[NMusers] RE: IMP method not stationary

From: Bauer, Robert <Robert.Bauer_at_iconplc.com>
Date: Tue, 24 Nov 2015 20:24:37 +0000

Carl:
You may want to try a couple of things.

First, NONMEM 7.3 should be used, as improvements on IMP are always made.

Next, after the SAEM step, try

$EST METHOD=IMP MAPITER=0 EONLY=1 NITER=20

Where MAPITER=0 has NONMEM not perform a MAP step even on the first itera=
tion, but borrows conditional means and variances as proposal density infor=
mation from the previous SAEM step.

If this still does not work, then go in the opposite direction and have MAP=
 done on every iteration:
$EST METHOD=IMP MAPITER=0 EONLY=1 NITER=20




Robert J. Bauer, Ph.D.
Vice President, Pharmacometrics R&D
ICON Early Phase
Office: (215) 616-6428
Mobile: (925) 286-0769
Robert.Bauer_at_iconplc.com<mailto:Robert.Bauer_at_iconplc.com>
www.iconplc.com<http://www.iconplc.com>

From: owner-nmusers_at_globomaxnm.com [mailto:owner-nmusers_at_globomaxnm.com] On=
 Behalf Of Panetta, Carl
Sent: Tuesday, November 24, 2015 9:35 AM
To: nmusers_at_globomaxnm.com
Subject: [NMusers] IMP method not stationary

I am having some issues with the IMP method using EONLY=1 option to gener=
ate an estimate of the objective function.

Specifically I am first running the SAEM method and the model appears to be=
 converging well (i.e. SAEMOBJ is stable). But when I then try to get the e=
stimate of the objective function using the IMP method the OBJ value is not=
 stable and actually is steadily increasing. (See the NONMEM output below.)=
 The data and model for this problem are a 3 compartment linear PK model wi=
th zero-order absorption (using ADVAN7) where each compartment represents e=
ither the parent drug or one of the two measured metabolites. Overall the f=
its look good and the parameter estimates are reasonable.

Note that if I only run the model with the parent drug and a one compartmen=
t model with zero-order absorption everything works as expected, i.e. the O=
BJ for the IMP step is very stable.

Are there setting for the IMP step that I should be modifying to help stabi=
lize the objective function with the more complex model (i.e. drug + metabo=
lite) vs the more basic model (parent drug alone)?

I appreciate any suggestions.

Thanks,

J. Carl Panetta, Ph.D.
Department of Pharmaceutical Sciences
St. Jude Children's Research Hospital
262 Danny Thomas Place
Memphis, TN 38105
Office: (901) 595-3172
Mobile: (901) 921-3740
Fax: (901) 595-3125
Carl.Panetta_at_stjude.org<mailto:Carl.Panetta_at_stjude.org>



NONMEM Command:
$EST PRINT=200 INTERACTION NOABORT METH=SAEM NBURN=600 NITER=800
MAPITER=0

Output:
Stochastic/Burn-in Mode
iteration -600 SAEMOBJ= 9495.99674546385
iteration -400 SAEMOBJ= 8603.58235406272
iteration -200 SAEMOBJ= 8723.87279412471
Reduced Stochastic/Accumulation Mode
iteration 0 SAEMOBJ= 8736.22287161761
iteration 200 SAEMOBJ= 8483.86390119489
iteration 400 SAEMOBJ= 8461.33184842421
iteration 600 SAEMOBJ= 8458.72312502707
iteration 800 SAEMOBJ= 8457.40631588857

NONMEM Command:
$EST PRINT=1 INTERACTION NOABORT METH=IMP EONLY=1 ISAMPLE=3000
NITER=15 MAPITER=0

Output:
iteration 0 OBJ= 11357.6200237143 eff.= 825098. Smpl.= =
   3000. Fit.= 0.99912
iteration 1 OBJ= 11480.4689169933 eff.= ******** Smpl.= =
   3000. Fit.= 0.99692
iteration 2 OBJ= 11794.3704982717 eff.= ******** Smpl.= =
   3000. Fit.= 0.99469
iteration 3 OBJ= 11877.2905908383 eff.= ******** Smpl.= =
   3000. Fit.= 0.99666
iteration 4 OBJ= 12350.7139467883 eff.= ******** Smpl.= =
   3000. Fit.= 0.99319
iteration 5 OBJ= 12822.3794461581 eff.= ******** Smpl.= =
   3000. Fit.= 0.98859
iteration 6 OBJ= 13338.5294969671 eff.= ******** Smpl.= =
   3000. Fit.= 0.98440
iteration 7 OBJ= 13800.6919759203 eff.= ******** Smpl.= =
   3000. Fit.= 0.98330
iteration 8 OBJ= 14072.5744654244 eff.= ******** Smpl.= =
   3000. Fit.= 0.98545
iteration 9 OBJ= 14703.6086363413 eff.= ******** Smpl.= =
   3000. Fit.= 0.98586
iteration 10 OBJ= 15703.8492953385 eff.= ******** Smpl.= =
   3000. Fit.= 0.97584
iteration 11 OBJ= 16114.3684511057 eff.= ******** Smpl.= =
   3000. Fit.= 0.96980
iteration 12 OBJ= 15968.9435824919 eff.= ******** Smpl.= =
   3000. Fit.= 0.98161
iteration 13 OBJ= 16309.4158350109 eff.= ******** Smpl.= =
   3000. Fit.= 0.96581
iteration 14 OBJ= 16001.2089520460 eff.= ******** Smpl.= =
   3000. Fit.= 0.97431
iteration 15 OBJ= 16920.1964345088 eff.= ******** Smpl.= =
   3000. Fit.= 0.95322 <br /><br /> ICON plc made the following annot=
ations. -----------------------------------------------------------------=
------------- This e-mail transmission may contain confidential or legall=
y privileged information that is intended only for the individual or entity=
 named in the e-mail address. If you are not the intended recipient, you =
are hereby notified that any disclosure, copying, distribution, or reliance=
 upon the contents of this e-mail is strictly prohibited. If you have rec=
eived this e-mail transmission in error, please reply to the sender, so tha=
t ICON plc can arrange for proper delivery, and then please delete the mess=
age. Thank You, ICON plc South County Business Park Leopardst=
own Dublin 18 Ireland Registered number: 145835


Received on Tue Nov 24 2015 - 15:24:37 EST

This archive was generated by hypermail 2.3.0 : Fri Sep 27 2019 - 16:47:27 EDT