Re: The theory of relativity for CL and V...
Dear Paolo,
If I have understood your problem correctly, I have pondered this for a similar situation. I formatted the table so that each subpopulation was on a different row and then reported CL/Frel, using the rel subscript (short for relative) to place emphasis on the F differing among subpopulations. I also included Frel as the last column of the table, so it is explicitly visible in the table...Then I placed an explanation in the Table caption, the methods section, the results section and the discussion section so no-one missed it. My opinion at the time was that the only way to make the table smaller was to lose important information.
If peer reviewing, I probably wouldn't object to report the CL values etc. for F=1 (which you call pure) but I prefer to explicitly acknowledge the role of differing F's than leave it for the reader to calculate. For readers concerned about AUC as a function of drug dose, it is the actual CL/Frel that is the driver of drug exposure.
I would want to see the subpopulation relative F's somewhere in the parameter table for clarity.
Best regards, James
Quoted reply history
On 07/01/2013 13:19, Paolo Denti wrote:
> Dear NMusers,
>
> we have a question we have been debating in our group, we have come to an agreement, and we would like the input from the NMusers community. It is about how to report values for clearances and volumes when the relative bioavailability is fixed to values different from 1 in some of the subjects.
>
> We developed a 2 compartment POPK model for an orally administered drug that is a substrate of polymorphic enzyme. Individuals are categorized as slow, intermediate and fast metabolizers.
>
> Since no IV data were available, relative bioavailability was fixed to 1 for slow metabolizers, and it was found to be 20% lower (and thus 0.8) in fast and intermediate metabolizers. On top of this difference in bioavailability, clearance was estimated to be 6, 11 and 14 L/h for slow, intermediate and fast metabolizers, respectively. The other clearances and volumes in the model (V, Q, V3) were the same in the three groups.
>
> Our question is on how to report clearances and volumes for each group of metabolizers.
>
> Since we only have oral data, we can only estimate oral clearance (e.g. CL/F), and normally this is what we would like to report, but in our case this means dividing the values of CL provided by NONMEM for the different metabolizers by their respective bioavailability. This would imply reporting 3 different values not only for CL/F, but also for V/F, Q/F and V3/F, resulting in a over-populated table.
>
> Moreover, since the model contains BSV both in CL and bioavailability, also the BSV of CL/F should be calculated if we want to report it.
>
> For these reasons, we decided it is best to report the "pure" values of CL, V, Q and V3 used in the NONMEM model, and then state that, to obtain CL/F, V/F, Q/F and V3/F, one should divide by the respective bioavailability. Does this make sense?
>
> If so, how would you indicate the parameters in the table? Calling it only CL gives the idea that we are providing the value of absolute CL, which is not possible without IV data. What we are indeed reporting are the values of CL (or V, Q, V3), if the bioavailability is fixed to a certain value (1 for slow metabolizers, and 0.8 for intermediate and fast). Is there a specific name that could be used for that? Would a simple explanation under the table be acceptable?
>
> What is the experience of the community with this? Any suggestions?
>
> Thank you in advance for your help,
> The pharmacometrics group at the University of Cape Town
--
James G Wright PhD,
Scientist, Wright Dose Ltd
Tel: UK (0)772 5636914