Back to the Top
I am seeking opinion from my colleagues as to what is the method of establishing steady state conditions from 3 morning trough plasma concentrations taken in a following clinical multiple rising dose study:
1 Full PK profile of the test drug is done on Day 1 from 0 to 72 hours.
2. Once a day dosing of the drug is done from Day 3 to Day 10 (ie for 7 days)
3. Morning plasma trough concentrations are taken on Days 7, 8 and 9.
4. Full PK profile of the test drug is again taken on Day 10 from 0 to 72 hours.
5 PK parameters obtained on Day 10 are compared to those obtained on Day 1.
6. How can one be sure that the steady state conditions were established by Day 10 by analysis of morning trough concentrations on Days 7, 8 and 9. I am particularly interested in different methods and approaches used.
7. By the way the test drug has a terminal t1/2 of about 12 hours but I do not want to use drug dosing for 5 half lives as a proof of steady state establishment.
Thank you all for your input and consideration.
Aziz Karim
Back to the Top
The following message was posted to: PharmPK
Hi Karim,
The easiest way would probably be to compare the AUC0-tau on Day 10 to the AUC0-infinity on Day 1. If at steady state, the two values should be similar.
Another way would be to estimate what your expected trough concentrations would be at steady state and see if your observed concentrations are at that range. The best way to do that would be to build a compartmental model and simulate the dosing schedule.
Toufigh
Toufigh Gordi, PhD
Clinical Pharmacology, PK/PD analysis consultant
www.tgordi.com
E-mail: tg.-at-.tgordi.com
Back to the Top
The following message was posted to: PharmPK
Dear Aziz,
A feature of PK Solutions (an easy to use Excel-based PK analysis
program) is to predict and graph multiple dose kinetics based on single
dose kinetics. Thus, PK Solutions can analyze the plasma profile obtained
on Day 1 from 0 to 72 hrs to determine the PK parameters of the first dose
and use these to calculate and graph the multiple dose projection through
steady state. This will allow you to visually and numerically observe the
trough concentrations as they rise to the steady state condition from
Day 1 to 10, and hence, resolve your concern (item 6).
The single dose to multiple dose projection feature of PK Solutions is
described and illustrated in the Tour of the program features available at
http://www.SummitPK.com/pksolutions/pksolutions.htm
Best Regards,
Dr. David S. Farrier
Summit Research Services
www.SummitPK.com
DFarrier.-a-.SummitPK.com
Back to the Top
The following message was posted to: PharmPK
Dear Dr. Aziz,
As you have done once a day dosing of the drug from Day 3 to Day 10 that means you have trough concentration for all the seven days (day-3 to day-10).
Now to know whether steady state has been reached or not you can perform linear regression on at least last three or four trough concentration values and calculate slope of the linear regression line. Then you have to test
H0: slope = 0 vs. H1: Slope <> 0. If H0 is accepted i.e. slope is statistically not significantly different from 0 then you can conclude that steady state has reached otherwise you have to continue the dosing in order to reach the steady state.
Thanks & Regards,
Ravi Shankar Pandey
Back to the Top
The following message was posted to: PharmPK
Ravi wrote:
> Now to know whether steady state has been reached or not you can perform linear regression on at least last three or four trough concentration values and calculate slope of the linear regression line. Then you have to test
> H0: slope = 0 vs. H1: Slope <> 0. If H0 is accepted i.e. slope is statistically not significantly different from 0 then you can conclude that steady state has reached otherwise you have to continue the dosing in order to reach the steady state.
This is a statistician's approach to the problem. It is doomed to failure because it does not recognize the underlying mechanism. Steady state is an asymptotic concept. You should describe the data with a PK model that recognizes the asymptotic nature of steady state. One can then define a criterion based on say the half-life for being close enough to steady state for the purpose of the study.
Nick
-- Nick Holford, Professor Clinical Pharmacology
Dept Pharmacology & Clinical Pharmacology
University of Auckland,85 Park Rd,Private Bag 92019,Auckland,New Zealand
Back to the Top
The following message was posted to: PharmPK
Dear Ravi,
You wrote to Aziz:
> Now to know whether steady state has been reached or not you can
>perform linear regression on at least last three or four trough
>concentration values and calculate slope of the linear regression
>line. Then you have to test
> H0: slope = 0 vs. H1: Slope <> 0. If H0 is accepted i.e. slope is
>statistically not significantly different from 0 then you can
>conclude that steady state has reached otherwise you have to continue
>the dosing in order to reach the steady state.
