10 replies to this topic

[Chetan Rathi]

Hello everyone,

I am interested in calculating PK-PD indices for 2 different dosing regimens.

a. 10mg/kg once daily

b. 5mg/kg two times daily ( total daily dose is 10mg/kg)

 

Steps followed:

1. I have the 24 hour data for 10mg/kg once daily study. I used the NCA model 220 for computing the indices.

2. Simulated the 5mg/kg twice daily using the PK model parameters as input from the 10mg/kg dose study to obtain the predicted conc.-time . Then performed the NCA model 220 for the predicted data.

 

My query is:

Time above the threshold in first case is 1.25 which looks fine if we see the predicted data.

But for the second case I am getting a value of 8.03 which does seems correct if we see the predicted conc.-time profile.

 

I am also attaching my file.

 

Any comments will be highly appreciated.

 

Thanks and regards,

Chetan [file name=QA_Indices.phxproj size=662277]http://pharsight.com/extranet/media/kunena/attachments/legacy/files/QA_Indices.phxproj[/file]

Attached Files

•  QA_Indices.phxproj   646.75KB   0 downloads

[Samer Mouksassi]

Hi Chetan,

Your simulation for BID might not be optimal as you only have one point after the second dose.

I am attaching a file showing how to do the simulation in a phoenix model as opposed to legacy PK model. Resutls are similar to what you got. The threshold seems very high and to have good target attainment you will have to increase the dose.

Phoenix models has better and expanded capabilities. You can easily include simulated concentrations at any timepoints you would like see how I did the table in the phoenix model.

 

 

Bests,

 

Samer [file name=NLMESIMULATION.phxproj size=1272445]http://pharsight.com/extranet/media/kunena/attachments/legacy/files/NLMESIMULATION.phxproj[/file]

Attached Files

•  NLMESIMULATION.phxproj   1.21MB   396 downloads

[Chetan Rathi]

Hi Samer:

 

I am not able to open the file. It gives me the following error:

 

The project you are trying to open was created with a newer version of Phoenix (6.3.0.394) and might not open correctly.

Would you like to try to open the project anyway?

 

Thanks,

Chetan

[Simon Davis]

Chetan,

   6.3 is the current version and can be downloaded freely by all currentl licensees from;

 

ftp://support_ftp:pt9JTmD9GdXQLtCgPkvL@ftp.certara.com/support/desktop/Phoenix/Phoenix_1.3.zip

 

I would suggest you speak with your internal IT contact to make sure he passes on information from Pharsight about these updates as this version was released in March of this year

 

Simon.

[Chetan Rathi]

Samer:

 

Thanks a lot for solving my problem. I installed Phoenix 6.3 and its working. But I have one more query:

The units are missing. I tried including the units in Phoenix model but still it shows no units in the NCA.

 

Please suggest.

 

Thanks,

Chetan

[Simon Davis]

Chetan, it's not too clear from your message but you may be dealing with the issued detailed in; Phoenix Product Notification 012, Unit conversion patch for Phoenix 1.3 on non-English OS.

This can be downloaded from the Support site together with the patch, it takes just a few seconds to apply. The document contents are reproduced below;

The following defect has been identified in Phoenix®1.3 for some users that do not use English (United States) Regional Options. It does not affect prior versions of Phoenix.
Valid units not recognized with some Regional Options that are non-US (QC 11920/CRM 136733):On machines where non-US Regional Options are used, this problem can be observed when importing a dataset with units from any source (PMO, CSV, DAT, XLS, *.PHXPROJ, etc.), or when specifyingunits on an already imported dataset. The worksheet headers of the data will get curly braces “{ }” around valid units,
indicating that the units are invalid. Since these units are marked as invalid, unit
conversions and relabeling cannot be done. The problem occurs in different locales
where NaN (i.e., “Not a Number”, shown as a blank cell in a worksheet) is represented as
more than one word. For example, in French NaN is represented by “Non Numerique”.
One workaround is to change the Regional Options to English (United States). However,
if the user wishes to use Regional Options other than English (United States), Pharsight is
providing a patch in conjunction with this Product Notification to resolve this problem.
To correct this issue, a new file “unitconverter.exe” can be downloaded using the link
below:
ftp://support_ftp:pt9JTmD9GdXQLtCgPkvL@ftp.certara.com/support/desktop/Phoenix/unitconverter.zip
Please download and unzip this file, then copy it over the existing one in
\application\Services\Data, e.g., C:\Program
Files\Pharsight\Phoenix\application\Services\Data.

[Chetan Rathi]

Simon:

 

I am using Windows 7 English version OS. Therefore, I think the unit conversion pack is not a problem because in other models it is displaying the units. I have also attached the image.

 

Thanks,

Chetan [file name=NCAPD_file.doc size=220160]http://pharsight.com/extranet/media/kunena/attachments/legacy/files/NCAPD_file.doc[/file]

Attached Files

•  NCAPD_file.doc   215KB   0 downloads

[Simon Davis]

Chetan,
the message "Insufficient units" appears when one or more of the input columns you have mapped to the model are missing units, take a look at your source data file and check there

Simon.

Attached Thumbnails

• 

[Chetan Rathi]

Hi,

 

Just like the previous case, I have to evaluate the PK/PD indices of another study in which the route of administration is subcutaneous. Model 7 of PK model gave the best fit.

 

My queries are:

 

1. Will it be appropriate to fit Model 7 (IV bolus as input) to a data for which the route of administration is subcutaneous instead of IV bolus?

2. While simulating the conc.-time profile using Phoenix model, what should I choose for Absorption (Intravenous or extra vascular)?

 

Please find my Phoenix file as attachment.

 

Thanks,

Chetan [file name=QA_pharsight_2.phxproj size=256594]http://www.pharsight.com/extranet/media/kunena/attachments/legacy/files/QA_pharsight_2.phxproj[/file]

Attached Files

•  QA_pharsight_2.phxproj   250.58KB   0 downloads

[Chetan Rathi]

Hello Samer,

 

I want to include the confidence interval across the conc.-time profile in the XY plot.

 

Please can you guide me how to do it.

 

Please find my Phoenix file (NLMESIMULATION) in earlier email of this thread.

 

Thanks and regards,

Chetan

[Samer Mouksassi]

Hi Chetan,

The first step is to compute summary statistics using the descriptive stats tool.

Make sure to use time as a sort variable so stats are computed by time.

Then you can send the results to a plot choosing the standard errors as the error bars.

 

If you need the confidence interval as opposite to just showing the SE this can be computed using the percentiles of your data ( provided you have lot of data) or using the normal approximation i.e. z alpha/2 * SE = + - 1.96 *SE

 

Hope this helps.

 

Samer