is that although they are beyond
the
calculated calibration line, they lie within the validated range so
they can
be considered as acceptable values and may be reported.
Maybe I'm missing something obvious here - and someone please tell me
if I
am, that's the point of this post - but this seems to completely defy
the
point of doing the calibration line in the first place. On this basis
why
bother with the calibration line at all, why not just compare the
results to
an agreed, validated calibration line acquired at the start of the
project?
Surely the point of running individual calibration lines is to take into
account the differing environmental and user conditions (e.g.
temperature,
detector response, operators etc)?
I appreciate that in reality this only affects a small number of
samples but
as a point of principal it seems completely wrong. Or is it just me
missing
the obvious?
Comments welcome.
Thanks!!
Back to the Top
Dear Fredrick and all,
Generally you would want to stick to your validation parameters.
For example, are you saying that the validated assay range is:
A) 0.1 - 100 ng/ml, or
B) 0.1 - 100 ng/ml WITH 15.0 % deviation
Once you validate a range, anything falling outside of the range is not
validated.
Another point would be relating to significant figures and tolerances
(see USP,
general notices). Make sure you're using the appropriate number of
significant
figures, and that you are consistent.
Also, you should comply with any company policy, SOP and/or protocol
already
established.
Now, regarding running a calibration curve with each assay.
It goes back to the validation and the way you've validated the assay.
Also, you're going to have QC samples and/or other controls that you
may want
run in each assay. These assay controls are usually used as criteria
for valid
assay (see Bioanalytical Method Validation Guidance, for example).
Hope this helps.
Roy
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)