Back to the Top
Hello,
May any one provide inputs about detecting a period affect and
sequence affect during a BE study.
the 'P' value obtained after ANOVA is less than 0.05(at 5% levelof
significance) for period affect.
after completion of BE study if the period affect was detected, what
impact it will have on the acceptance of the study by the regulatory
authorities.
In those cases do we need to provide any justification for having
period affect?
and in the similar manner, what impact it has on sequence affect and
what are the reasons for getting period and sequence affects.
Thanks in advance,
D.Chetia.
Back to the Top
The following message was posted to: PharmPK
Dear D.Chetia!
> [...] In those cases do we need to provide any justification for
having
> period affect?
No - unless you have to deal with a really stupid regulator - I
haven't seen
one - did you?
In a randomized cross-over design the treatment effect can be separated
from the period effect. Just give it a try: take the data from any
study,
multiple all values of the second period by 5 and run the BE-
calculation.
Both PE and the CI will be exactly the same.
Now for a 'gedankenexperiment': a 2x2 cross-over (RT/TR), test and
reference
of equal true bioavailability of 1 giving a PK response of 100 units
(left pannel). Due to some external influence (let's call it the
'phase of
the moon') the measured PK responses in the second period are just 50
units
(right pannel):
1 2 1 2
sequence 1: RT 100 100 sequence 1: RT 100 50
sequence 2: TR 100 100 sequence 2: TR 100 50
period mean 100 100 period mean 100 50
reference mean 100 reference mean 75
test mean 100 test mean 75
PE = T/R = 100/100 = 1 PE = T/R = 75/75 = 1
The treatment effect is not influenced by the period effect, and an
unbiased PE is obtained.
> [...] and in the similar manner, what impact it has on sequence
affect
> and what are the reasons for getting period and sequence affects.
A nuisance? A statistical artefact? A true sequence effect effect, a
subject-by-formulation interaction, an unequal carry-over, an error in
randomization (and a mixture of a all or some of them) in a 2x2 cross-
over
lead to a biased estimate of the teratment effect. The important point
is
that it's impossible to deal with a 'significant sequence effect' based
on the data from the study. You can only avoid it by a proper design
(e.g.,
not an endogenous compound, long enogh wash-out, no pre-dose levels in
period II). That's for the bad news.
No for the good news. If the your study was properly planned according
to
conditions given above, simply forget it. First of all forget about even
testing for it. Apart from theoretical considerations (see Stephen Senn
in the link below), it was shown in a large metastudy, that you will get
a significant result (if testing at a 10% level as suggested by
Grizzle in
1965) in roughly 10% of studies.
http://forum.bebac.at/mix_entry.php?id=1368
See also
http://www.jerrydallal.com/LHSP/crossovr.htm
especially Section 4.
The topic arises quite frequently, but we all should learn to forget
it. ;-)
For other threads see:
http://forum.bebac.at/search.php?search=sequence+effect&=Search&ao=phrase
Best regards,
Helmut
--
Ing. Helmut SchAtz
BEBAC - Consultancy Services for
Bioequivalence and Bioavailability Studies
Neubaugasse 36/11
1070 Vienna, Austria
e-mail helmut.schuetz.-at-.bebac.at
web http://bebac.at/
contact http://bebac.at/Contact.htm
forum http://forum.bebac.at
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 "Period and sequence affects" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)