Is there a automatic way to include the AUCtmax without having to do it with a labor intensive copy & paste method. Will the partial AUC definitions allow the use of the Tmax varible?
WNL output of AUCtmax
Posted 07 January 2013 - 07:41 PM
You can use a Data Wizard (in PHX 1.3) or data steps (previous to 1.3) to create a dataset as input for the partial area set up that indicates to calculate the area up to tmax (calculated from a previous NCA). I have done this in the project attached so you can see what is required. Please note this project was created in Phoenix WinNonlin 6.3 and would not open in previous version. If you need a previous version project, let me know.
Hope this is helpful,
Ana [file name=AUCtmax.phxproj size=477362]http://pharsight.com/extranet/media/kunena/attachments/legacy/files/AUCtmax.phxproj[/file]
Posted 07 January 2013 - 08:18 PM
Jim, here is another way to do it that does not require a second NCA object, and instead uses Inner Join to extract AUCtmax from the Summary table. Simon. [file name=AUCtmax_bw.phxproj size=514575]http://pharsight.com/extranet/media/kunena/attachments/legacy/files/AUCtmax_bw.phxproj[/file]
Edited by Simon Davis, 04 September 2018 - 08:33 PM.
Posted 09 January 2013 - 10:44 AM
Do I count wrongly?
[…] does not require a second NCA object
Caution: There are two different methods for early exposure stated in Canada's and US' guidances. The one you use here is none of them. ;-)
[ul][li]The Canadian one requires truncation at the subject's tmax-values of the reference – not independent from formulation like in you worklow.[/li][li]FDA's requires truncation at the median of tmax-values of the reference.[/li][/ul]
Posted 09 January 2013 - 01:01 PM
Happy New Year Helmut,
You're correct the attached project had two NCAs, (since i just added to Ana's original project), however only the first was used to feed into "Join to extract AUCtmax from Summary table" where it takes the ACTUAL Tmax for that profile and returns the corresponding AUC to that time point.
Good tips those regarding regulatory requests for these calculations - both of them would be better suited to a modification of Ana's method (Data Wizard and second linked NCA) in my opinion, although you could just enter the values obtained as requested partial areas in the first NCA (downside of the latter being that the workflow is not automatically refreshing if data changes.
Data wizard filter to just get reference, then use that value for test and reference.
FDA's requires truncation at the median of tmax-values of the reference.
Data wizard filter to just get reference, then desc stats to get median and finally use that value for test and reference.
Posted 10 January 2013 - 02:28 AM
Simon, I like your solution better.
Jim, you might want to use Simon's join statement instead as it only necessitates one NCA.
Yes, but no agency would accept that. I built a WF based on yours – for HPB and FDA (attached).
First run a simple NCA and request only tmax – no need to filter other metrics afterwards.
In the respective 2nd NCAs you can include all the usual stuff, fit the elimination, blah blah.
Hint: If a truncation time point does not exactly equal a measured one (time deviations or a consequence of FDA's median method) PHX interpolates lin/lin in increasing sections of the profile and lin/log in decreasing ones (regardless which trapezoidal method is set). Might be a good idea to use lin-up/log-down trapezoidal for consistency.
BTW, made a funny discovery: If you don't state a unit of dose, no units of the other metrics are given.
[file name=pAUC_CAN_US.phxproj size=812031]http://www.pharsight.com/extranet/media/kunena/attachments/legacy/files/pAUC_CAN_US.phxproj[/file]
P.S.: Is it possible to move the thread to the PHX/WNL category?
Edited by Simon Davis, 05 September 2018 - 05:29 PM.
0 user(s) are reading this topic
0 members, 0 guests, 0 anonymous users