Yura
★    

Belarus,
2016-01-13 07:51
(2998 d 04:21 ago)

Posting: # 15817
Views: 5,902
 

 Immunoglobulins, normal human, for intravascular [Power / Sample Size]

Hi
How to calculate the sample size for the non-comparative trials for biosimilar, or what are the approaches to solving this problem?
Aim - assessment of treatment effectiveness (improvement in symptoms).
Best regards,
Yura
BE-proff
●    

2016-01-14 12:15
(2996 d 23:56 ago)

@ Yura
Posting: # 15826
Views: 4,848
 

 Immunoglobulins, normal human, for intravascular

Hi Yura,

IMHO, the best option is to involve qualified statistician who has an experience in such calculations.
Dr_Dan
★★  

Germany,
2016-01-14 14:37
(2996 d 21:34 ago)

@ Yura
Posting: # 15827
Views: 4,783
 

 Immunoglobulins, normal human, for intravascular

Dear Yura
In a non-controlled (non-comparative) study no control/comparative group (in which subjects are given comparator drug) is used. Is it this what you mean? Pilot study for Proof-of-Concept to evaluate if some Parameters change?
This would be an exploratory study and no formal sample size calculation would be needed. If you want to assess treatment efficacy you need to compare to something (placebo, comparator drug or no treatment at all) and for such a (confirmatory) study you can make a sample size calculation based on the desired outcome (non-inferiority, equivalence or superiority). Please advise.
Kind regards
Dr_Dan

Kind regards and have a nice day
Dr_Dan
Yura
★    

Belarus,
2016-01-15 09:53
(2996 d 02:18 ago)

@ Yura
Posting: # 15828
Views: 4,769
 

 Immunoglobulins, normal human, for intravascular

Hi
There are data in which the effectiveness of the similar biological product up to 80%. Can we compare the test biological product with these data and how should be the size of the sample?
Best regards,
Yura
Dr_Dan
★★  

Germany,
2016-01-18 18:17
(2992 d 17:54 ago)

@ Yura
Posting: # 15838
Views: 4,572
 

 Immunoglobulins, normal human, for intravascular

Dear Yura
Sorry, but with such sparse information it is not possible to help.

In general you will need at least the following Information:
  • Probability of not getting a Type II error (not getting a false negative)
  • Probability of a Type I error (false positive)
  • Expected difference
  • Standard Deviation
Kind regards
Dr_Dan

Kind regards and have a nice day
Dr_Dan
UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,636 registered users;
105 visitors (0 registered, 105 guests [including 3 identified bots]).
Forum time: 12:12 CET (Europe/Vienna)

With four parameters I can fit an elephant,
and with five I can make him wiggle his trunk.    John von Neumann

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