balakotu
★    

India,
2015-08-20 11:34
(3143 d 14:14 ago)

Posting: # 15312
Views: 3,632
 

 Sample size for Clinical endpoint BE studies [Regulatives / Guidelines]

Dear All,

Anyone please help me on calculation of sample size for clinical endpoint bioequivalence studies.

In PK endpoint Bioequivalence studies standard parameters required for Sample sizes calculations are ratios, alpha, intra/inter CV.

What are the common parameters required for clinical endpoint bioequivalence studies (With respect to mean & proportions)?

Is there any SAS Program or any related software to construct the sample size for clinical endpoint bioequivalence studies?

Regards
Kotu
d_labes
★★★

Berlin, Germany,
2015-08-20 12:51
(3143 d 12:58 ago)

@ balakotu
Posting: # 15314
Views: 3,115
 

 Sample size for Clinical endpoint BE studies

Dear Kotu,

that's a very broad field and can't be answered in short and generally.

The parameters needed for sample size estimation depend on the type of clinical endpoint you intend to use.
For metric endpoints you need an estimate or educated guess of the means and variability.
Moreover you need an equivalence margin, i.e. a measure how big the means may deviate but are considered yet as equivalent. Complicated, I promise you. See EMA guideline on non-inferiority margin f.i.
Next you need a decision how to test equivalence. Via difference in means or via ratio of means.

If you aim to test equivalence via ratios think about the presumed distribution of your clinical endpoint. Usually they are not evaluated after log-transformation (log-normal distribution not assumed), as is the standard for most PK endpoints, but assumed as normal-distributed untransformed.
Please refer to:

Hauschke, Steinijans, Pigeot
Bioequivalence Studies in Drug Development
Wiley & Sons, Chichester 2007
Chapter 10 "Equivalence assessment for clinical endpoints"


Sample size estimation for this evaluation may be done via R-package PowerTOST, functions power.RatioF() and sampleN.RatioF().

For binary endpoints (proportions) you need an estimate or educated guess of the proportions under Test or Reference treatment. And a choice of which equivalence test to use, equivalence of proportion ratios, proportion difference or such wired things like odds ratio. There are numerous variants of tests out there.
Moreover you need a criterion how big a difference is considered yet as equivalent, the equivalence margin.

Other types of clinical endpoints are for instance time-to-event data or ordinal categories. They require other distinct types of power/sample size calculations.
That all goes beyond the scope of this forum, I think.

Refer to

Chow, Shao, Wang
"Sample Size Calculations in Clinical Research"
Second edition
Chapman & Hall/CRC
Boca Raton, London, New York 2008


to get an impression.

If you look for a SAS solution for sample size estimation:
Proc Power is your friend. Consult the help pages for that procedure.
But be warned. Proc Power is a complicated friend :cool:.
Especially if it comes to cross-over studies and/or equivalence of proportions.

Regards,

Detlew
nobody
nothing

2015-08-20 16:37
(3143 d 09:11 ago)

@ d_labes
Posting: # 15315
Views: 2,969
 

 Sample size for Clinical endpoint BE studies

...and don't forget to validate your clinical parameter regarding sensitivity to detect relevant (what ever that is) differences between formulations tested.

Good luck! :-)

Kindest regards, nobody
UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,636 registered users;
88 visitors (0 registered, 88 guests [including 12 identified bots]).
Forum time: 00:49 CET (Europe/Vienna)

Nothing shows a lack of mathematical education more
than an overly precise calculation.    Carl Friedrich Gauß

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5