Back to the Top
Hello everyone,
I have the following question (that I've been debating with various
people lately and there is no consensus of course!)
When people read the FDA guidelines, everyone interpret the
guidelines somewhat differently.
My question is the following:
1. Do you normally analyze QC LOQ samples with every run of your
validation?
2. Do you determine the assay precision at the QC LOQ level (inter
day and within day)?
All comments/feedbacks are more than welcome
Best regards,
Sylvain
Back to the Top
Hi Sylvain,
Yes you have to analyze QC LOQ samples in all the 4 or 5 PA batches
only and is not required to do in stability studies viz. bench top,
freeze thaw, plasma long term.
yes we have to determine assay precision and accuracy both for intra
and inter day.
hope this helps you
Rajareddy
Back to the Top
The following message was posted to: PharmPK
Hi Sylvain,
For a partial validation we run 5 x QCLOQ-QCLOW-QCMID-QCHIGH and
determine
the precision and accuracy within day.
For a complete validation we run 3 or 5 days 3x QCLOQ-QCLOW-QCMID-
QCHIGH and
determine the precision and accuracy inter day.
Fabrice Guillet
XENOBLIS
Back to the Top
LLOQ and ULOQ can be run as part of the accuracy and precison runs.
Guidance recommends that all stability experiments use a minimum of
the LQC and HQC at each time point. If there are issues you are
aware of with the LLOQ then it would be good to include this in the
stability as well. No requirement but....
Back to the Top
It' s necessary to analyse the QC LOQ during intra and inter day
precision. This results refers the reproducibility of your method.
Thanks .
A.Karthik
Back to the Top
Dear Fabrice:
Then, having done all that, what do you DO with what you
have determined? I think you simply decide that the assay is socially
acceptable, and then I get you store the info away and nothing
further happens with it. Such an exercise in nothing! Why not use
the3 data you get to fit a polynomial to your results so you can get
a good estimate of assay SD over its entire working range, and then
weigh the data (each result) by its Fisher information.
Very best regards,
Roger Jelliffe
Roger W. Jelliffe, M.D. Professor of Medicine,
Division of Geriatric Medicine,
Laboratory of Applied Pharmacokinetics,
USC Keck School of Medicine
2250 Alcazar St, Los Angeles CA 90033, USA
Phone (323)442-1300, fax (323)442-1302, email= jelliffe.aaa.usc.edu
Our web site= http://www.lapk.org
Back to the Top
Dear A. Karthik:
Then, having established your inter and intra day precision,
what do you DO with the info? Just decide if it is OK or not? What
about trying to define the random errors, composed of both
components, so that you can come up with an overall error that will
pertain to any sample as it comes randomly through your lab, and
assign real quantitative meaning to it, such as Fisher information?
You are not doing enough, in my opinion, when pharmacokinetic work
(or any other, for that matter) is concerned. All I get from you
guys is the same old stuff, but no response, no scientific
justification for what you do.
Speak up, fellas, please!
Roger Jelliffe
Roger W. Jelliffe, M.D. Professor of Medicine,
Division of Geriatric Medicine,
Laboratory of Applied Pharmacokinetics,
USC Keck School of Medicine
2250 Alcazar St, Los Angeles CA 90033, USA
Phone (323)442-1300, fax (323)442-1302, email= jelliffe.-a-.usc.edu
Our web site= http://www.lapk.org
Back to the Top
The following message was posted to: PharmPK
These are predictors of future success or failure. If the failures are
too high you may want to rethink the assay. On the other hand if you
meet or better the CV and Bias requirements you may want to tighten them
up or extend the range (re-validate to a lower LLOQ and/or higher ULOQ,
etc)
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
Dear Roger,
With my 3 (5 data is better) I calculate mean, SD and CV%. Then we have
criteria for these results (bias and precision) which are included in
the
final validation report. I think it's a standard practice for assay
validation to determine accuracy and precision inter-day.
The acceptance criteria are well determined in our SOPs.
All data are archived.
It's always a pleasure to play with HPLC or LC-MS/MS but an analysis
has a
cost, and we try to do only what is necessary and never throw away our
results (or our money)...
Very best regards,
Fabrice Guillet
XENOBLIS
Back to the Top
The following message was posted to: PharmPK
Dear Roger,
With my 3 (5 data is better) I calculate mean, SD and CV%. Then we have
criteria for these results (bias and precision) which are included in
the
final validation report. I think it's a standard practice for assay
validation to determine accuracy and precision inter-day.
The acceptance criteria are well determined in our SOPs.
All data are archived.
It's always a pleasure to play with HPLC or LC-MS/MS but an analysis
has a
cost, and we try to do only what is necessary and never throw away our
results (or our money)...
Very best regards,
Fabrice Guillet
XENOBLIS
Back to the Top
Dear Fabrice:
Thanks for your note. The thing that is new and different in
our approach is that we do not simple want to determine if an assay
is acceptably precise or not - we want to quantify its credibility by
determining what the random (inter and intra day) errors do to a
sample as it randomly comes through the lab assay system, so we can
determine - with its Fisher information with the assay error
polynomial - what the effect of both factors is, operating together.
Good for you for never throwing any info away. When you do
another 3 samples, you can make them different concentrations, so you
can have more data points in your polynomial then. You can also begin
to tell if the assay is drifting over time or not.
All the best,
Roger Jelliffe
|
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 "Question regarding QC LOQ" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)