Jump to content


Photo

Lack of IntersubjectCV for untransformed data in BE analysis

IntersubjectCV BE untransformed

  • Please log in to reply
16 replies to this topic

#1 ZhangYong

ZhangYong

    Newbie

  • Members
  • Pip
  • 8 posts
  • LocationNanjing, China

Posted 04 September 2017 - 09:46 AM

Dear Simon and all,

When I perform BE analysis using orignal PK data (none Ln transformed), I find that PHX does not output IntersubjectCV in PHX WNL 7.0
 
Can anyone help me to explain why WNL PHX doesn't calculate this? Since Inter CV is mathematically possible, I don't understand why PHX does not calculate it.
 
Be sure, analysis of BE using Original PK data without Ln()-transformation.
 
Thank you for your help and clarification. or will it be provided in future version?


#2 Simon Davis

Simon Davis

    Advanced Member

  • Administrators
  • 1,316 posts

Posted 04 September 2017 - 10:20 AM

Dear Zhang, Please can you expand a little on your study design, to whom you would like to submit and which BE model you are using?  Even better please post your project, Phoenix/WinNonlin’s default in BE is:

fixed:  Sequence+Formulation+Period
Random: Subject(Sequence)

 

In which case the intersubject CV% is found in Final Variance Parameters when log-transforming the data using the option on the Fixed Effects tab which is what the guidances expect.  Maybe I am missing something in your question ?

 Simon.

 



#3 Helmut Schütz

Helmut Schütz

    Advanced Member

  • Members
  • PipPipPip
  • 316 posts
  • LocationVienna, Austria

Posted 04 September 2017 - 11:41 AM

Hi Simon,

 


Maybe I am missing something in your question ?
 

 

You do. Not you fault. ;-)

 

@Zhang Yong. If you cross-post, please refer to the ongoing thread at the BEBA-Forum.

 


 Best regards,
Helmut

https://forum.bebac.at/


#4 Simon Davis

Simon Davis

    Advanced Member

  • Administrators
  • 1,316 posts

Posted 04 September 2017 - 11:55 AM

Thanks - that background helps  - if I summarise the thread correctly we don't currently have a published reference to calculate CV% in untransformed data?   If you do have one please let us know.

 

ALso we may need to consider reviewing this formula when using untransformed data.

» intrasubject CV = sqrt(Var(Residual)) / RefLSM

 

 Simon.



#5 Helmut Schütz

Helmut Schütz

    Advanced Member

  • Members
  • PipPipPip
  • 316 posts
  • LocationVienna, Austria

Posted 04 September 2017 - 12:21 PM

Hi Simon,

 

… we don't currently have a published reference to calculate CV% in untransformed data? If you do have one please let us know.

 

ALso we may need to consider reviewing this formula when using untransformed data.

» intrasubject CV = sqrt(Var(Residual)) / RefLSM

 

As Michael wrote, the formula is given as (34) on p. 85 of the PHX/WNL 7.0 User’s Guide. He assumed that it is referring to the book by Chow & Liu (which is given on p. 415). Can’t find the formula in there… I guess the correct reference is Phillips, which is also on p. 415. However, I think that it is wrong.

I would drop the “intra-subject CV” of untransformed data from the output in PHX.


 Best regards,
Helmut

https://forum.bebac.at/


#6 ZhangYong

ZhangYong

    Newbie

  • Members
  • Pip
  • 8 posts
  • LocationNanjing, China

Posted 04 September 2017 - 01:11 PM

Hi Simon,

 

 

You do. Not you fault. ;-)

 

@Zhang Yong. If you cross-post, please refer to the ongoing thread at the BEBA-Forum.

 

Dear Helmut,

Thanks for your citing giving our Simon the calrification of the background.



#7 ZhangYong

ZhangYong

    Newbie

  • Members
  • Pip
  • 8 posts
  • LocationNanjing, China

Posted 04 September 2017 - 01:35 PM

Thanks - that background helps  - if I summarise the thread correctly we don't currently have a published reference to calculate CV% in untransformed data?   If you do have one please let us know.

