Dear all,
I'm trying to fit concentrations of a drug simultaneously measured in plasma as
well as intracellularly (in peripheral blood mononuclear cells). I observe a
high accumulation in cells, but with a delayed absorption. I'm trying to fit
the intracellular data dependent of the plasma pharmacokinetics. The volume of
distribution of the cellular compartment can be considered negligeble compared
to the volume of distribution in plasma (no distribution to a second
compartment can be observed in the plasma pharmacokinetics. However, cellular
accumulation depends on plasma concentrations.
I'm fitting the data with a first order oral absorption and elimination,
basically it can be summarized as:
$Model (DOSE) (CENTRAL) (CELL)
I have a rich sampling dataset of 11 individuals with at each timepoint an
observation in the cellular and plasma compartment. I have tried several
approaches for modelling cellular accumulation. I have tried fixing the
cellular volume to a very small volume.
$PK
k12=theta(1)
v1=theta(2)
cl=theta(3)
k20=cl/v1
v3=0.0001
k23=theta(4)
k32=theta(5)
This didn't work. K23 was estimated to be very small and k32 was estimated to
be very large, giving the same fit as estimating an accumulation ratio, which
is not a good fit, since a delayed absorption in the cellular compartment was
observed. There is some mass transport going on between the central and
cellular compartments, that I do not want. My cellular pharmacokinetics depend
on the plasmapharmacokinetics, but I don't want my cellular pharmacokinetics to
influence my plasma pharmacokinetics, since this effect is likely negligible.
Does anyone have a smart idea on how to code this?
Sincerely,
Rob ter Heine
_______________________________
Rob ter Heine, MSc, PharmD
Department of Pharmacology, Slotervaart Hospital
Amsterdam, The Netherlands
E: [EMAIL PROTECTED]
T: +31-20-5124737
Modelling intracellular and plasma data
Dear Rob,
I am not an expert on intracellular PK models, but your data appear to
be similar to a typical effect compartment PK system with a delayed time
to Cmax in the effect compartment relative to the time of Cmax in the
plasma compartment. If the PK in the intracellular compartment does not
influence the PK in the plasma compartment (as you stated below), then
you cannot estimate both k23 and k32 independently. I suggest trying
the following modifications.
k32=theta(4); similar to Keo in an effect compartment model
k23=k32*v3/v1
Good luck!
Steve
Steven M. Troy
Senior Director
Global Clinical Pharmacology and Pharmacokinetics
Shire Pharmaceuticals
725 Chesterbrook Boulevard
Wayne, PA 19087-5637
Office: 1.484.595.8780
Mobile: 1.484.375.3692
Email: [EMAIL PROTECTED]
Quoted reply history
-----Original Message-----
From: [EMAIL PROTECTED] [mailto:[EMAIL PROTECTED]
On Behalf Of Rob ter Heine
Sent: Tuesday, December 09, 2008 8:28 AM
To: [email protected]
Subject: [NMusers] Modelling intracellular and plasma data
Dear all,
I'm trying to fit concentrations of a drug simultaneously measured in
plasma as well as intracellularly (in peripheral blood mononuclear
cells). I observe a high accumulation in cells, but with a delayed
absorption. I'm trying to fit the intracellular data dependent of the
plasma pharmacokinetics. The volume of distribution of the cellular
compartment can be considered negligeble compared to the volume of
distribution in plasma (no distribution to a second compartment can be
observed in the plasma pharmacokinetics. However, cellular accumulation
depends on plasma concentrations.
I'm fitting the data with a first order oral absorption and elimination,
basically it can be summarized as:
$Model (DOSE) (CENTRAL) (CELL)
I have a rich sampling dataset of 11 individuals with at each timepoint
an observation in the cellular and plasma compartment. I have tried
several approaches for modelling cellular accumulation. I have tried
fixing the cellular volume to a very small volume.
$PK
k12=theta(1)
v1=theta(2)
cl=theta(3)
k20=cl/v1
v3=0.0001
k23=theta(4)
k32=theta(5)
This didn't work. K23 was estimated to be very small and k32 was
estimated to be very large, giving the same fit as estimating an
accumulation ratio, which is not a good fit, since a delayed absorption
in the cellular compartment was observed. There is some mass transport
going on between the central and cellular compartments, that I do not
want. My cellular pharmacokinetics depend on the plasmapharmacokinetics,
but I don't want my cellular pharmacokinetics to influence my plasma
pharmacokinetics, since this effect is likely negligible. Does anyone
have a smart idea on how to code this?
Sincerely,
Rob ter Heine
_______________________________
Rob ter Heine, MSc, PharmD
Department of Pharmacology, Slotervaart Hospital
Amsterdam, The Netherlands
E: [EMAIL PROTECTED]
T: +31-20-5124737
______________________________________________________________
This message has been scanned for all viruses by BTnet VirusScreen.
The service is delivered in partnership with MessageLabs.
This service does not scan any password protected or encrypted
attachments.
If you are interested in finding out more about the service,
please visit our website at
http://www.btignite.com/internetservices/btnet/products_virusscreen.htm
==============================================================
Seasons Greetings from your friends at Shire!
As part of our ongoing commitment to Corporate Responsibility and environmental
awareness, it is a Shire tradition not to send cards at this time of the year.
