Back to the Top
Dear All,
We have two Test formulation (T1 and T2) and Two Reference
formulations (R1 and R2). We want to test for Bioequivalence of each
test formulation against both the reference formulation i.e. T1 vs.
R1, T1 vs. R2, T2 vs. R1 and T2 vs. R2. Please suggest me the
appropriate study design which can be used to do all the comparisons
in one study. Please also provide some details regarding the
statistical consideration and references.
Waiting for your suggestions.
Thanks & Regards,
Ravi Shankar Pandey
Back to the Top
Dear Ravi,
Go for 4 period 4-seq 4-way cross over so that all the sub will get
all the treatment and you specify clearly in the protocol itself the
comparison of interest of ur treatments.
T1 vs R1, T1 vs R2, T2 vs R1, T2 vs R2 one more advantage in this
design is you can compare ur test to T1 vs T2 or T2 vs T1.
rgrds,
venkatesan
Back to the Top
Dear Ravi,
The best design may be "balanced, randomised, four treatment, four
period, four sequence, four way cross over" design. T
Regards
C.Raghavender
Back to the Top
The following message was posted to: PharmPK
Dear Shanker,
Two BE studies with three way cross over design should suffice the
requirement
thanks
Back to the Top
The following message was posted to: PharmPK
Dear Mr.Ravi Shankar,
The appropriate study design is Balanced, randomized, four period, four
treatment, four sequence, four way cross over study. You can do
comparison of all the four products in one study itself by this design.
Regards,
Dr.S.Gunasakaran, MBBS, MD
Medical Affairs
Back to the Top
The following message was posted to: PharmPK
Hi Ravi,
Let's look at the two possibilities:
1) What I can suggest is to perform 2 different studies rather than one.
In which your 3 drugs should be R1, T1, T2 and in another study you
compare R2, T1 and T2.
2) You can also try 4 formulations at a time. However I am unsure what
sequences should be used and what sense it will make? I doubt it will
make any sense except for reduce sample size and cost.
Anyone of the groups feels that the design 2 would be more scientific.
Please let me know.
Regards,
Yousuf
Back to the Top
The following message was posted to: PharmPK
Dear Ravi,
Design
--
4-way crossover study with 4 sequences defined according to a Williams
design, for instance:
Seq#1: R1-R2-T1-T2
Seq#2: T1-R1-T2-R2
Seq#3: T2-T1-R2-R1
Seq#4: R2-T2-R1-T1
(Adapted from: Chow and Liu, Design and Analysis of Bioequivalence
Studies, Marcel Dekker Inc., 1992)
Statistical analysis
--
SAS code would be something like this:
PROC MIXED;
CLASSES SEQ SUBJ PER TRT;
MODEL Y = SEQ PER TRT;
RANDOM SUBJ(SEQ) / SUB=SUBJ;
ESTIMATE 'T1 vs. R1' TRT -1 0 1 0 / CL ALPHA=0.1;
ESTIMATE 'T2 vs. R1' TRT -1 0 0 1 / CL ALPHA=0.1;
ESTIMATE 'T1 vs. R2' TRT 0 -1 1 0 / CL ALPHA=0.1;
ESTIMATE 'T2 vs. R2' TRT 0 -1 0 1 / CL ALPHA=0.1;
Good luck,
Fabrice Nollevaux
www.sgs.com/cro
Back to the Top
The following message was posted to: PharmPK
Dear Yousuf,
Please find below the logic for preference for design 2 with four
formulations.
(1). 3 Treatment, 3 Period, 3 sequence crossover design is not variance
balanced.
(2). 3 Treatment, 3 Period, 6 sequence crossover design is variance
balance but its efficiency is (80.44, 44.44) which is less than the
efficiency (90.61, 62.50) of 4 Treatment, 4 period, 4 sequence crossover
design.
(3). Moreover cost and sample size for 4 Treatment, 4 period, 4 sequence
crossover design will be less as compared for 3 Treatment, 3 Period, 6
sequence crossover design.
Reference for Point (2). is Page 159 of "Design and Analysis of
Cross-Over trials" 2nd Edition by Byron Jones and Michael Kenward.
Thanks & Regards,
Ravi Shankar Pandey
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 "Four treatment study design for Bioequivalence" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)