From: "Jing Li" jli29@jhmi.edu
Subject: [NMusers] simulation problem
Date: Wed, 22 Nov 2006 12:30:34 -0500
Dear users,
I am doing a PK-PD simulation with effect compartment model to simulate
the drug concentrations and PD biomarker concentrations simultaneously
under different dose levels. The drug is administered as intravenous
infusion. The PK is a three-compartment model with Michaelis-Menton
elimination from the central compartment, and the PD is the stimulatory
Emax model.
After simulation, I found there is something wrong with the simulated
drug and biomarker concentrations. In general, the simulated drug
concentration-time profiles look good, but there is not always a
consistent increase in the concentration (e.g., Cmax at the end of
infusion) with the increase of the dose. The simulated biomarker
concentrations exhibit an erratic profile. I would very much appreciate
if someone can point out where I am doing wrong in this PK-PD
simulation.
The code is as followings:
$PROB SIMULATE PK AND PD DATA TO SIMPLE PK-LINK-PD MODEL
$DATA DRUG_SIM_3.csv IGNORE=@
$INPUT ID AMT=DOSE RATE TIME DV MDV EVID SEX BSA FLAG CMT
$SUBROUTINES ADVAN6 TOL=5
$MODEL NCOMP = 4
COMP=(CENTRAL,DEFDOSE,DEFOBS)
COMP=(PERIPH1)
COMP=(PERIPH2)
COMP=(EFFECT)
$PK
VM = 122*(1+0.502*(2-SEX))*EXP(ETA(1))
KM = 111*EXP(ETA(7))
V1 = 8.15*(BSA/1.8)**0.69*EXP(ETA(2))
V2 = 7.54*EXP(ETA(3))
V3 = 25.5*EXP(ETA(4))
Q2 = 5.01*EXP(ETA(5))
Q3 = 2.74*EXP(ETA(6))
K12 = Q2/V1
K21 = Q2/V2
K13 = Q3/V1
K31 = Q3/V3
S1 = V1
;EFF COMP
K14 = .001*K12
K40 = THETA(1)*EXP(ETA(8))
S4 = S1*K14/K40
EMAX = THETA(2)*EXP(ETA(9))
C50 = THETA(3)*EXP(ETA(10))
BSL = THETA(4)*EXP(ETA(11))
;OBS = XDV
$DES
C1 = A(1)/V1
DADT(1) = - A(1)*VM/(KM+C1)- (K12+K13+K14)*A(1)+K21*A(2)+K31*A(3)
DADT(2) = K12*A(1) - K21*A(2)
DADT(3) = K13*A(2) - K31*A(3)
DADT(4) = K14*A(1) - K40*A(4)
$ERROR
TMPC=0
IF(CMT.EQ.1) TMPC=1
TMPE=1-TMPC
DEL = 0
IF(F.EQ.0) DEL = .000001
IF (TMPC.EQ.1) IPRED = LOG(F + DEL)
IF (TMPC.EQ.1) W=.269
IF (TMPE.EQ.1) IPRED = LOG(BSL*(1+EMAX*F/(C50+F)))
IF (TMPE.EQ.1) W = THETA(5)
IRES = DV-IPRED
IWRES = IRES/W
Y = IPRED+ERR(1)*W
$THETA .321 ;1 K40
$THETA 2.75 ; 2 EMAX
$THETA 659 ; 3 C50
$THETA 46.2 ;4 BL
$THETA .267 ;5 W
$OMEGA 0 FIX ;1 VM
$OMEGA 0 FIX ;2 V1
$OMEGA 0 FIX ;3 V2
$OMEGA 0 FIX ;4 V3
$OMEGA 0 FIX ;5 Q2
$OMEGA 0 FIX ;6 Q3
$OMEGA 0 FIX ;7 KM
$OMEGA 0 FIX ;8 K40
$OMEGA 0 FIX ;9 EMAX
$OMEGA 0 FIX ;10 C50
$OMEGA 0 FIX ;11 BSL
$SIGMA 1 FIX
$SIMULATION (9215690) ;seed 1-7 digits
$TAB ID AMT TIME DV SEX BSA CMT ONEHEADER NOPRINT FILE=mytab14
The datafile for the first three subjects is as the following:
#ID AMT RATE TIME DV MDV EVID SEX BSA FLAG CMT
1 354 1074 0 . 1 1 1 1.8 0 1
1 0 0 0 . 0 0 1 1.8 3 4
1 0 0 0.33 . 0 0 1 1.8 2 1
1 0 0 0.33 . 0 0 1 1.8 3 4
1 0 0 1 . 0 0 1 1.8 2 1
