mittyri
★★  

Russia,
2014-12-21 19:04
(3384 d 16:52 ago)

Posting: # 14127
Views: 9,369
 

 Different strengths – different ISV? [Power / Sample Size]

Dear All,

Are there any "rules of thumb" about the ISV prediction for a study with a new strength?
For exapmle, the formulation 50mg is registered and we're going to conduct the study with the formulation 100 mg.
The drug has linear PK to 120 mg according to the literature.
Could we plan our study with ISV derived from the previous data? Is it safe enough?
Thanks in advance!

Kind regards,
Mittyri
ElMaestro
★★★

Denmark,
2014-12-21 21:42
(3384 d 14:15 ago)

@ mittyri
Posting: # 14129
Views: 7,727
 

 Different strengths – different ISV?

Hi again mittyri,

❝ For exapmle, the formulation 50mg is registered and we're going to conduct the study with the formulation 100 mg.

❝ The drug has linear PK to 120 mg according to the literature.

❝ Could we plan our study with ISV derived from the previous data? Is it safe enough?


A difficult topic. I am aware of people have a theoretical concern about ISV varying with strength but I don't remember ever having seen solid data quantifying the phenomenon.
I think I would just use the ISV from the previous study if that's available.

Pass or fail!
ElMaestro
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2014-12-22 04:07
(3384 d 07:50 ago)

@ ElMaestro
Posting: # 14134
Views: 7,801
 

 Different strengths – different ISV?

Hi ElMaestro & Mittyri,

❝ I am aware of people have a theoretical concern about ISV varying with strength…


Are these the ones wearing waist-belts plus suspenders to exercise caution?

❝ … but I don't remember ever having seen solid data quantifying the phenomenon.


So do I.

❝ I think I would just use the ISV from the previous study if that's available.


Yep – with one possible exception. If you enter nonlinear PK (high doses: saturation of enzymes and/or transporters) genetics and day-to-day variability may become increasingly important. Add some % to the %. Don’t ask me how much. If in doubt, run a pilot.

@Mittyri: I would not worry in your case.

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
mittyri
★★  

Russia,
2014-12-22 21:33
(3383 d 14:24 ago)

@ Helmut
Posting: # 14140
Views: 7,641
 

 Different strengths – thanks for advice

Dear Helmut and ElMaestro,

Many thanks for useful advice!

Kind regards,
Mittyri
mittyri
★★  

Russia,
2015-02-11 20:23
(3332 d 15:34 ago)

@ Helmut
Posting: # 14407
Views: 7,091
 

 Planning the study based on the pre­vi­ous results

Hi ElMaestro & Helmut,

I would like to continue the discussion on the practical level.
So we conducted the BEQ study on the Test product (50 mg).
Our results for Cmax were:
90%CI=[0.847; 1.134] CV=38%, PE=0.98, N=36
Does it mean that with a probability 90% the real GMR of Cmax lies between 0.847 and 1.134? Does it mean that the mentioned probability normally distributed in this interval?

We're planning to perform the second study with the same API, but another strength (100mg).
Which GMR would you propose? 0.98% (optimistic from previous study)? 0.95% (standard)?

Kind regards,
Mittyri
ElMaestro
★★★

Denmark,
2015-02-11 21:29
(3332 d 14:27 ago)

@ mittyri
Posting: # 14408
Views: 7,255
 

 Planning the study based on the pre­vi­ous results

Hi Mittyri,

❝ Our results for Cmax were:

❝ 90%CI=[0.847; 1.134]

❝ CV=38%, PE=0.98, N=36

❝ Does it mean that with a probability 90% the real GMR of Cmax lies between 0.847 and 1.134? Does it mean that the mentioned probability normally distributed in this interval?


It means that if the assumptions are correct (the residual is normal) and if the true GMR is 0.98 and if the CV is 38% then there is a 90% chance that the estimated GMR will be between your observed limits when you conduct a study on 36 subjects.

❝ We're planning to perform the second study with the same API, but another strength (100mg).

❝ Which GMR would you propose? 0.98% (optimistic from previous study)? 0.95% (standard)?


If you can afford it, it might be a good idea to go even a little lower. After all, this is a new formulation, right?
There are no rules, and you'll be going by gut feeling. If you intend the next study to be pivotal then the choice of GMR guess should reflect a ton of factors including biology, dissolution, gut feeling, Saturn's moons, your biorhythms, personal horoscope (you've gotta ask yourself one question: "Do I feel lucky?" Well, do ya, punk?) etc but not your management's desire to save money.

Pass or fail!
ElMaestro
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2015-02-12 14:31
(3331 d 21:26 ago)

@ mittyri
Posting: # 14411
Views: 7,182
 

 Conservatism

Hi Mittyri,

❝ Does it mean that with a probability 90% the real GMR of Cmax lies between 0.847 and 1.134?


Yes.

❝ Does it mean that the mentioned probability normally distributed in this interval?


No. These are backtransformed values. They are normal distributed with a mean of ln(0.98) = -0.0202 and 90% limits of ln(0.847, 1.134) = -0.1661, +0.1258. These limits are symmetrical around ln(PE): (-0.1661+0.1258)/2 = -0.0202.

❝ Which GMR would you propose? 0.98% (optimistic from previous study)? 0.95% (standard)?


First of all I agree with what ElMaestro wrote. As you know I’m not a friend of post hoc power, but in the previous study with 61% you were lucky to show BE. You are dealing with a HVD / HVDP (CV 38%). For these drugs the PE “jumps around” between studies. Therefore, the two Lászlós recommended in their paper about sample size estimation a conservative T/R of 0.90 – even if you have ob­served a “better” one in a previous study.

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, 1,639 registered users;
76 visitors (1 registered, 75 guests [including 9 identified bots]).
Forum time: 11:57 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