Bioequivalence and Bioavailability Forum

rahul dixit ☆ 2011-01-03 16:17 (5222 d 03:08 ago) Posting: # 6388 Views: 19,576	Montelukast BE study [Design Issues] Post reply
	Hello Everyone Has any body worked on the Montelukast BE study. Please let me know the number of volunteers required and the variability observed with it. Regards Rahul Edit: Category changed. [Helmut]

bjkim97
☆

Korea / Seoul,
2011-01-06 07:45
(5219 d 11:41 ago)

@ rahul dixit
Posting: # 6399
Views: 18,196

Montelukast BE study

Post reply

Hi Ragul

I recommend the this web site
- Montelukast Bioequivalence

Many good Bioequivalence study report.

And this other I suggest to my opinion

A sample size of 26 completers (13 subjects per sequence) will be required to provide 80% power that the 90% confidence interval for the ratio of Test to Reference for AUClast will lie within the acceptance region of (80%, 125%) and 80% power that 90% confidence interval for the ratio of Test to Reference treatment for Cmax will lie within the acceptance region of (80%, 125%). Consequently, this study has at least 80% power overall to demonstrate bioequivalence of the Test treatment to the Reference treatment (i.e., equivalence in both AUClast and Cmax]. This estimate is based on the assumption that the true ratio between Test and Reference treatments for both AUClast and Cmax is 1.05. An intra-subject CV estimate of approximately 15.80% for AUC and an intra-subject CV estimate of approximately 23.90% for Cmax, were obtained from literature (Pharmaceutical Research. 2004 Sep; 21(9): 1539-1544) and used in this power calculation.

This sample-size calculation was performed using the PASS® software.

Assuming an approximate dropout rate of 20%, 6 subjects will be recruited into the study to ensure 32 completers.

This is my opinion... But you change the possible dropout rate 10%
If so assuming an approximate dropout rate of 10%, 4 subjects will be recruited into the study to ensure 30 completers.

I hope this helps
Kind regards

Byung-Ju Kim.
Bioequivalence Scientist
Tel: +82 2 317 2081 / +82 10 3955 1601

Edit: Full quote removed. Please delete anything from the text of the original poster which is not necessary in understanding your answer; see also this post! [Jaime]

d_labes
★★★

Berlin, Germany,
2011-01-07 14:51
(5218 d 04:34 ago)

@ bjkim97
Posting: # 6402
Views: 17,245

Opinions ...

Post reply

Dear Byung-Ju,

❝ And this other I suggest to my opinion

❝ ...

A man having his own opinion is always preferable :cool:

.
But as John Moore said: "Your opinion is your opinion, your perception is your perception - do not confuse them them with 'facts' or 'truth' ... " (the rest of this quote may be somewhat bothering).

You are totally right if the point estimator and the CV you have used in your sample size estimation (taken from a particular study - why that and not the PAR studies?) are the true values, are "carved in stone" - Helmut's favourite terminology.

Have a look at this thread for another strong opinion :-D

and go through the category "Sample Size" of this forum - discussions of sample size in connection with pilot study.
Eventually are some thoughts mentioned there worth considering them.

—
Regards,

Detlew

bjkim97
☆

Korea / Seoul,
2011-01-11 07:47
(5214 d 11:38 ago)

(edited on 2011-01-11 09:47)
@ d_labes
Posting: # 6408
Views: 17,015

Opinions ...

Post reply

Dear d_labes

Thanks your good advice... You take a carrot and stick approach. ;-)

I accept with your opinion. It was hasty of me to do such a thing

From now on I do carefully considered when expressed my opinion

(The boughs that bear most hang lowest. :cool:

)

Very very thanks...
Byung-Ju Kim

Edit: Full quote removed. Please delete anything from the text of the original poster which is not necessary in understanding your answer; see also this post! [Ohlbe]

joy_fm ☆ Indonesia, 2015-02-05 14:10 (3728 d 05:16 ago) @ bjkim97 Posting: # 14378 Views: 14,909	Montelukast BE study Post reply
	Hi Byung-Ju Kim, would you mind to send the journal publication as you mentioned above, since I can't access the link. Best regards joy_fm

Helmut
★★★

Vienna, Austria,
2015-02-05 14:24
(3728 d 05:01 ago)

@ joy_fm
Posting: # 14379
Views: 14,970

MHRA’s PARs

Post reply

Hi Joy,

❝ would you mind to send the journal publication as you mentioned above, since I can't access the link.

Byung-Ju was referring to MHRA’s public assessment reports – not a paper. The URL is not active any more. Try this search.

