Back to the Top
Dear all,
I am seeking information about how to conduct a bioequivalence
study using parallel design instead of the cross-over design. Thanks for
your help!
Nasir Idkaidek,Ph.D.
College of Pharmacy, Irbid, JORDAN.
dekaidek.-a-.just.edu.jo
>>CAUTION: PLEASE NOTE MY CORRECT E-MAIL ADDRESS BELOW!!
******************************************************
Nasir M. Idkaidek, Ph.D.
College of Pharmacy
JUST, Irbid, Jordan.
Tel. 962-2-295111 ext 3542
Fax. 962-2-295019
E-mail: dekaidek.at.just.edu.jo
http://www.geocities.com/Athens/Academy/4418/
Back to the Top
Conducting a parallel design is probably easier tha a cross-over design
because you dont have to worry about drop-outs after the first period :-)
The problem is convincing the statisticians that subjects are different from
occasion to occasion and that the assumption of increased statistical power
from within-subject correlation may not be be as true as is often promoted.
I would suggest that the ideal way to design such a study and convince
everyone that the parallel design is sufficiently powerful is to simulate
the trial design under various assumptions of between and within subject
variability. If you have real data from earlier studies which allows you to
estimate the between and within subject variability then you should be able
to make a strong case for either the crossover or paralle design. If the
designs are approximately the same in terms of power then choose the one
that is easier to do (see above).
--
Nick Holford, Center for Drug Development Science
Georgetown University, 3900 Reservoir Rd NW, DC 20007-2197
email:n.holford.-at-.auckland.ac.nz tel:(202)687-1618 fax:687-0193
http://www.phm.auckland.ac.nz/Staff/NHolford/nholford.htm
Back to the Top
You may want to consider such design for drugs with very long half-life
which is very difficult with cross-over design due to drop-out.
You probably need more subjects since the variance is composed of the
between and within subject variability, and the between subject
variability is usually larger than within.
You probably want to select an homogeneous population (i.e., healthy male
or female but not both, with normal hepatic, renal function). If the drug
being studied is subject to polymorphism, you probably to want to phenotype
to exclude abnormal patients (e.g., 2D6 deficient).
The analysis of data is usually done using one way ANOVA with treatment as
the only factor after log transformation of the parameters. If parameters
are not normally distributed, you could perform non-parametric analysis.
Eric Masson, Pharm.D.
Scientific Director
Anapharm Inc.
2050, boul. Ren=E9-L=E9vesque Ouest,
5e =E9tage,
Sainte-Foy, QC,
G1V-2K8
Tel: (418) 527-4000
FAX: (418) 527-3456
Email: emasson.aaa.anapharm.com
Back to the Top
The FDA has giuidelines about what they expect for bioequivalence studies.
try the website.
Back to the Top
Hi Niser
FDA requires that the ANOVA should be performed to separate the
effect of Sequence (group), subject effect, period effect, and
treatments. In your case, you have 2 groups on the same day,
however, you should look for the effect of the group and subjects
within group.
I do know if the FDA accept this design, if you want to conduct this
study for your own interest, large number of subjects needed based on
variability of this drug.
Please contact me if I can be any help.
Please say Hi to Dr. Bassam Tashtosh in your college
Nabil B. Darwazeh, Ph.D.
Wyeth-Ayerst Research
401 N. Middletown Rd
205/314
Pearl River, NY 10965
darwazn.aaa.war.wyeth.com
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)