Mikkabel
☆    

Belgium,
2017-02-16 13:18
(2596 d 20:52 ago)

Posting: # 17068
Views: 7,337
 

 Replicate or not? [Design Issues]

Hello,

I am new as user of the forum (but not as a viewer) and it is my first question, so please be undestanding with me.....;-)

My company plan to perform a 4-arm PK study with two arms for two different doses of the test formulations and two arms for two different batches of the reference treatment.

Even if two different batches of the reference treatment will be used, could you please advice me if I can considered this design as a replicate design?
Furthermore, on a statistical point of view, is it correct to use the within-subject variability of the reference obtained with this design to widen the acceptance criteria for Cmax?

thank you in advance for your response,
have a nice day!
DavidManteigas
★    

Portugal,
2017-02-16 14:04
(2596 d 20:07 ago)

@ Mikkabel
Posting: # 17069
Views: 6,629
 

 Replicate or not?

Hi Mikkabel

To be replicated, the reference product must be administered to the same subject at least 2 times. Is that the case in your study? Not sure if I understood the study design... why do you use 4 arms, being 2 of them with the reference product?

Regards,
David
Mikkabel
☆    

Belgium,
2017-02-16 14:50
(2596 d 19:21 ago)

@ DavidManteigas
Posting: # 17071
Views: 6,675
 

 Replicate or not?

Hi David,

Thank you for your quick reply!

I know that to be replicated the reference product must be administered to the same subject at least two times, but my question was to know if it is mandatory to use the same batch of the reference product or if it allowed to use two different batches of the reference product.

Indeed, we propose to use two different batches of the reference products because we observed in previous PK studies that the reference product is very variable from batch to batch.

I hope it is more clear. ;-)


Edit: Full quote removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post #5! [Helmut]
ElMaestro
★★★

Denmark,
2017-02-16 14:48
(2596 d 19:22 ago)

@ Mikkabel
Posting: # 17070
Views: 6,724
 

 Replicate or not?

Hi Mikkabel,

❝ My company plan to perform a 4-arm PK study with two arms for two different doses of the test formulations and two arms for two different batches of the reference treatment.


It is not an optimal design; if you want to scale the acceptance range, then it is potentially tricky that you are using different batches of the reference product. It may -in the eyes of a regulator- be seen as a design feature intended to inflate the swr and force the CV upwards for scaling.
Your design is one often chosen to companies a bit new in BE, perhaps more well versed with innovator development or formulation technologies, and who wish to squeeze a lot info out of a single trial. It is too optimistic, but you can't readily find info about it in the public domain.

If you are new to BE then be aware that these aspects about scaling are some of the most heavily debated parts of the latest guideline revision. Therefore I don't recommend you to gamble that an expert of sorts can make the trouble disappear if a regulator starts asking questions.

Have a good day.

Pass or fail!
ElMaestro
nobody
nothing

2017-02-16 15:10
(2596 d 19:00 ago)

@ ElMaestro
Posting: # 17072
Views: 6,675
 

 Replicate or not?

Hi!

As the master of all masters indicated: The approach you are planning is everything but conservative considering intraindividual CV. I would never do something like this without Sci Adv from relevant (!) authority/ies... ;-)

Kindest regards, nobody
Mikkabel
☆    

Belgium,
2017-02-16 15:12
(2596 d 18:58 ago)

@ ElMaestro
Posting: # 17073
Views: 6,569
 

 Replicate or not?

Hi El maestro,

Thank you for your clear explanation!

So if I understand well, the best way to justify the scaling is to perform a "classical" replicate design i.e using the same batch administered two times to the same subject and I agree with it to avoid some bias in the statistical analysis (in particular within subject variability forced).

Furthermore, as I mentioned in a previous post the rationale to use two different batches of the reference product is to demonstrate that the reference product is variable from batch to batch and maybe not BE between them. The idea was to take two batches presenting two different profile in-vitro (opposite profiles but still within the specifications) resulting in two different PK profile. What do you think about this method?

Thank you and have a good day¨!


Edit: Full quote removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post #5! [Helmut]
ElMaestro
★★★

Denmark,
2017-02-16 15:24
(2596 d 18:46 ago)

@ Mikkabel
Posting: # 17074
Views: 6,593
 

 Replicate or not?

Hi Mikkabel,

❝ Furthermore, as I mentioned in a previous post the rationale to use two different batches of the reference product is to demonstrate that the reference product is variable from batch to batch and maybe not BE between them. The idea was to take two batches presenting two different profile in-vitro (opposite profiles but still within the specifications) resulting in two different PK profile. What do you think about this method?


Yes, this is a sneaky attempt, one that I have mainly seen with innovator companies going generic when pipelines dry up. "Let us show the regulators that two Ref batches exists which are hardly BE to each other, let's demonstrate how wrong the requirements are, then they will agree with us and relax their requirements for our product", that kind of thing.

Tested and tried, and not a generally healthy way forward. I am not saying it will never work, but I am saying it does not generally work. A famous publication not too long ago was one by Elise B Getz and colleagues from Oriel who went that way with DPI's containing Fp and Sx. This isn't for the fainthearted.

I would either give the regulators what they ask for, or fold my cards if that's not possible. At least some SA's are called for. Depending on your choices you risk getting stuck exactly where 20 other generic companies are currently stuck. Be very, very careful now.

Pass or fail!
ElMaestro
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2017-02-16 15:50
(2596 d 18:20 ago)

@ ElMaestro
Posting: # 17075
Views: 6,675
 

 Can of worms

Hi ElMaestro & Mikkabel,

❝ I am not saying it will never work, but I am saying it does not generally work. A famous publication not too long ago was one by Elise B Getz and colleagues from Oriel who went that way with DPI's containing Fp and Sx. This isn't for the fainthearted.


❝ Be very, very careful now.


Exactly. You should not try to open the can of worms.*


  • Burmeister Getz E, Carroll KJ, Jones B, Benet LZ. Batch-to-Batch Pharmacokinetic Variability Confounds Current Bioequivalence Regulations: A Dry Powder Inhaler Randomized Clinical Trial. Clin Pharm Ther. 2016;100(3):223–231. doi:10.1002/cpt.373. [image] free resource

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
Mikkabel
☆    

Belgium,
2017-02-16 16:45
(2596 d 17:25 ago)

@ ElMaestro
Posting: # 17076
Views: 6,676
 

 Replicate or not?

Thank you again for your feedback and for express your point of view!
Interestingly you mentioned the publication from Elise B Getz for the Fp/Sx DPI combination that I have recently read. In the same way, there is the publication of Bude/Formo from Orion pharma!


Edit: Full quote removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post #5! [Helmut]
nobody
nothing

2017-02-16 19:05
(2596 d 15:05 ago)

@ Mikkabel
Posting: # 17077
Views: 6,619
 

 Replicate or not?


Kindest regards, nobody
UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,640 registered users;
87 visitors (0 registered, 87 guests [including 6 identified bots]).
Forum time: 10:11 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