Back to the Top
Dear colleagues,
If I were using patients instead of healthy volunteers for
bioequivalence assessment, is still the classical +/-20% rule a
rational criterion for decision-making. I was thinking about the
greater between- (and also within-) subject variability from patients,
which will probably have a notable impact on statistical basis of the
comparison between both products (i.e., reference and test
formulation). That is, in order to avoid any problem in meeting
westlake's 90% confidence intervals may I acommodate this to a wider
range?. If so, how wide should be?. In short, does anyone of you know
of any other statistically-supported method for assessing
bioequivalence when patients have to be used?. Thanks in advance.
Regards,
Jorge Duconge
Jorge Duconge, PhD. MSC.
Center for Biological Evaluations & Researches,
Institute of Pharmacy and Foods
University of Havana, Cuba
Havana 36 CP 13600.
Tel. 537 271 9532
Fax 537 33 6811
e-mail:duconge.-at-.cieb.sld.cu
Back to the Top
Dear Jorge,
To my knowledge, the bioequivalence criteria do not change based on the
type of subjects used in the study.
radu
Back to the Top
Dear Jorge
I believe you have to do BE study in patients and can not do in normal
subjects
There are various options you can consider
Have any data on variability in patients or any justification for a
proposed variability in patients? Compare this variability with that
from normal volunteers? How different are patients likely to be
behaving from normal volunteers? here inter-subject variability will be
of prime concern? If wash out period is long then effect of disease
state would also come into picture, therefore keep wash out period as
short preferably even shorter than a week (take care of other issues
related to this like total blood loss, normality of patients in terms
of hemogram, fatigue etc.)
If it is possible to include more number of volunteers based on the
variability in patients - this is preferred approach and would face
less resistance from regulators.
If for any reason, like unacceptable number of patients, lack of
availability these many patients in reasonable period of time, or
ethics issues, you can not do study with higher no of patients then go
for a wider range for BE conclusion.
Such provision is already there in UK guidelines. If this study is to
be submitted to FDA then take opinion of BE division before commencing
study.
In addition to variability due to patients vis a vis normal volunteers
check if there are variables due to drug properties as well. These will
increase your burden on number of volunteers or the range for
acceptance.
In any case you cannot propose a range, which would be say 40 – 160 %.
Therefore you may have to use a combined approach go for a little wider
range and higher number of subjects than usual 24.
thanks and regards
Prashant
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)