Back to the Top
Hi,
Would request if someone could explain whether PKPD and/or PBPK
modeling could be used to predict fate of test drug product with
reference in bioequivalence studies.
I did some preliminary search on the subject but it appears that
these modeling techniques are more suitable for new chemical entities
(NCEs) rather than drug products already available in the market,
hence may have limited use in bioequivalence setup.
Would you agree with the above statement?
Are there some predictive modeling techniques available to test
bioequivalence in-silico to have improved (rather positive) BE study
outcome?.
Conventionally, dissolution profile of test versus reference drug
product is the only tool available to take a go/no go decision with a
typical BE study. Not to mention that in case a BE study fails, it
becomes very very difficult to find the root cause of failure. It is
indeed a very very expensive way of learning!!!
Would request comments/suggestions from the group on the above.
Thanks
Alp
Back to the Top
The following message was posted to: PharmPK
Dear Alp,
A frequent use of GastroPlus(tm) is to conduct virtual bioequivalence
trials. These are stochastic simulations that simulate a population of N
subjects for both test and reference products. Our customary approach
is to
run three virtual trials for each (3 test and 3 reference) at both
fasted
and fed states. We do this because all virtual trials are random
processes,
so the distribution of plasma concentration-time curves within each
trial
will be different for each one (just as actual trials with the same
number
of different subjects would vary). In each set of three test trials
one will
have the highest Cmax and AUC, one will have the lowest, and the
other will
be in between. The same will be true for the three reference trials.
(Again,
this is repeated for fasted and fed conditions.)
We then compare each of the test trials with each of the reference
trials
(for a given number of subjects) to estimate the likelihood of
demonstrating
bioequivalence with that number of subjects. For example, in a
particular
trial, the lowest result from the three fasted state test runs might be
bioequivalent to the lowest of the reference fasted state trials, but
might
not be for the other two trials. Across the matrix of 9 comparisons,
you can
see the number of positive (bioequivalent) and negative results. This
process is repeated for a variety of numbers of subjects to determine
the
number of subjects that would likely be required to demonstrate
bioequivalence (if any). For example, at 20 subjects, perhaps only 3
of 9
would show bioequivalence. At 50 subjects, perhaps 7 of 9 would show
bioequivalence. At 100 subjects, perhaps all 9 of 9 would show
bioequivalence.
Or, perhaps there is little or no predicted bioequivalence at any
number of
subjects - the formulation is simply not going to be successful.
The advantage of this approach is that it can include all of the
complexities in oral absorption - saturable absorption, saturable
metabolism, formulation effects, gastric emptying effects, fasted and
fed
states, and so on. For approved products, the amount of data
available for
such modeling is usually very good compared to NCE's.
In recent studies for customers, this method has show that planned
studies
had low probability of success, and our customers were able to design
new
formulations that were predicted to have high likelihood of success
with a
reasonable number of subjects. Obviously, the time and cost savings from
avoiding failed trials is significant.
Walt Woltosz
Chairman & CEO
Simulations Plus, Inc. (AMEX: SLP)
42505 10th Street West
Lancaster, CA 93534-7059
U.S.A.
http://www.simulations-plus.com
Phone: (661) 723-7723
FAX: (661) 723-5524
E-mail: walt.-a-.simulations-plus.com
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 "Predictive tools in bioequivalence studies" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)