alejandro_jo1990
☆    

Spain,
2022-09-01 16:40
(574 d 04:24 ago)

Posting: # 23252
Views: 2,576
 

 N calculation for trial about PK superiority [Power / Sample Size]

Hello to all! first thanks for this wonderful forum, here a young MD, clinical pharmacologist recently moving from clinical trials conduct to design, tryinto apply my (modest) knowledge of clinical PK and biostatistics to phase I trial design. I was discussing with a former colleague. If we were to design a trial to try to support a change in SmPC for a certain product, ie. a faster absorption:

- If we were to demostrate in overall a faster absorption, we would need to design a superiority trial and calculate N based on expected diff in logCmax using logsd (difference or more conservative the higer Cmax maybe?). Suggested log to use a diff in means to calculate N and Cmax as is the parameter to include both extentand and rate of absorption (I deem Tmax as useless).

-If we were to demonstrate a faster absorption at X point (10 minutes), before Tmax, shall I need to use the differences in logAUC with same SD assessment? or concentration at 10 minutes would suffice (or maybe any approach is not correct at all).

I assume I would calculate N based on the difference in means as in any study.

PS. I swear I conducted an intensive research through the forum

BR
Alex
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2022-09-01 17:02
(574 d 04:01 ago)

@ alejandro_jo1990
Posting: # 23253
Views: 2,326
 

 Bracketing approach

Hi Alex,

❝ Hello to all! first thanks for this wonderful forum, here a young MD, clinical pharmacologist recently moving from clinical trials conduct to design, tryinto apply my (modest) knowledge of clinical PK and biostatistics to phase I trial design.


Welcome to the club!

❝ I was discussing with a former colleague. If we were to design a trial to try to support a change in SmPC for a certain product, ie. a faster absorption:


❝ - If we were to demostrate in overall a faster absorption, we would need to design a superiority trial and calculate N based on expected diff in logCmax using logsd (difference or more conservative the higer Cmax maybe?). Suggested log to use a diff in means to calculate N and Cmax as is the parameter to include both extentand and rate of absorption (I deem Tmax as useless).


Since both formulations are IR, only the rate of absorption changes. The extent of absorption should not be affected (unless you have pre-systemic metabolism and run into saturation with the ‘faster’ formulation).

❝ -If we were to demonstrate a faster absorption at X point (10 minutes), before Tmax, shall I need to use the differences in logAUC with same SD assessment?


In principle, yes.

❝ or concentration at 10 minutes would suffice (or maybe any approach is not correct at all).


Not sure, what you mean. If you want to avoid a clinical trial and bridge to the safety/efficacy data of the ‘old’ formulation, consider bracketing. That means, non-superiority for Cmax and equivalence for AUC.

❝ I assume I would calculate N based on the difference in means as in any study.


Yes.

❝ PS. I swear I conducted an intensive research through the forum


Although that’s rare, I believe it. :-D

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
alejandro_jo1990
☆    

Spain,
2022-09-01 18:18
(574 d 02:46 ago)

@ Helmut
Posting: # 23255
Views: 2,233
 

 SmPC change is the whole thing

»Dear Helmut, vielen dank for the fast answer (I truly miss my times as assistenzarzt in München)

I will try to give a broader scope. It's, at the and, a discussion between my colleagues from medical affairs versus clinical pharmacology/regulatory
A formulation is commercialized and appears to be absorbed faster than the free substance (its a lysinate as I recall). That is known from in vitro release, chemical data and some bad studies or maybe because their parents told them that (so SmPC gives the same clinical pharmacology properties in PK as the free substance).

Medical Affairs position: if wanted to include some marketing in the package to say that absorption is faster than other substance. So, if we design a trial and demosntrate superiority of therapeutic effect at 10 minutes we are good (not probably aware of the monstruous N, but ok).

R&D: A therapeutic trial only lets your marketing go in that direction. You cannot use a therapeutic action to support PK changes in the clinical pharmacology section of SmPC. You need a Phase I trial. If we want faster absorption in overall, use a T test to calculate an N for the differences in Cmax as expected or after a pilot (assuming it falls into the therapeutic range) or if you specifically wan it, you might need to demonstrate faster absorption at than point (ie. 10m). There the question: if we aimed to design it (technical stuff, I'm aware of the terrible N, especially as the clinical setting would call for a parallel design), would we need to adress superiority of the PK endpoint (logAUC10 or logConc a 10m???) uing t-test.

❝ Not sure, what you mean. If you want to avoid a clinical trial and bridge to the safety/efficacy data of the ‘old’ formulation


The point is (above) to demonstrate faster abosrption in overall or at acertain point. Your statement makes me raise some questions:
1. Being an IR, hence should not eiste the risk of dose dumping, do we also need to demonstrate BE for AUC (extent?) or only superiority of PK is ok?
2. If we fall into the therapeutic range (if it's wide we can assume that, what if its narrow?), can we still use Cmax and assume an increase for calculations? Or even if it's at at ceratin point, the Cmax will be increased accordingly, so all my doubts apply again.
4. So what if the therapeutic range is narrow?
5. If all is wrong, how do you design a trial to demonstrate faster absorption in overall or at a certain point? I've seen many FIH seeing that just because it's seen, but to design specifically for that??


