**From vfc24106@ggr.co.uk Wed Aug 30 09:58:48 1995
**

Hi,

I have again a problem to submit to the NONMEM users group. I would like to know if it is possible to model a saturable absorption (Michaelis-Menten type) with NONMEM. I want to fit together parent drug and metabolite plasma levels. But before fitting my data, I am trying simulations with a simpler model. For that, I am using parameters values from literature (Nick Holford's paper in JPB 1992 : Models for describing Absorption Rate and Estimating Extent of Bioavailability : Application to Cefetamet Pivoxil). Followed the NM-TRAN control I used :

$PROB SIMULATION MM ABSORPTION (FROM HOLFORD)

$INPUT ID TIME DV AMT

$DATA

$SUBROUTINE ADVAN6 TOL=6

$MODEL

COMP=("MMCPT" DEFDOSE INITIALOFF) COMP=(CENTRAL DEFOBS NOOFF)

$PK

CALLFL=1

VM=THETA(1)

KM=THETA(2)

CL=THETA(3)

V=THETA(4)

ALAG1=THETA(5)

S2=V

K23=CL/V

$DES

DADT(1)=-VM*A(1)/(KM+A(1))

DADT(2)=VM*A(1)/(KM+A(1))-K23*A(2)

$ERROR

Y=F

$THETA 17.2 5.73 15.8 57.3 0.9

$TABLE ID TIME

Because of the unit of KM, VM in Nick's paper, I multiplied them by the volume.

I obtained very strange simulations : predictions up to 10 hours only increase.

Tell me what I have done wrong im my code.

Thanks in advance.

Regards,

Valerie COSSON

GLAXO-WELLCOME

VFC24106@GGR.CO.UK

******
**

**From alison Wed Aug 30 14:33:32 1995
**

Ms. Cosson did not supply data, so I created a data file with a single bolus dose to the MMCPT compartment. After the lag time, the central compartment concentration rises fairly steeply till approx. TIME=15, stays reasonably constant, then falls steeply at approx. TIME=55. The code looks fine. I think the answer to the question "[is it] possible to model a saturable absorption (Michaelis-Menten type) with NONMEM?" is yes. If the time scale seems too long, there may be some problem with the units. I have not read Holford's paper, so I cannot comment on the values used for VM, KM, CL, and V.