—
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

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

joy_fm ☆ Indonesia, 2015-02-10 18:38 (3723 d 00:47 ago) @ Helmut Posting: # 14398 Views: 14,682	MHRA’s PARs Post reply
	Dear Helmut, thank you for your information. I'm sorry, but I still can't find journal which states CV AUCt for Montelukast. Regards joy_fm

Helmut
★★★

Vienna, Austria,
2015-02-11 03:05
(3722 d 16:20 ago)

@ joy_fm
Posting: # 14399
Views: 14,791

CV from the CI

Post reply

Hi Joy,

❝ I'm sorry, but I still can't find journal which states CV AUCt for Montelukast.

Some people think to be very clever and don’t give the CV in the results. However, the CV can be calculated from the CI and sample size (formulas for 2×2 crossover studies).
Example from the first hit in : C_max 99.7% (89.0–110.4%), n 20 gives CV ~19.8%.

I suggest to get the free statistical software

and the package PowerTOST (you will need it sooner or later, believe me)… Once you installed it, try the nice function CVfromCI():

library(PowerTOST)

cat("CV =", round(CVfromCI(lower=0.890, upper=1.104, n=20)*100, 1), "\n")

—
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

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

d_labes
★★★

Berlin, Germany,
2015-02-11 10:16
(3722 d 09:09 ago)

@ Helmut
Posting: # 14400
Views: 14,615

CI2CV

Post reply

Dear Joy, dear Helmut

❝ Example from the first hit in : C_max 99.7% (89.0–110.4%), n 20 gives CV ~19.8%.

[peer_review] :cool:

"Abstract
... tablets in twenty-four healthy volunteers ..."

library(PowerTOST)

cat("CV =", round(CI2CV(lower=0.890, upper=1.104, n=24)*100, 1), "\n")

CV = 22

[/peer_review]

BTW: CI2CV() is the same function as CVfromCI() and was only "invented" because my brain was not able to remember which name was implemented by me.

—
Regards,

Detlew

Helmut
★★★

Vienna, Austria,
2015-02-11 14:01
(3722 d 05:24 ago)

@ d_labes
Posting: # 14402
Views: 14,597

Beyond the abstract

Post reply

Dear Detlew,

❝ [peer_review]

❝ "Abstract

❝ ... tablets in twenty-four healthy volunteers ..."

❝ [/peer_review]

[response_to_reviewer]

Although the study was planned for 24 subjects, the paper states on p.88:
- Two female volunteers did not show up for the study due to personal issues; so two male volunteers were retired to keep pair sequences and complete statistical treatment; without affecting the final result after having performed the corresponding randomization test. Thus; the study was performed on twenty subjects.
20 is also mentioned in Figure 2 and Tables 2&3 (p.89). Why two male subjects were “retired” (to keep balance?) – thus decreasing power – is beyond me.

[/response_to_reviewer] :smoke:

❝ BTW: CI2CV() is the same function as CVfromCI() and was only "invented" because my brain was not able to remember which name was implemented by me.

Understandable. I always use CI2CV() as well.

—
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

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

ElMaestro ★★★ Denmark, 2015-02-11 14:36 (3722 d 04:50 ago) @ Helmut Posting: # 14403 Views: 14,448	Beyond the abstract Post reply
	Hi Hotzi, ❝ Why two male subjects were “retired” (to keep balance?) – thus decreasing power – is beyond me. ❝ [/response_to_reviewer] [humor] Because they used Kinetica [/humor] — Pass or fail! ElMaestro

Helmut
★★★

Vienna, Austria,
2015-02-11 15:27
(3722 d 03:59 ago)

@ ElMaestro
Posting: # 14404
Views: 15,014

Beyond the abstract & Kinetica

Post reply

Hi ElMaestro,

❝ [humor]

❝ Because they used Kinetica :crying:

❝ [/humor]

Nope: “For the statistical analysis of data derived from this in vivo study; general parametric procedures for linear models (normal theory) were used and an analysis of variance (ANOVA) was performed using the statistical application Phoenix™ WinNonlin^®; version 6.3; 2013.”

BTW (Dec 24 2014, source)…

Kinetica 5.1 SP1 has been released
Thermo Fisher Scientific is very pleased to announce the release of Kinetica 5.1 SP1. Kinetica now supports unbalanced data sets for two period, two treatments, two sequences Bioequivalence studies. This issue was first reported in the Journal of the American Association of Pharmaceutical Scientists (AAPS) article entitled, "Reference Datasets for 2-Treatment, 2-Sequence, 2-Period Bioequivalence Studies. The article described the validation of software used to assess bioequivalence and reported that results from Kinetica software differed from other similar software when unbalanced datasets were used (results for balanced datasets were consistent across all of the software packages). Please find the […] issues fixed below:
Issue Description
Incorrect equations used for bioequivalence determination in unbalanced datasets – Point Estimate and 90% Upper and Lower limit calculation for Latin Square 2 Formulations statistical analysis.