PS Im truly aware of the whole thing being 0 clinically relevant, but I see it as an opportunity to learn.

Sorry for the long long post,
Alex
dshah
★★  

India/United Kingdom,
2022-09-02 13:52
(573 d 07:12 ago)

@ alejandro_jo1990
Posting: # 23262
Views: 2,149
 

 N calculation for trial about PK superiority

Hi Alex!

I am wondering why PK superiority? I am not getting if you do not prove faster onset of action with PD superiority, you can have advantage. Already the discussion for Tmax assessment (EMA PSG) is discussed in forum. Could you provide more details for not focusing on PD superiority?
Regards,
Divyen Shah
alejandro_jo1990
☆    

Spain,
2022-09-02 20:57
(573 d 00:06 ago)

@ dshah
Posting: # 23263
Views: 2,182
 

 The efocus is faster absorption

❝ I am wondering why PK superiority? I am not getting if you do not prove faster onset of action with PD superiority, you can have advantage. Already the discussion for Tmax assessment (EMA PSG) is discussed in forum. Could you provide more details for not focusing on PD superiority?


Hi Diven thanks for the answer!
If you have the link to the thread reagarding the Tmax discussion, could you paste it here please? Regarding the issue, the want to demonstrate faster absorption not faster onset of action, therefore the whole discussion about if AUC, therapeutic ranges etc
BR
Alex
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2022-09-03 02:46
(572 d 18:18 ago)

@ alejandro_jo1990
Posting: # 23264
Views: 2,132
 

 The efocus is faster absorption

Hi Alex,

❝ If you have the link to the thread reagarding the Tmax discussion, could you paste it here please? Regarding the issue, the want to demonstrate faster absorption not faster onset of action, therefore the whole discussion about if AUC, therapeutic ranges etc


Preliminary stuff in this thread. An introduction in this presentation and more detailed in this article.
Instead of equivalence you are interested in a one-sided test.

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
alejandro_jo1990
☆    

Spain,
2022-09-06 02:26
(569 d 18:38 ago)

@ Helmut
Posting: # 23271
Views: 2,056
 

 Non-parametric N

Dear Helmut thanks for answering

❝ Instead of equivalence you are interested in a one-sided test.


So we would calculate our N based on a non-parametric test for T-max differences (I have heard some epidemologists saying its ok to calculate with T-Test and increase like 15%)
I swear its thing that would have come to my mind was using Tmax.
Kind of related, I guess no concern exists about the almost sure increase in Cmax if it doesnt go beyond the therapeutic range inst it?


Edit 2022-09-06 15:56:
I apologize and reply myself/give a maybe more accurate answer after reading again the article. Not possible to calculate N for Tmax but to perform certain simulations with some parameters. So better to run a pilot without and seein what comes? if you are right Tmax will come sooner (with a higher Cmax). Would it support a SmPC change?


Edit: Merged with a later (now deleted) post. You can edit your posts for 24 hours. [Helmut]
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2022-09-06 19:02
(569 d 02:02 ago)

@ alejandro_jo1990
Posting: # 23274
Views: 2,070
 

 Non-parametric N

Hi Alex,

❝ So we would calculate our N based on a non-parametric test for T-max differences (I have heard some epidemologists saying its ok to calculate with T-Test and increase like 15%)


That’s black magick. Why 15%?
  • If (if!) the distribution is normal, the Wilcoxon signed-rank test (for paired differences) and the Mann–Whit­ney U test (for independent samples) have an asymptotical power of \(\small{3/\pi\approx95.5\%}\).
  • If the distribution is not normal (e.g., tmax), power of nonparametric tests are higher than the ones of the corresponding t-tests.

❝ Kind of related, I guess no concern exists about the almost sure increase in Cmax if it doesnt go beyond the therapeutic range inst it?


No idea. In the EMA’s bracketing approach you perform a Non-Superiority test with an upper margin of 1.25. All of this stuff (including BE) is not related to the therapeutic range at all.

❝ Not possible to calculate N for Tmax but to perform certain simulations with some parameters.


Even then it’s not an easy job.

❝ So better to run a pilot without and seein what comes?


Yes.

❝ if you are right Tmax will come sooner (with a higher Cmax).


Likely. But as I wrote before: Under the premise of no saturable pre-systemic metabolism.

❝ Would it support a SmPC change?


No idea. Once you have the result of a pilot, consider a scientific advice. Sweden?

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
alejandro_jo1990
☆    

Spain,
2022-09-15 16:56
(560 d 04:08 ago)

@ Helmut
Posting: # 23297
Views: 1,961
 

 Non-parametric N

Thanks for the answer! truly clarified :-)
UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,636 registered users;
91 visitors (0 registered, 91 guests [including 10 identified bots]).
Forum time: 20:04 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