 

ALso we may need to consider reviewing this formula when using untransformed data.

» intrasubject CV = sqrt(Var(Residual)) / RefLSM

 

 Simon.

Dear Simon,

 

First question:

------------------------

Let look back to the currently used formula by PHX 7.0 (for untransfrormed data BE analysis)

 

intrasubject CV = sqrt(Var(Residual)) / RefLSM

Yes, I guess the above formula (for untransformed data) may be cited from Page 153 of Chow and Liu's book:

Design and Analysis of Bioavailability and Bioequivalence Studies, Third Edition
Only this book discussed deeply the BE analysis using untransformed data with Examples.
But the formula is lack of supporting from other published papers.
 
Anyway, PHX 7.0 is using this formula.
Whether it will be used in next version of not, let us look forward to the PHX 8.0, 9.0, 10.0 ...........
 
BTW, if you input the dataset from Chow and Liu's book for BE analysis without ln-transform, the above formual will get the CV=15.66% which is identical to that in the book.
 
Attached are the dataset from Chow and Liu's book, and the pages extracted from the book for your reference.
 
 

Second question:

------------------------

IntersubjectCV for BE analysis with untransformed data. If no reference can be cited, Will PHX calculate it in the next version?

I guess the answer is NO. ^_^

Attached Files



#8 ZhangYong

ZhangYong

    Newbie

  • Members
  • Pip
  • 8 posts
  • LocationNanjing, China

Posted 04 September 2017 - 01:42 PM

Hi Simon,

 

 

As Michael wrote, the formula is given as (34) on p. 85 of the PHX/WNL 7.0 User’s Guide. He assumed that it is referring to the book by Chow & Liu (which is given on p. 415). Can’t find the formula in there… I guess the correct reference is Phillips, which is also on p. 415. However, I think that it is wrong.

I would drop the “intra-subject CV” of untransformed data from the output in PHX.

Attached is the paper form Phillips, these is no formula for IntrasubjectCV or IntersubjectCV in this paper.

If the PHX cann't provide more references for "currently used formula" for IntrasubjectCV, maybe I will drop the “intra-subject CV” of untransformed data from the output in PHX too.

 

At this situation, I think PHX will not provide IntersubjectCV in the next version.

Attached Files



#9 mittyright

mittyright

    Advanced Member

  • Members
  • PipPipPip
  • 98 posts

Posted 04 September 2017 - 02:08 PM

Dear Zhang,

 

1.

you are completely right regarding your results, they are the same as in the book.

By the way the question is: could we blindly trust what Chow does say?

As I wrote before: try to substitute Test and Reference treatments in the Bioequivalence object interface and execute it. Your results will be different. But that's a nonsense! CV cannot change just because you reverse the order of the drugs!

 

2.

I think that's correct

No justification = no implementation

So as far as you (or someone else) can justify which formula is appropriate for interCV using untransformed data (using math or scientific links), Simon can ping some devs responsible for this.

BTW there are many things which cannot be done in current PHX release and they do not have so simple solution as just dividing one variable by another one

 

Bests,

Mittyright



#10 ZhangYong

ZhangYong

    Newbie

  • Members
  • Pip
  • 8 posts
  • LocationNanjing, China

Posted 04 September 2017 - 02:23 PM

Dear Zhang,

1.
you are completely right regarding your results, they are the same as in the book.
By the way the question is: could we blindly trust what Chow does say?
As I wrote before: try to substitute Test and Reference treatments in the Bioequivalence object interface and execute it. Your results will be different. But that's a nonsense! CV cannot change just because you reverse the order of the drugs!

