Re: How to account for M-M in the presence of Ka>Ke constraint in PopPK model
Dear Lei,
I feel your pain, cuz I have also battled with models with stubborn flip-flop and I started concocting all sort of codes similar to yours to prevent it - you can probably find some desperate posts of mine about this on NMUSERS :).
In my experience, the only code that sort of works without causing too many side-effects is the one that prevents flip-flop of the typical values (the THETAs). Unfortunately, this does not work every time, but all the other codes accounting for individual parameters (such as the one you propose) introduce funny correlations between the parameters and the ETAs, they reshape the between-subject variability, and they may end up causing more trouble than they solve.
In your case with nonlinear clearance, it may be even more complicated than usual.
In my experience, the best option is to use priors on ka and volume, even weakly informative. These should help stabilising your model. Adding priors may seem "artificial", but if you think about it, it is doing exactly what you are trying to achieve with all these tricky codes. One cannot solve the flip-flop problem only with data from oral administration, the only way is to add external information, like the fact that you expect ka to be larger than ke, or include IV data that helps you identify the correct value of volume. With the priors you do just that, and in a more natural way than "cheating" NONMEM with funny codes.
Good luck!
Paolo
Quoted reply history
On 2015/07/08 22:41, Lei Diao wrote:
Dear NONMEM Users,
I have a popPK model for which the Ka is constrained to be larger than Ke at the individual level to avoid flip-flop. The question is that if there is an additional nonlinear clearance component (M-M), how should I constrain between the absorption rate (KA) and terminal phase elimination rate (KE) since nonlinear clearance causes the KE to change with time? Is there any reference on this topic?
KA and KE constraining in the absence of nonlinear clearance in NM code:
KE=((K+K23+K32)-SQRT((K+K23+K32)*(K+K23+K32)-4*K*K32))/2
KA=KE+THETA(5)*EXP(ETA(3))
Thanks a lot for your input!
Lei Diao
Biogen
--
------------------------------------------------
Paolo Denti, PhD
Pharmacometrics Group
Division of Clinical Pharmacology
Department of Medicine
University of Cape Town
K45 Old Main Building
Groote Schuur Hospital
Observatory, Cape Town
7925 South Africa
phone: +27 21 404 7719
fax: +27 21 448 1989
email: paolo.denti_at_uct.ac.za<mailto:paolo.denti_at_uct.ac.za>
------------------------------------------------
________________________________
UNIVERSITY OF CAPE TOWN
This e-mail is subject to the UCT ICT policies and e-mail disclaimer published on our website at http://www.uct.ac.za/about/policies/emaildisclaimer/ or obtainable from +27 21 650 9111. This e-mail is intended only for the person(s) to whom it is addressed. If the e-mail has reached you in error, please notify the author. If you are not the intended recipient of the e-mail you may not use, disclose, copy, redirect or print the content. If this e-mail is not related to the business of UCT it is sent by the sender in the sender's individual capacity.