Question for Non-Compartment Anaylsis

From: Wei Lu <wlu01@ureach.com> Subject: Question for Non-Compartment Anaylsis Date: Fri, 12 Jan 2001 11:54:58 -0500 Hi, There, I have a set of PK data with IV bolus and IV infusion adminstration, and am going to do a non-compartmental analysis. However, WinNonLin seems only be be able to handle the single dosing data (either for bolus or infusion) in non-compartmental analysis. Could anybody let me know what software should I use for this analysis? Thanks Wei

From: Jeff Wald <jwald@pharsight.com> Subject: Re: Question for Non-Compartment Anaylsis Date: Fri, 12 Jan 2001 12:50:13 -0500 Wei - Selection of software should first start with a definition of what your objectives are in the analysis. If a simple slope,height, area, moment (SHAM) characterization is sufficient WinNonlin will work in this case. However, depending on the design of your study, current theory for noncompartmental approaches may leave you without any interpretable results for most PK parameters.....regardless of the software that you select! My opinion, and no doubt that of the collective group of participants on this list, is that a model based analysis is the most suitable approach. I think the only debate on the issue is whether this is true 100% percent of the time or in a large majority of cases. Furthermore, in most cases you will get more value from your data using a population approach. Feel free to respond to me directly (off the list) if you want any more info about WinNonlin or WinNonMix and support for problems like this. Regards, Jeff

From: Nick Holford <n.holford@auckland.ac.nz> Subject: Re: Question for Non-Compartment Anaylsis Date: Sat, 13 Jan 2001 08:35:41 +1300 Jeff Wald wrote: > Wei - Selection of software should first start with a definition of what > your objectives are in the analysis. If a simple slope,height, area, > moment (SHAM) characterization is sufficient WinNonlin will work in this > case. Wonderful! I think the SHAM acronym is a very appropriate description term for this kind of Cmax,Tmax,AUC PK analysis (aka Walt Disney School of PK Methodology). > My opinion, and no doubt that of the collective group of participants on > this list, is that a model based analysis is the most suitable > approach. We must not forget that SHAM PK is model based but not compartmental. A structural PK model which can describe the full time course of concentrations should be a primary goal of PK analysis. SHAM analyses discard a lot of this information by integrating out time or only considering design specific aspects. > I think the only debate on the issue is whether this is true > 100% percent of the time or in a large majority of cases. Furthermore, > in most cases you will get more value from your data using a population > approach. If there is more than 1 subject then we should be considering a population approach. Apart from the target concentration intervention setting (aka TDM) this means all PK analyses (and indeed any other kind of model based analysis) should be viewed from a population perspective. -- Nick Holford, Divn Pharmacology & Clinical Pharmacology University of Auckland, 85 Park Rd, Private Bag 92019, Auckland, New Zealand email:n.holford@auckland.ac.nz tel:+64(9)373-7599x6730 fax:373-7556 http://www.phm.auckland.ac.nz/Staff/NHolford/nholford.htm

From: michael.looby@pharma.Novartis.com Subject: Re: Question for Non-Compartment Anaylsis Date: Mon, 15 Jan 2001 09:53:50 +0100 Dear All If NONMEM was the acronym in PK of the last millenium, it certainly has now been eclipsed by SHAM ml