Back to the Top
The following message was posted to: PharmPK
Happy New Year.
I would be grateful if some one could assist me by explaining how to
calculate the sample size for BE study.
I have conducted a preliminary bioavailabilty study for loratadine on
6 healthy volunteers. From the study, I am able to calculate the
variance. How should I calculate the sample size ? What are the inputs
? What equation to be used ?
Appreciate for your reply. Thank you very much.
Regards,
Peh
Lecturer
School of Pharmaceutical Sciences,
Universiti Sains Malaysia,
11800 Penang,
Malaysia
Back to the Top
The following message was posted to: PharmPK
Dear Peh,
> [...] some one could assist me by explaining how to calculate the
> sample size for BE study.
>
Well, based on the point estimate and CVintra (if your pilot was a
crossover) or CVtotal (parallel).
> I have conducted a preliminary bioavailabilty study for loratadine on
> 6 healthy volunteers. From the study, I am able to calculate the
> variance. How should I calculate the sample size? What are the inputs?
> What equation to be used ?
You cannot calculate the sample size directly (i.e., there's no
formula). You only may calculate power (=1-beta) for a given combination
of acceptance range, alpha, CV, expected PE and given sample size. In an
iterative procedure the sample size is increased until power >= desired
(let's say 80%).
See some posts (including links and code for free software) here:
http://forum.bebac.at/mix.php?category=1
Some R code here:
http://forum.bebac.at/mix.php?category=19
Your sample size was *very* low... So keep in mind, that *both* the PE
as well as the CV are uncertain - not carved in stone. Don't get trapped
by simply using these numbers as they are and feeding some software. ;-)
It's possible to calculate an upper confidence interval of the CV based
on the Chi-distribution (but don't be shocked by the outcome for n=6).
Also allow for a safety margin of the PE in the pivotal BE study.
In the literature for loratadine CVintra of 30%-50% (Cmax) are reported,
so be warned again not to rely on your pilot study's data *alone*. A low
CV value and a 'nice' PE may be pure chance.
ICH E9
http://www.ich.org/LOB/media/MEDIA485.pdf
calls for a sensitivity analysis in study planning.
Best regards,
Helmut
-
Ing. Helmut Schuetz
BEBAC - Consultancy Services for
Bioequivalence and Bioavailability Studies
Neubaugasse 36/11
1070 Vienna, Austria
|
Want to post a follow-up message on this topic?
If this link does not work with your browser send a follow-up message to PharmPK@boomer.org with "Sample size for BE study" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)