FW: FW: [NMusers] Question of fitting population PK model using summary statistics of data instead of raw data

From: Penny Zhu <penny.zhu_at_novartis.com>
Date: Fri, 11 Sep 2015 10:38:36 -0700

Dear Dr Gibiansky
Thank you very much for the suggestion. I largely agree with you that it s=
eems to be an trial and error thing to make the variability match the model=
 prediction if we have a strong assumption about the model structure.

I was also wondering whether it is possible to simulate individual patient =
data at each timepoint based on the mean, steandard deviation, and an using=
 an assumption that within patients (especially in adjacent timepionts) the=
 Pk concentrations are more correlated compared to between patients. Then =
use these simulated data to fit the population PK model.

Best regards.

Penny Zhu



> -----Original Message-----
> From: Leonid Gibiansky [mailto:lgibiansky_at_quantpharm.com]
>
> Sent: Thursday, September 10, 2015 5:10 PM
> To: Zhu, Penny; nmusers_at_globomaxnm.com
> Subject: Re: FW: [NMusers] Question of fitting population PK
> model using summary statistics of data instead of raw data
>
> It is likely impossible without strong assumptions. I would
> first fit the population model (fixed effects only) and then
> start to simulate with different assumption trying to match
> observed SD or CV for peaks and troughs. You may need to
> assume the structure and the magnitude of the error model
> and the structure of the IIV model (ETAs on CL, or V, or
> both equal, etc.). You may get some rough idea about the
> magnitude of the IIV but you may need strong assumptions
> about the residual and IIV model.
> Leonid
>
>
> --------------------------------------
> Leonid Gibiansky, Ph.D.
> President, QuantPharm LLC
> web:    www.quantpharm.com
> e-mail: LGibiansky at quantpharm.com
> tel:    (301) 767 5566
>
>
>
> On 9/10/2015 2:06 PM, Penny Zhu wrote:
> > Dear Dinko
> > Thank you for the suggestion.  It seems this NAD
> approach only uses the mean data and does not estimate
> inter-subject variability using the standard deviation
> data.
> >
> > My intention is to establish a population PK/PD model
> with appropriate estimation of intersubject variability
> based on the mean and standard deviation data at each
> timepoint.
> >
> > A major assumption is that we have good knowledge of
> the base
> > structure of the model (e.g. biexponential), and won't
> run the risk
> > mistaking 2 mono exponential models for a biexponential
> model
> >
> > Your help and discussions will be very much
> appreciated.
> >
> > Penny
> >
> >
> >   -----Original Message-----
> >   From: Rekic, Dinko [mailto:Dinko.Rekic_at_fda.hhs.gov]
> >   Sent: Thursday, September 10, 2015
> 10:41 AM
> >   To: Zhu, Penny
> >   Subject: RE: [NMusers] Question of
> fitting population PK
> >   model using summary statistics of data
> instead of raw data
> >
> >   See the link and text below.
> >
> >   http://accp1.org/pharmacometrics/theory_popmeth.htm#np=
d
> >
> >
> >   Naive averaged data approach (NAD)
> >
> >       A model without BSV and
> BOV is fitted to the
> >   mean data from all individuals.
> >
> >       Features
> >
> >       
>    -Specialized software not
> >   necessary.
> >
> >       Disadvantages
> >
> >           -Does not
> distinguish between
> >   BSV and WSV.
> >
> >       
>    -Inappropriate means lead to
> >   biased parameter estimates.
> >
> >           -May
> produce model distortion
> >   i.e., 2 mono exponential equations
> averaged together can
> >   yield a biexponential.
> >
> >           -Covariate
> modeling cannot be
> >   performed.
> >
> >   Kind regards
> >   Dinko
> >   _________________________________
> >   Dinko Rekić, Ph.D., MSc(Pharm)
> >   Pharmacometrics reviewer
> >   Division of Pharmacometrics
> >   Office of Clinical Pharmacology
> >   Office of Translational Science
> >   Center for Drug Evaluation and
> Research
> >   U.S. Food and Drug Administration
> >   10903 New Hampshire Ave
> >   Silver Spring, MD 20993
> >   WO Bldg 51, Rm 3122
> >   Office phone: (8)240 402-3785
> >
> >   "The contents of this message are mine
> personally and do not
> >   necessarily reflect any position of
> the Government or the
> >   Food and Drug Administration."
> >
> >   -----Original Message-----
> >   From: owner-nmusers_at_globomaxnm.com
> >   [mailto:owner-nmusers_at_globomaxnm.com]
> >   On Behalf Of Penny Zhu
> >   Sent: Thursday, September 10, 2015
> 9:49 AM
> >   To: nmusers_at_globomaxnm.com
> >   Subject: [NMusers] Question of fitting
> population PK model
> >   using summary statistics of data
> instead of raw data
> >
> >   Dear all
> >   Assuming the population PK or PD data
> are log-normally (or
> >   normally) distributed, if you have the
> mean and standard
> >   deviation of a readout at each
> timepoint but do not have the
> >   actual raw data (assuming all pateints
> are with the same
> >   dosing regimen, etc),  is it
> possible to establish a
> >   well fitted population PK or PD
> model?  How would one
> >   get about doing it?
> >
> >   Your help is very much appreciated
> >
> >   Penny
> >
>
Received on Fri Sep 11 2015 - 13:38:36 EDT

This archive was generated by hypermail 2.3.0 : Fri Sep 27 2019 - 16:46:01 EDT