7 replies to this topic

[bauer]

Hi,

the parameter AUCF% is not an output of WinNonlins NCA. Is there a way to derive the parameter in WinNonlin? 

 

Thanks,

 

Robert

[Simon Davis]

Hi Robert,

 

   Any secondary parameter should be creatable by using a Custom Transformation step in a Data Wizard after your initial NCA.  Regarding youe request for AUCF%, if you can quote the formula (and reference/regulatory source) then perhaps people can guide you if you are having problems.

 

  If it's a Fluctuation related parameter, please take a look at the manual as there are a few of those already in NCA when performed at Steady State which might be helpful as starting points e.g.

 

Cavg                 :   AUC from dose time to Tau divided by Tau.
%_Fluctuation   :  100*(Cmax-Cmin)/Cavg, for Cmin and Cmax between dose time and Tau

 

  Simon

[bauer]

Hi Robert,

 

   Any secondary parameter should be creatable by using a Custom Transformation step in a Data Wizard after your initial NCA.  Regarding youe request for AUCF%, if you can quote the formula (and reference/regulatory source) then perhaps people can guide you if you are having problems.

 

  If it's a Fluctuation related parameter, please take a look at the manual as there are a few of those already in NCA when performed at Steady State which might be helpful as starting points e.g.

 

Cavg                 :   AUC from dose time to Tau divided by Tau.
%_Fluctuation   :  100*(Cmax-Cmin)/Cavg, for Cmin and Cmax between dose time and Tau

 

  Simon

Hi Simon,

thank you for your message.

The formula is AUCF(%)=((AUC(above Cave) + AUC(below Cave))/AUC)*100

with Cave=AUC of the dose interval / Dose interval.

The reference is the book by Cawello: "Parameters for Compartment-free Pharmacokinetics"

and the Population by the AGAH: http://www.agah.eu/f..._group_2004.pdf

 

Best,

Robert

[Simon Davis]

Robert, writing the formula is easy enough in Phoenix, I am using the therapeutic response option to set the threshold to Cavg.  However I may be misunderstanding it because if it is (AUCtau_above_Cavg + AUCtau_below_Cavg)/AUCtau

Isn't that always 1?

 

 Simon.

Edited by Simon Davis, 04 August 2015 - 11:18 AM.

[bauer]

Hi Simon,

thank you very much. I agree. The result should be 1.

I have to do more research on AUCF%. At the moment, the only thing I have is a SAS macro for the derivation of AUCF%. That macro treats trapezials having e.g. one concentration below Cavg and the other one above Cavg in a special way. That leads typically to other results then 1.   

 

Best,

Robert

[Simon Davis]

perhaps if you can show the SAS Macro code we can see how that differs from the formula you gave - are you sure the SAS macro is what you want, if it differs from your reference ?!

 

  Simon.

[Helmut Schütz]

Hi Robert,

 

be aware that PHX interpolates the intersection of Cavg (or any other “fixed” value like the MIC) and the profile linear between increasing concentrations and loglinear between decreasing ones. This is independent from the trapezoidal rule you selected. Makes sense to me. Might explain discrepancies to your SAS-macro. Check how the interpolation is done there.

 

BTW, the linear trapezoidal should go into the waste bin of NCA anyway. I hope that Pharsight will make the lin-up/log-down the default setting in future releases of PHX.

[Simon Davis]

.... I hope that Pharsight will make the lin-up/log-down the default setting in future releases of PHX.

I doubt the factory default will change, however remember that since v6.3 you can use Edit>Preferences>Default Templates to set your own preferred version of any operational object.  Which means you could also set your preferred Parameter names etc., graph formats as well.

 

  Simon.