Back to the Top
Dear All,
I would like to provide some statistical input to a standard PK study in
infant of 1 month to 4 years old in the context of the draft FDA guideline
on pediatric PK studies. We are looking to collect some PK parameters in
these children (AUC, Cmax, ...). Our drug is not bound to plasma protein in
adult (<10%). It is mostly excreted in urine unchanged (66%) and 24% as an
acidic compound (not via CYP 450).
As a biostatistician, I am now wondering if a analysis with age
subclassification (1-3 months, 3-6 months and 6 months-2 years) would be
more efficient than using some regression techniques with a set of
covariates like age, weight, length, bsa, creatinine clearance, ... . I
would also appreciate some advice regarding the collection of covariables
that could reflect the renal function maturity and correlate some PK
parameters.
Any kind of help would be highly appreciated.
Thanks in advance,
Edouard Ledent
Statistician
UCB S.A. Pharma Sector
Chemin du Foriest
B-1420 Braine L'Alleud / Belgium
Phone : + 32 2 386 25 16
Fax : + 32 2 386 30 44
SMTP : Edouard.Ledent.at.UCB-Group.com
Back to the Top
The best way, I think, to do a study like you mentioned is to
estimate the individual pk parameters (by using SAAMII
or the same kind software) as much as you could, I mean, try to get
the parameters from individual as much as
possible and then do the multivariate analysis to find out if there
is any potential correlation between your pk parameters
and the subjects' physical or medical condition. If there is any then
you can re-run your pk profile by using a real
population pk program like NONMEM . If your individual
concentration-time data is too small to estimate individual
pk parameters you have to run population pk .
Zhao
Zhao Wang, M.D
Northwestern University
Medical School
Anesthesia Research
z-wang.-a-.nwu.edu
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)