1 0 0 1 . 0 0 1 1.8 3 4
1 0 0 2 . 0 0 1 1.8 2 1
1 0 0 2 . 0 0 1 1.8 3 4
1 0 0 4 . 0 0 1 1.8 2 1
1 0 0 4 . 0 0 1 1.8 3 4
1 0 0 8 . 0 0 1 1.8 2 1
1 0 0 8 . 0 0 1 1.8 3 4
1 0 0 12 . 0 0 1 1.8 2 1
1 0 0 12 . 0 0 1 1.8 3 4
1 0 0 24 . 0 0 1 1.8 2 1
1 0 0 24 . 0 0 1 1.8 3 4
1 0 0 48 . 0 0 1 1.8 2 1
1 0 0 48 . 0 0 1 1.8 3 4
2 709
2149 0 . 1 1 1 1.8 0 1
2 0 0 0 . 0 0 1 1.8 3 4
2 0 0 0.33 . 0 0 1 1.8 2 1
2 0 0 0.33 . 0 0 1 1.8 3 4
2 0 0 1 . 0 0 1 1.8 2 1
2 0 0 1 . 0 0 1 1.8 3 4
2 0 0 2 . 0 0 1 1.8 2 1
2 0 0 2 . 0 0 1 1.8 3 4
2 0 0 4 . 0 0 1 1.8 2 1
2 0 0 4 . 0 0 1 1.8 3 4
2 0 0 8 . 0 0 1 1.8 2 1
2 0 0 8 . 0 0 1 1.8 3 4
2 0 0 12 . 0 0 1 1.8 2 1
2 0 0 12 . 0 0 1 1.8 3 4
2 0 0 24 . 0 0 1 1.8 2 1
2 0 0 24 . 0 0 1 1.8 3 4
2 0 0 48 . 0 0 1 1.8 2 1
2 0 0 48 . 0 0 1 1.8 3 4
3 1418 4298 0 . 1 1 1 1.8 0 1
3 0 0 0 . 0 0 1 1.8 3 4
3 0 0 0.33 . 0 0 1 1.8 2 1
3 0 0 0.33 . 0 0 1 1.8 3 4
3 0 0 1 . 0 0 1 1.8 2 1
3 0 0 1 . 0 0 1 1.8 3 4
3 0 0 2 . 0 0 1 1.8 2 1
3 0 0 2 . 0 0 1 1.8 3 4
3 0 0 4 . 0 0 1 1.8 2 1
3 0 0 4 . 0 0 1 1.8 3 4
3 0 0 8 . 0 0 1 1.8 2 1
3 0 0 8 . 0 0 1 1.8 3 4
3 0 0 12 . 0 0 1 1.8 2 1
3 0 0 12 . 0 0 1 1.8 3 4
3 0 0 24 . 0 0 1 1.8 2 1
3 0 0 24 . 0 0 1 1.8 3 4
3 0 0 48 . 0 0 1 1.8 2 1
3 0 0 48 . 0 0 1 1.8 3 4
Thanks in advance!
Jing
simulation problem
From: "Jing Li" jli29@jhmi.edu
Subject: Re: [NMusers] simulation problem
Date: Wed, 22 Nov 2006
Hi Sam,
Thanks for the corrections! However, the corrected model does not look
very helpful for the simulated PD biomarker concentrations. Here I am
attaching the simulated biomarker concentration-time profiles for you to
review. I don't understand why the biomarker concentrations increased
again at 48 h in some subjects and there is not an apparent
dose-dependent response.
Any thoughts? Thanks again.
Jing
From: Sam Liao sliao@pharmaxresearch.com
Subject: Re: [NMusers] simulation problem
Date: Wed, 22 Nov 2006 16:40:56 -0500
Hi Jing:
I tried your model and I think there are few minor corrections
needed in your model as shown below:
$DES
C1 = A(1)/V1 DADT(1) = - C1*VM/(KM+C1)- (K12+K13+K14)*A(1)+K21*A(2)+K31*A(3) DADT(2) = K12*A(1) - K21*A(2) DADT(3) = K13*A(1) - K31*A(3)
DADT(4) = K14*A(1) - K40*A(4)
I used C1 instead of A(1) in the first term of DADT(1). I replaced K13*A(2) with K13*A(1) in
the first term of DADT(3). Hope this will work for you.
Best regards,
Sam Liao
Pharmax Research
From: Sam Liao sliao@pharmaxresearch.com
Subject: Re: [NMusers] simulation problem
Date: Wed, 22 Nov 2006 19:16:26 -0500
Hi Jing:
I think it is because of the random error. It is already very near the baseline. When I fixed your
SIGMA1 to zero, I got about the same values for time 0, 24 and 48 hr.
Best regards,
Sam Liao
Pharmax Research
_______________________________________________________