Back to the Top
Dear PharmPK group:
I want to gather opinions on the preferred way to apply NCA to a
rodent toxicokinetic study with repeated sampling from each subject
(non-destructive). In this scenario, there are also sporadic missing
time points from a small number of subjects. The overall goal is to
determine differences in exposure between subsets (male vs. female,
dosage level). Here are the specifics:
36 rodents (18M,18F) treated at 3 dosage levels (A,B,C), divided into
two sampling groups (group 1 - 0.5, 2, 8, 24hrs; group 2 - 1, 4,
12hrs). Three males and females in each sampling group treated at
each of 3 dosage levels: [(3+3)*2]*3=36 rats.
The following subjects had missing C-t data -
Female X, Dosage Level B, group 1, 2hr time point
Female Y, Dosage Level B, group 2, 4hr time point
Female Z, Dosage Level C, group 2, 4hr time point
My first impression is to use a compartmental approach and apply
NONMEM to test the effect of covariates. However, to maintain
consistency in toxicokinetic data reports, we prefer to apply NCA, so
compartmental approaches are out for the moment. My second thought is
to use Bailer's method, or a derivation thereof. Yet this method
assumes that each observation is from a separate subject
(destructive), and does not account for intra- and inter-subject
variation. I would prefer to use this straightforward Bailer method,
but I have doubts about its performance in the repeated sampling
situation. Of note, Hing et al. apparently applied the Bailer
approach to TK data sets with repeated samples per subject to identify
subset differences, and compared this to NONMEM (J Appl Toxicol. 2002
Nov-Dec;22(6):437-43.PMID: 12424748). Their conclusion: "Point
estimates for exposure differ significantly between NC and MEM
methods, but exposure ratios estimates between subsets of animals are
broadly similar. . ." If one is confident of minimal inter-subject
variation, then I guess one could apply the Bailer method to repeated
samples, but one does not know of such variability a priori. CIs in
this instance would probably be so wide as to only allow detection of
grossly different exposures between subgroups, > 2-fold perhaps.
Finally, my least preferred approach - conduct NCA on the subjects in
each sampling group, drop subjects with missing data points
altogether, and compare AUC0-t, Cmax, and Tmax between the subsets
using standard parametric or non-parametric statistical tests within
(but not between) the sampling groups.
Advice or comments would be appreciated. Also, please chime in on
what approach you have taken, or are currently taking, with such
repeat sample TK data.
Thanks,
Burgess
Back to the Top
Hi Burgess, for the NCA data analysis determine how much each point
contributes to the total AUC. If the contribution is minimal then you
can probably just compare the AUCs.
Neil
Back to the Top
The following message was posted to: PharmPK
Dear Dr Freeman,
If you are using WinNonlin you may be interested to look at the
Sparse Sample option for NCA that uses the subject information to
calculate standard errors that will account for any correlations in the
data resulting from repeated sampling of individual animals. It is
switched on my clicking a check box as you apply the methods 200-202 and
will give you a Subject ID to map in the Data variables window.
The method will calculate standard error for the mean concentration
curve's maximum value (Cmax), and for the area under the mean
concentration curve from dose time through the final observed time.
- Standard error of the mean Cmax will be calculated as the
sample standard deviation of the y-values at time Tmax divided by the
square root of the number of observations at Tmax.
- Standard error of the mean AUC will be calculated as described
in Nedelman and Jia (1998), using a modification in Holder (2001), and
will account for any correlations in the data resulting from repeated
sampling of individual animals.
The full references are below - I hope you find this is an
interesting and useful alternative to a full 'pop' approach.
Nedelman and Jia (1998). An extension of Satterthwaite's approximation
applied to pharmacokinetics. J Biopharm Stat 8(2):317-28.
Holder (2001). Comments on Nedelman and Jia's extension of
Satterthwaite's approximation applied to pharmacokinetics. J Biopharm
Stat 11(1-2):75-9.
Best regards,
Simon.
--
Simon Davis
Senior Scientific Consultant
PharmPK Discussion List Archive Index page
Copyright 1995-2009 David W. A. Bourne (david@boomer.org)