2.
I think that's correct
No justification = no implementation
So as far as you (or someone else) can justify which formula is appropriate for interCV using untransformed data (using math or scientific links), Simon can ping some devs responsible for this.
BTW there are many things which cannot be done in current PHX release and they do not have so simple solution as just dividing one variable by another one

Bests,
Mittyright

Dear Mittyri, Happy to meet you here.
Yes. I agree with you. PHX is a tool, not a statistician or a biostatistician. It's clever for PHX not involved into an ambiguous issue.

Edited by ZhangYong, 04 September 2017 - 02:30 PM.


#11 ZhangYong

ZhangYong

    Newbie

  • Members
  • Pip
  • 8 posts
  • LocationNanjing, China

Posted 04 September 2017 - 02:24 PM

Dear Zhang,

1.
you are completely right regarding your results, they are the same as in the book.
By the way the question is: could we blindly trust what Chow does say?
As I wrote before: try to substitute Test and Reference treatments in the Bioequivalence object interface and execute it. Your results will be different. But that's a nonsense! CV cannot change just because you reverse the order of the drugs!

2.
I think that's correct
No justification = no implementation
So as far as you (or someone else) can justify which formula is appropriate for interCV using untransformed data (using math or scientific links), Simon can ping some devs responsible for this.
BTW there are many things which cannot be done in current PHX release and they do not have so simple solution as just dividing one variable by another one

Bests,
Mittyright

Dear Mittyri, Happy to meet you here.
Yes. I agree with you. PHX is a tool, not a statistician or a biostatistician. It's clever for PHX not involved into an ambiguous issue.

Dear Simon, sorry for replicate reply. Thank you for deleting one.

Edited by ZhangYong, 04 September 2017 - 02:30 PM.


#12 Helmut Schütz

Helmut Schütz

    Advanced Member

  • Members
  • PipPipPip
  • 316 posts
  • LocationVienna, Austria

Posted 04 September 2017 - 03:05 PM

Hi Yong,

 

Let look back to the currently used formula by PHX 7.0 (for untransfrormed data BE analysis)

 

intrasubject CV = sqrt(Var(Residual)) / RefLSM

Yes, I guess the above formula (for untransformed data) may be cited from Page 153 of Chow and Liu's book:

Design and Analysis of Bioavailability and Bioequivalence Studies, Third Edition

 

 

I missed this in Chow & Liu’s book. Not their only crime. Mr Chow joined the FDA recently. ;-)

 

See the attached project. Since the study was balanced, I could simply use the global mean.

CVintra (based on R): 0.1566

CVintra (based on T): 0.1611

CVintra (based on global mean): 0.1588

After log-transformation: 0.1947

 

Now look at the second workflow. I divided all T-data by 10. Now:

CVintra (based on R): 0.1688

CVintra (based on T): 1.7357

CVintra (based on global mean): 0.1711

After log-transformation: 0.1947

 

Since we are playing around with mean (which we shouldn’t) the CVs change. No surprise. On the other hand, the CV in log-scale is exactly the same.

Simply: If you want the CV for sample size estimation (what else?) the estimate will depend on the mean(s). That’s crap – unless you expect in the next study exactly the same T–R. Do you?

Attached Files


 Best regards,
Helmut

https://forum.bebac.at/


#13 ZhangYong

ZhangYong

    Newbie

  • Members
  • Pip
  • 8 posts
  • LocationNanjing, China

Posted 05 September 2017 - 12:38 AM

Hi Yong,

 

 

 

I missed this in Chow & Liu’s book. Not their only crime. Mr Chow joined the FDA recently. ;-)

 

See the attached project. Since the study was balanced, I could simply use the global mean.

CVintra (based on R): 0.1566

CVintra (based on T): 0.1611

CVintra (based on global mean): 0.1588

After log-transformation: 0.1947

 

Now look at the second workflow. I divided all T-data by 10. Now:

CVintra (based on R): 0.1688

CVintra (based on T): 1.7357

CVintra (based on global mean): 0.1711

After log-transformation: 0.1947

 

Since we are playing around with mean (which we shouldn’t) the CVs change. No surprise. On the other hand, the CV in log-scale is exactly the same.

