ABE1580
☆    

India,
2012-10-09 17:42
(4188 d 05:46 ago)

Posting: # 9358
Views: 9,936
 

 Failure of BE in fasting studies [Study As­sess­ment]

Dear All

In our BE studies we are finding that most of the times fed studies are passing BE in comparison with fasting studies. What could be the probable reason of such observations? Is it formulation that is playing a role or variable behavior of drug moleule in fasting state or study design? What are the factors to be considered in such cases for fasting study design?


Regards


Edit: Category changed. [Helmut]
drgunasakaran1
★★  
avatar

2012-10-09 22:07
(4188 d 01:21 ago)

@ ABE1580
Posting: # 9364
Views: 9,013
 

 Failure of BE in fasting studies

Dear Mr ABE,
It may be due to the variable absorption of the drug under fasting conditions. Sometimes, food enhances the absorption of the molecule and its pharmacokinetic parameters will show less variations under fed conditions. In these cases, the formulation will pass under fed conditions, but will fail in fasting study.
In these cases, it is better to do a pilot study in Fasting condition first, if the literature shows high variability and lesser absorption under fasting condition. If the formulation passes in Fasting condition, then it has higher chances of passing under Fed condition.
Say for example, if the formulation shows highly variable nature under fasting conditions, doing a routine two period, two way, cross over design is not going to help. You need to change your design to a Replicate design so that the formulation passes under fasting condition too.
In the above scenario, I will suggest you to go for Replicate design for Fasting study and Two period, Two way, Cross over under Fed conditions.

Dr S Gunasakaran MBBS MD
Disclaimer: The replies/posts are my personal opinions and it does not represent my company views on the same.
cakhatri
★    

India,
2012-10-10 09:34
(4187 d 13:54 ago)

@ drgunasakaran1
Posting: # 9369
Views: 8,996
 

 Failure of BE in fasting studies

Dear Dr.Gunasakaran,

Can you confirm if the regulatory will accept two different designs for fast and fed study, to the best of my understanding both the studies have to be of the same design. In case of replicate design, the criteria for establishing BE will differ based on the ISCV obtained from reference.

Regards
Chirag
Dr_Dan
★★  

Germany,
2012-10-10 15:09
(4187 d 08:19 ago)

@ cakhatri
Posting: # 9374
Views: 9,047
 

 Failure of BE in fasting studies

Dear Chirag
why shouldn't regulatory authorities accept two different designs? Bioequivalent is bioequivalent regardless if this is demonstrated by replicate or two way cross-over design. The reason why two different designs (+ blood sampling time points) were choosen is easily explained. To be on the safe side you should use the same batches of test and reference in both studies conducted at the same CRO (one site and one analytical method).
Kind regards
Dan

Kind regards and have a nice day
Dr_Dan
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2012-10-10 15:48
(4187 d 07:40 ago)

@ cakhatri
Posting: # 9376
Views: 8,983
 

 Different designs: why not?

Dear Chirag,

agree with what Dan said above. We once performed a fasting study in a parallel design (long half life according to the literature). The long half life turned out to be nonsense. I guess the analytical method in the publication had some selectivity issues and measured parent + one minor metabolite. Consequently we performed the fed study in a cross-over. No problem.

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
cakhatri
★    

India,
2012-10-12 12:42
(4185 d 10:46 ago)

@ Helmut
Posting: # 9403
Views: 8,811
 

 Different designs: why not?

Dear Dr_Dan and Helmut,
Thanks for the confirmation, it helps....
Regard
Chirag
drgunasakaran1
★★  
avatar

2012-10-10 21:14
(4187 d 02:14 ago)

@ cakhatri
Posting: # 9380
Views: 8,918
 

 Failure of BE in fasting studies

Dear Dr Chirag,
I fully agree with answers of Dr_Dan and Mr Helmut. Regulatory will accept two different study designs for fasted and fed conditions.

Dr S Gunasakaran MBBS MD
Disclaimer: The replies/posts are my personal opinions and it does not represent my company views on the same.
ABE1580
☆    

India,
2012-10-11 08:10
(4186 d 15:18 ago)

@ drgunasakaran1
Posting: # 9381
Views: 8,869
 

 Failure of BE in fasting studies

Dear All,

Thank you for all your suggestions. What are the factors to be considered in designing the protocol for fasting study in above cases?


Regards,
drsinghs
☆    

India,
2012-10-11 11:28
(4186 d 12:00 ago)

@ drgunasakaran1
Posting: # 9382
Views: 8,892
 

 Failure of BE in fasting studies

❝ Say for example, if the formulation shows highly variable nature under fasting conditions, doing a routine two period, two way, cross over design is not going to help. You need to change your design to a Replicate design so that the formulation passes under fasting condition too.


Dear Dr. Gunasakaran,

Can you elaborate more on the above statement with some illustrations (If possible)?

