Back to the Top
Dear All
I am an graduate student doing a PK-PD modeling using mixed effect
model(NonMEM). However, the PD data is survival data which means they
in the format of time to the endpoint. We are trying to approve the
association between PK exposure and the survival. Also we gonna test
some clinical measurements' significance on the patient survival.
The biggest problem we have now is that some PD data are censored,
which means the patients could die or lost follow-up for some reasons
before the study endpoint. For pure statistic survival analysis,
there are certain methods to handle those censored data. But I could
not find any information in PK-PD modeling. Does anyone has
experience regarding this issue? Thank you very much in advance for
your help!
Jiayin Huang
Graduate student
hjypharm.at.gmail.com
Back to the Top
The following message was posted to: PharmPK
Allan,
There are a couple of papers that have reported how to do time to
event analysis using NONMEM. This allows the use of time courses of
concs or other biomarkers to predict the hazard of the event. I also
recommend the book by Collett for understanding parametric hazard
models.
Nick
1. Cox E, Veyrat-Follet C, Beal S, Fuseau E, Kenkare S, Sheiner L. A
population pharmacokinetic-pharmacodynamic analysis of repeated
measures time-to-event pharmacodynamic responses: the antiemetic
effect of ondansetron. J Pharmacokinet Biopharm. 1999;27(6):625-44.
2. Cox E, Sheiner LB. Repeated measures time to event with time-
varying concentration and hazard. 2001 ftp://ftp.globomaxnm.com/
Public/nonmem/non_continuous
3. Hu C, Sale ME. A joint model for nonlinear longitudinal data with
informative dropout. J Pharmacokinet Pharmacodyn. 2003 Feb;30(1):83-103.
4. Collett D. Modelling survival data in medical research. 2nd ed.
Boca Raton: CRC Press; 2003.
--
Nick Holford, Dept Pharmacology & Clinical Pharmacology
University of Auckland, 85 Park Rd, Private Bag 92019, Auckland, New
Zealand
email:n.holford.at.auckland.ac.nz tel:+64(9)373-7599x86730 fax:373-7556
http://www.health.auckland.ac.nz/pharmacology/staff/nholford/
Back to the Top
Allan,
One thing you need to ask is whether the event is repeated or not.
If the event is not repeated there is no need to use a program like
Nonmem for the analysis. There are many, many statistical programs
that do the analysis much easier.
One way to do the analysis if cox proportional hazards, but I have
found that interpreting a hazard is difficult to explain to
laypeople. This is the standard approach. A better approach, and
one that I am starting to use almost exclusively, is accelerated
failure time models in which the parameters are directly
interpretable in terms of surivival times.
The BEST book on learning survival analysis is Kleinbaum and Klein
Survival Analysis: A self-Learning Text by Springer and another good
book is Statistical Methods for Survival Data Analysis by Lee and
Wang.The former discusses AFT models more so than the latter, but
they do complement each other well.
Good luck,
pete bonate
Back to the Top
The following message was posted to: PharmPK
Peter,
> One thing you need to ask is whether the event is repeated or not.
> If the event is not repeated there is no need to use a program like
> Nonmem for the analysis.
Even if you have only one event per person there is still the
possibility that the fixed effects are best explained by a non-linear
function of the explanatory variable.
>From what I know (very little) of SAS and Splus it is not possible
to specify non-linear models e.g. using SAS lifereg. NONMEM can do
this because it explicitly states the hazard model.
Nick
--
Nick Holford, Dept Pharmacology & Clinical Pharmacology
University of Auckland, 85 Park Rd, Private Bag 92019, Auckland, New
Zealand
email:n.holford.aaa.auckland.ac.nz tel:+64(9)373-7599x86730 fax:373-7556
http://www.health.auckland.ac.nz/pharmacology/staff/nholford/
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 "PK/PD survival analysis" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)