Simply: If you want the CV for sample size estimation (what else?) the estimate will depend on the mean(s). That’s crap – unless you expect in the next study exactly the same T–R. Do you?

Dear Helmut,

Thanks for your kind calculation and demostration.

 

But my results are:

 

See the attached project. Since the study was balanced, I could simply use the global mean.

CVintra (based on R): 0.1566

CVintra (based on T): 0.1611

CVintra (based on global mean): 0.1588

After log-transformation: 0.1947

 

Now look at the second workflow. I divided all T-data by 10. Now:

CVintra (based on R): 0.1688

CVintra (based on T): 1.7357

CVintra (based on global mean): 0.3076  tihis is different from yours, would you like to confirm this?

After log-transformation: 0.1947

 

I like to calculate the parameters by hand (in MS Excel via Excel's VBA) to verify the formula used in PHX and other software. Just for my hobby.



#14 mittyright

mittyright

    Advanced Member

  • Members
  • PipPipPip
  • 98 posts

Posted 06 September 2017 - 09:56 PM

Dear Zhang,

 

just want to confirm your findings, there's a little mistake in data mapping in the project:

CVintra using global mean = sqrt(194.12374)/45.29 = 0.3076
I like to calculate the parameters by hand (in MS Excel via Excel's VBA) to verify the formula used in PHX and other software.

 

please look into R world. More fun. I promise.

 

Mittyright



#15 Helmut Schütz

Helmut Schütz

    Advanced Member

  • Members
  • PipPipPip
  • 316 posts
  • LocationVienna, Austria

Posted 06 September 2017 - 11:29 PM

Dear both,

 

just want to confirm your findings, there's a little mistake in data mapping in the project:

CVintra using global mean = sqrt(194.12374)/45.29 = 0.3076

 

I stand corrected.

 

 

please look into R world. More fun. I promise.

 

Yessir! Absolutely.


 Best regards,
Helmut

https://forum.bebac.at/


#16 ZhangYong

ZhangYong

    Newbie

  • Members
  • Pip
  • 8 posts
  • LocationNanjing, China

Posted 07 September 2017 - 04:42 AM

Dear Zhang,

 

just want to confirm your findings, there's a little mistake in data mapping in the project:

CVintra using global mean = sqrt(194.12374)/45.29 = 0.3076

please look into R world. More fun. I promise.

 

Mittyright

 

Thank you for your confirmation and recmmendation.


Edited by ZhangYong, 07 September 2017 - 05:06 AM.


#17 0521

0521

    Advanced Member

  • Members
  • PipPipPip
  • 46 posts

Posted 08 September 2017 - 05:31 AM

Dear Helmut,

Thanks for your kind calculation and demostration.

 

But my results are:

 

See the attached project. Since the study was balanced, I could simply use the global mean.

CVintra (based on R): 0.1566

CVintra (based on T): 0.1611

CVintra (based on global mean): 0.1588

After log-transformation: 0.1947

 

Now look at the second workflow. I divided all T-data by 10. Now:

CVintra (based on R): 0.1688

CVintra (based on T): 1.7357

CVintra (based on global mean): 0.3076  tihis is different from yours, would you like to confirm this?

After log-transformation: 0.1947

 

I like to calculate the parameters by hand (in MS Excel via Excel's VBA) to verify the formula used in PHX and other software. Just for my hobby.

Dear Zhang,
 
Happy to meet you here.
 
I see a lot of your wonderful Excel table in 'DXY'.
 
I have been puzzled by the nonparametric test of WinNonlin's "Crossover object". Can you use Excel to demonstrate how to reproduce WinNonlin's "Crossover object" function?
 
Thanks in advance.
0521






Also tagged with one or more of these keywords: IntersubjectCV, BE, untransformed

0 user(s) are reading this topic

0 members, 0 guests, 0 anonymous users