Dear all,Back to the Top
We are trying to validate an HPLC technique for
quantifying the analyte in plasma.
We have one problem with the meaning and the way to
calculate the accuracy of the assay.
We have to compare the true value of the analyte
considering its concentration with the value obtained
by the method. However, which is the real value?
If I do not have international standards, can I
calculate the accuracy?
Sometimes we read that the accuracy is determined
considering the real value as a sample prepared by
spiking plasma (for example) with known amounts of the
analyte. Then, which is the value obtained by the
method?
Thank you,
Paula
Paula,Back to the Top
It sounds like you need to validate two parameters, precision and
accuracy. this can be done by preparing a standard curve in the matrix that you
will be analyzing and QCs at the high, middle and low end of your curve.
Example
Curve at 1, 2, 10, 50, 250, 500, 900 and 1000 ng/mL
6 QC replicates at 3, 100, 750 ng/mL
there are two standards on each end of the QC so that the QC is
bracketed by the standard curve if one of the endpoints is dropped.
Repeat on three occasions calculating the accuracy and precision. The
concentration of the QC is calculated by fitting to the standard curve
and then divided by the theoretical concentration. Calculate the intra and
extra run relative standard deviations.
You should also examine the extraction efficiency at the high and low
end of the curve.
Depending upon your goals you can scale this back, but this should
stand up for most applications.
Good Luck
Mike
Michael D. Cameron Ph.D.
CellzDirect, Inc.
8609 Cross Park Dr, Suite 100
Austin, TX 78754
Phone: 512-615-2286
Fax: 512-834-7767
michaelc.aaa.cellzdirect.com
First of all, true values are normally the added amount, which isBack to the Top
already
known based on the preparation of standard solution and presumably
accurate
piptetting.
The obtained value is based on a calibration curve, from which the
signinal
(peak area, absorbance and et al) can be converted to the
concentration/amount
of analyte. Normally, for biofluid analysis, an internal standard is
necessary
for adminishing the errors caused by sample handling. If you don't have
an
internal standard, you have to gurantee the accuracy and precision of
each
step in your sample processing, such as the extent of vertexing,
liquid-liquid
extraction, the volume of HPLC injection.
--
314 Room, 19 Russell Street
Faculty of Pharmacy
University of Toronto
Toronto, Ontario M5S2S2
(Tel) 416-978-6996
(Fax)416-978-8511
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)