The funds that would be used to create and send these cards are contributed to
charitable organizations around the globe in regions where Shire employees live
and work. Please visit our Shire Corporate Responsibility website
(www.shirecr.com) where you can find out more information about the charities
we have contributed to this season, under Latest News. Heres wishing you a
very Merry Christmas, Happy Holiday and peaceful New Year!
Please consider the environment before printing this e-mail
This email and any files transmitted with it are confidential and
may be legally privileged and are intended solely for the use of
the individual or entity to whom they are addressed. If you are
not the intended recipient please note that any disclosure,
distribution, or copying of this email is strictly prohibited and may
be unlawful. If received in error, please delete this email and any
attachments and confirm this to the sender.
www.shire.com
Rob,
I would start by modeling plasma concentrations separately. Then, when you know the model that gives you good description, you can treat predicted concentrations in plasma as a driving force of your other system, that is, intracellular compartment. You can work in $DES terms, then you do not need to introduce any mass transfer from your main system to the second one.
There are many ways to model delayed absorption. You may try transit compartment model, or effect-compartment type model. Once you figured out how to model is, you can combine two parts together.
Example:
$DES
DADT(1) = -K*A(1) ; plasma C1=A(1)/V1
DADT(2) = K2*A(1)-K23*A(2) ;"transit"(can be duplicated to increase delay)
DADT(3) = K23*A(3)-K30*A(3) ; intracellular C3=A(3)/V3
Thanks
Leonid
--------------------------------------
Leonid Gibiansky, Ph.D.
President, QuantPharm LLC
web: www.quantpharm.com
e-mail: LGibiansky at quantpharm.com
tel: (301) 767 5566
Rob ter Heine wrote:
> Dear all,
>
> I'm trying to fit concentrations of a drug simultaneously measured in plasma as well as intracellularly (in peripheral blood mononuclear cells). I observe a high accumulation in cells, but with a delayed absorption. I'm trying to fit the intracellular data dependent of the plasma pharmacokinetics. The volume of distribution of the cellular compartment can be considered negligeble compared to the volume of distribution in plasma (no distribution to a second compartment can be observed in the plasma pharmacokinetics. However, cellular accumulation depends on plasma concentrations.
>
> I'm fitting the data with a first order oral absorption and elimination,
> basically it can be summarized as:
>
> $Model (DOSE) (CENTRAL) (CELL)
>
> I have a rich sampling dataset of 11 individuals with at each timepoint an
> observation in the cellular and plasma compartment. I have tried several
> approaches for modelling cellular accumulation. I have tried fixing the
> cellular volume to a very small volume.
>
> $PK
> k12=theta(1)
> v1=theta(2)
> cl=theta(3)
> k20=cl/v1
> v3=0.0001
> k23=theta(4)
> k32=theta(5)
>
> This didn't work. K23 was estimated to be very small and k32 was estimated to
> be very large, giving the same fit as estimating an accumulation ratio, which
> is not a good fit, since a delayed absorption in the cellular compartment was
> observed. There is some mass transport going on between the central and
> cellular compartments, that I do not want. My cellular pharmacokinetics depend
> on the plasmapharmacokinetics, but I don't want my cellular pharmacokinetics to
> influence my plasma pharmacokinetics, since this effect is likely negligible.
> Does anyone have a smart idea on how to code this?
>
> Sincerely,
>
> Rob ter Heine
>
> _______________________________
> Rob ter Heine, MSc, PharmD
> Department of Pharmacology, Slotervaart Hospital
> Amsterdam, The Netherlands
> E: [EMAIL PROTECTED]
> T: +31-20-5124737
Dear all,
I received some useful suggestions for modelling intracellular pharmacokinetics
andI'd like to thank everyone who replied in the list or personally.
Ultimately, visual inspection of my data suggested there was a delay in
absorption and elimination from the cellular compartment. However, individual
estimation of these parameters using an effect compartment or just plain
modelling of intracellular pharmacokinetics independent of plasma data, did not
improve the model compared to estimation of an accumulation ratio.
For future searching purposes I'd like to sum up the suggestions I received:
1 - The effect compartment in ADVAN5 (Steven Troy)
If the PK in the intracellular compartment does not
influence the PK in the plasma compartment (as you stated below), then
you cannot estimate both k23 and k32 independently. I suggest trying
the following modifications.
k32=theta(4); similar to Keo in an effect compartment model
k23=k32*v3/v1
2 - The effect compartment in ADVAN 6 (Nele Plock/Atul Bhattaram/Jacob Brogren)
you might try an effect compartmental approach. Use your central
compartment, and then describe the cellular concentrations as being in the
effect compartment (the good thing ist that the effect compartment does
not include any mass transfer).
here's the code for a two-compartment model:
DADT(1)= -K12*A(1)
DADT(2)= K12*A(1)-K23*A(2)+ K32*A(3) -K20*A(2)
DADT(3)= K23*A(2)-K32*A(3)
DADT(4)= KEO1*(A(2)/V2-A(4))
3 - No mass transfer in ADVAN6 (Ron Keizer)
dadt (1) = -ka*a(1)
dadt (2) = ka*a(1) - k20 *a(2)
dadt (3) = k23*a(2) - k32*a(3)
Cheers,
Rob
_______________________________
Rob ter Heine, MSc, PharmD
Department of Pharmacology, Slotervaart Hospital
Amsterdam, The Netherlands
E: [email protected]
T: +31-20-5124737