Is it possible that a formulation will pass in replicate design and same will fail in 2 way crossover?

What impact will be on ratio and 90% CI, If I test same formulation in 2 way crossover and replicate design?

Regards,
drsinghs
Dr_Dan
★★  

Germany,
2012-10-11 13:43
(4186 d 09:45 ago)

@ drsinghs
Posting: # 9385
Views: 8,937
 

 Failure of BE in fasting studies

Dear drsinghs

❝ Is it possible that a formulation will pass in replicate design and same will fail in 2 way crossover?


❝ What impact will be on ratio and 90% CI, If I test same formulation in 2 way crossover and replicate design?


If the replicate design and the 2 way crossover have the same power and the same intra-subject variability then these two studies will have the same chance to pass (or fail). In your case (same formulations) it solely depends on sample size:
  1. Theta (%): the target ratio in average BA between the two formulations expressed in percentage of the average reference BA.
  2. Power (%): the least statistical power to detect (1-Power) differences between the Test and the Reference formulation.
  3. CV (%): the intra-subject coefficient of variation for crossover(or replicate crossover) study.
  4. Lower acceptance limit represents the lower BE criteria to be accepted;
    the upper acceptance limit is equal to 1/(lower acceptance limit). Usually the lower limit can be 80.000% or, 70.000% in some cases.
  Upper acceptance limit = 125 %
  Lower acceptance limit = 80 %
      Expected ratio T/R = 95.00 %
            Target power = 80.00 %
        Intra-subject CV = 20.0 %

  study     2x2x2        2x2x3       2x2x4
  design  crossover   replicated   replicated
 ------- ----------- ------------ ------------
    N        20          16           10

         Estimated power = 83.46802 %


For a given intra-subject CV of for example 20.0 % you need in a full replicate design 10 subjects, in a half replicate dasign 16 subjects and in a 2 way crossover design 20 subjects to show bioequivalence in the general acceptance range of 80 - 125% with a power of more than 80% each.
I hope this helps
Kind regards
Dan

Kind regards and have a nice day
Dr_Dan
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2012-10-11 15:05
(4186 d 08:23 ago)

@ Dr_Dan
Posting: # 9389
Views: 8,900
 

 1–β not necessarily = Δ

Dear Dan!

❝ 2. Power (%): the least statistical power to detect (1-Power) differences between the Test and the Reference formulation.


Not necessarily. The regulatory BE limits in the general case are based on a 20% difference Δ [where BElower = 1–Δ and BEupper=(1–Δ)-1], which is not related to the producer’s risk (β). The fact that with 80% power both the producer’s risk and the difference to detect are 20% is just coincidental.

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
d_labes
★★★

Berlin, Germany,
2012-10-11 15:10
(4186 d 08:18 ago)

@ Dr_Dan
Posting: # 9390
Views: 8,907
 

 Sample size

Dear Dr_Dan

  Upper acceptance limit = 125 %

❝   Lower acceptance limit = 80 %
❝       Expected ratio T/R = 95.00 %
❝             Target power = 80.00 %
❝         Intra-subject CV = 20.0 %


❝   study     2x2x2        2x2x3       2x2x4
❝   design  crossover   replicated   replicated
❝  ------- ----------- ------------ ------------
❝     N        20          16           10


❝          Estimated power = 83.46802 %


That looks like Bear if I'm right?
Bear has implemented only the power and sample size for the 2x2x2 crossover based on the non-central t-distribution approximation.
For the 3-period and 4-period replicate designs the rules n(2x2x3)=0.75*n(2x2x2) and n(2x2x4)=0.5*n(2x2x2) are used. But rules are rules only.

PowerTOST (exact via Owen's Q-function) gives you for your above settings:
Design   N    power
2x2x2   20  0.834680
2x2x3   14  0.817926
2x2x4   10  0.843312


Of course qualitatively and practically nearly the same.

Regards,

Detlew
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2012-10-11 19:53
(4186 d 03:35 ago)

@ d_labes
Posting: # 9394
Views: 8,904
 

 +partial replicate

Dear Detlew & Dan

❝ PowerTOST (exact via Owen's Q-function) gives you for your above settings:

Design   N    power

2x2x2   20  0.834680

2x2x3   14  0.817926

2x2x4   10  0.843312


Nitpicking: “2×2×3” is the full replicate 3-period design (TRT|RTR). The fashionable partial replicate (TRR|RTR|RRT) would give:
2x3x3   15  0.844011 – since the sample size has to be a multiple of 3.

Dif-tor heh smusma 🖖🏼 Довге життя Україна! [image]
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,636 registered users;
76 visitors (0 registered, 76 guests [including 7 identified bots]).
Forum time: 22:29 CET (Europe/Vienna)

Nothing shows a lack of mathematical education more
than an overly precise calculation.    Carl Friedrich Gauß

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5