I don't think this is a valid test, and your interpretation of the
statistical inference is incorrect. If the slope is not significantly
different from 0, H0 cannot be rejected. However, this does not imply
that H0 is correct, and you cannot conclude that steady state has
reached.
To illustrate this: in the case of large variability between
measurements, you would conclude that steady state is reached, even
after a few doses. On the other hand, in the case of low variability,
you might conclude that concentration still increases, slightly, but
statistically significance. This illustrates that this test is not
valid.
best regards,
Hans Proost
Johannes H. Proost
Dept. of Pharmacokinetics, Toxicology and Targeting
University Centre for Pharmacy
Antonius Deusinglaan 1
9713 AV Groningen, The Netherlands
Back to the Top
The following message was posted to: PharmPK
Thank you Dr Proost for your thoughtful comments. I brought this subject to the attention of PK forum because we routinely do multiple rising dose study in drug development using the study design outlined by me, yet there is a diverse opinion on how to test for establishment of steady state conditions.
With your wide experience in PK analysis, what are your thoughts on how to test for establishment of steady state conditions in a multiple rising dose studies.
With best wishes and regards,
Aziz
Back to the Top
The following message was posted to: PharmPK
Dear Dr. Holford and Dr. Proost,
Please have a look at Page no. 17 (lines 11 & 12 of first Paragraph) in
the link below
http://www.fda.gov/downloads/AnimalVeterinary/GuidanceComplianceEnforcemen t/GuidanceforIndustry/UCM052363.pdf
The guideline in the above link quotes that "To determine a steady state
concentration, the CMIN values should be regressed over time and the
resultant slope should be tested for its difference from zero."
You can also check this link
http://onbiostatistics.blogspot.com/2009/05/pharmacokinetics-steady-state- analysis.html
I hope this will clear my point. Please correct me if I am wrong.
Thanks & Regards,
Ravi Shankar Pandey
Assistant Biostatistician
Back to the Top
Ravi,
The FDA is not always a reliable source of scientific information. There are some good scientists there, some of whom I know very well. But there are also those who do not know what they are doing and especially those who apply statistics to pharmacology without understanding pharmacology. If you have are forced to follow the FDA rules and regulations because of your job then so be it. But I prefer to discuss these issues openly without the conflict of commercial interest pressures that often contaminate discussion of FDA utterances.
Pharmacokineticists understand the principles involved in describing the time course of concentrations and the approach to steady state. These principles are based on physical and chemical concepts. They are not based on statistics. You need to understand the principles first before trying to answer questions about pharmacokinetic steady state.
Nick
-- Nick Holford, Professor Clinical Pharmacology
Dept Pharmacology & Clinical Pharmacology
University of Auckland,85 Park Rd,Private Bag 92019,Auckland,New Zealand
Back to the Top
The following message was posted to: PharmPK
Dear Ravi,
Thank you for sending the links to the FDA guidelines that recommend the
procedure you proposed. Your interpretation of these guidelines is correct.
>The guideline in the above link quotes that "To determine a
>steady state concentration, the CMIN values should be
>regressed over time and the resultant slope should be tested
>for its difference from zero."
In my opinion, this procedure is incorrect. It is simply the wrong
statistics. This would indeed be the right test to demonstrate whether
steady state has NOT been reached. But I presume that nobody is interested
in that test.
Again, if the null hypothesis cannot be rejected, this does not imply that
it is true.
Please note the similarity with bioequivalence testing, where the null
hypothesis is that the products are not equivalent; if the null hypothesis
is rejected, bioequivalence is concluded. But the statistical procedure is
more complicated, and not applicable to the assessment of steady state.
I fully agree with Nick Holford:
>Steady state is an asymptotic concept. You should describe the data
>with a PK model that recognizes the asymptotic nature of steady
>state. One can then define a criterion based on say the half-life for
>being close enough to steady state for the purpose of the study.
best regards,
Hans Proost
Johannes H. Proost
Dept. of Pharmacokinetics, Toxicology and Targeting
University Centre for Pharmacy
Antonius Deusinglaan 1
9713 AV Groningen, The Netherlands
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 "Establishment of steady state conditions" as the subject | Support PharmPK by using the |
Copyright 1995-2011 David W. A. Bourne (david@boomer.org)