Back to the Top
Hello All,
Can anyone tell me whether there is an accepted SOP for analyzing plasma
samples from clinical pharmacokinetic studies (actually any Pk study
clinical or otherwise). Specifically, does running each experimental
sample in duplicate make sense? Would confidence in the data be increased
or would the Pk parameter estimates be improved enough to warrant the extra
effort. This is assuming that the appropriate QC and calibration standards
are analyzed and that the assay itself has been properly validated. The
samples are serially collected so they represent single points in a plasma
vs time profile as opposed to single time points collected to monitor
therapeutic drug levels.
Any info would be greatly appreciated.
Jim Peggins
**************************
James O. Peggins, Ph.D.
Food & Drug Administration
Center for Veterinary Medicine
Division of Residue Chemistry
8401 Muirkirk Rd
Laurel, MD 20708
Voice: (301) 827-8092
Fax: (301) 827-8170
***************************
Back to the Top
Jim,
We are rarely required to run assays in duplicate for our clients at
this time.
Glenn Hanson
Covance Laboratories.
Back to the Top
Dr. Peggins:
If your definition of "duplicate" entails the serial collection of
blood samples from two subjects at the same time points, or the serial
collection of blood samples from the same subject at the same time points on
two different occasions, then the "duplicate" concentration measurements may
be used to estimate within and between subject variances. These statistical
parameters are useful for the quantitative estimation of intra-subject or
inter-subject pharmacokinetic variability.
Within and between subject variances are commonly used in
differential weighting of individual data according to its quality and
quantity. Probability distributions of individual pharmacokinetic parameter
estimates may also be modeled given within and between subject variances.
Replicate experimental designs have been informally endorsed by
FDA's CDER in a draft guidance entitled "In Vivo Bioequivalence Studies
Based on Population and Individual Bioequivalence Approaches".
Regards,
Howard S Mehler PhD
Howard S Mehler PhD JD & Associates Inc
6399 Wilshire Boulevard Suite 310
Los Angeles, CA 90048
Voice: (310) 271-0755
FAX: (323) 782-9342
e-mail: hmehler.aaa.mehler.com
Back to the Top
Dear Jim:
The real issue is - what is the relative credibility of eacj and every
data point you are fitting? This can be assessed by weighting each data
point by its Fisher information, which is the reciprocal of the variance of
each point. The thing to do is to empirically describe the relationship
between the assay and the SD at each point. You can do this by measuring
(in at least quadruplicate) a blank (including positive and negative
results), a low level, a medium one, a high one, and a very high one. Then
you can describe the relationship between concentration and SD with a
polynomial such as SD=A1 + A2C + A3C**2 + A4C**3. With this polynomial you
now have a cost-effective way to have a good estimate of the SD of any
other single specimen, and can fit each data poing by its Fisher
information. Actually, the weight should most properly be given to the
expected value instead of the measured value, but the differences are small
in our experience. You might look at
Therap Drug Monit 15: 380-393, 1993. Yes, there are many other sources of
noise in the system, but they can be used to scale the assay error
polynomial, as the assay error os probably the major determinant of
credibility of the measured data.
Does this help?
Roger Jelliffe
Roger W. Jelliffe, M.D.
USC Lab of Applied Pharmacokinetics
(323)442-1300, fax (323)442-1302
jelliffe.aaa.hsc.usc.edu
************************************
You might check our web site for info on our
new PK/PD workshop in LA on Feb 26-27, 1999.
Our web address is www.usc.edu/hsc/lab_apk
************************************
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)