Do you expect that Thermo fixed the bug for parallel designs as well? I don’t.

—
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

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

d_labes
★★★

Berlin, Germany,
2015-02-11 15:59
(3722 d 03:26 ago)

@ Helmut
Posting: # 14405
Views: 14,485

Beyond the abstract & Kinetica

Post reply

Dear Helmut,

@Abstract:
This is a case for showing the effectiveness of a 'peer' review :-D

.

@Kinetica:

❝ Kinetica 5.1 SP1 has been released

❝ ... This issue was first reported in the Journal of the American Association of Pharmaceutical Scientists (AAPS) article entitled, "Reference Datasets for 2-Treatment, 2-Sequence, 2-Period Bioequivalence Studies...

We know it better :not really:

. A lie is a lie is a lie ...

—
Regards,

Detlew

ElMaestro
★★★

Denmark,
2015-02-11 16:49
(3722 d 02:36 ago)

@ d_labes
Posting: # 14406
Views: 14,390

Partial credit

Post reply

Hi d_labes and Hötzi,

at least we should give TFS partial credit because they acknowledge the existence of the issue and they are doing something about it. They could have decided to keep silent about it and do nuffin. No big deal. After all, only 0.0000000000000000000000001 (plusminus 10) % of the studies submitted for approval are evaluated with Kinetica :-D

Life is beautiful.

—
Pass or fail!
ElMaestro

ioanam
★

Romania,
2016-01-22 15:17
(3377 d 04:08 ago)

@ ElMaestro
Posting: # 15845
Views: 11,214

Partial credit

Post reply

Dear ElMaestro, dear all

since I have the "privilege" to work sometimes with Kinetica, I obtained the new version 5.1 SP1 pack released by Thermo Fisher in 2014, where they fixed the statistical part for the unbalanced studies.

Compared to the previous version, they mention now at the end of analysis:
- ANOVA analysis: Point estimate and confidence interval derived from the least squares estimate (marginal means). - for a balanced study
and for incomplete block analysis: Point estimate and confidence interval derived from the arithmetic means.

Perhaps I am wrong, but calculation based by arithmetic means cannot give bias in such studies?

Wish you a happy weekend.

Ioana

ElMaestro
★★★

Denmark,
2016-01-22 15:58
(3377 d 03:27 ago)

@ ioanam
Posting: # 15846
Views: 11,164

Partial credit

Post reply

Hi Ioana,

❝ since I have the "privilege" to work sometimes with Kinetica, I obtained the new version 5.1 SP1 pack released by Thermo Fisher in 2014, where they fixed the statistical part for the unbalanced studies.

May I recommend to Format C:

❝ Compared to the previous version, they mention now at the end of analysis:

❝ - ANOVA analysis: Point estimate and confidence interval derived from the least squares estimate (marginal means). - for a balanced study

❝ and for incomplete block analysis: Point estimate and confidence interval derived from the arithmetic means.

❝

❝ Perhaps I am wrong, but calculation based by arithmetic means cannot give bias in such studies?

Yes, this sounds odd. Arithmetic means would work for parallel trials, but not necessarily for crossovers. It could be a misunderstanding or a matter of wording. Try and look up in the documentation what is going on - how it is calculated in practice, and please keep us posted :-)

—
Pass or fail!
ElMaestro

Helmut
★★★

Vienna, Austria,
2016-01-22 16:06
(3377 d 03:19 ago)

@ ElMaestro
Posting: # 15847
Views: 11,240

Partial credit

Post reply

Hi Ioana,

❝ ❝ since I have the "privilege" to work sometimes with Kinetica, I obtained the new version 5.1 SP1 pack released by Thermo Fisher in 2014, where they fixed the statistical part for the unbalanced studies.

❝

❝ May I recommend to Format C:

BTW, Kinetica 5.1 SP1 did not fix the bug for parallel studies with unequal group sizes. Wrong as it ever was.

—
Dif-tor heh smusma 🖖🏼 Довге життя Україна!
Helmut Schütz

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

Montelukast BE study [Design Issues]

Montelukast BE study

Opinions ...

Opinions ...

Montelukast BE study

MHRA’s PARs

MHRA’s PARs

CV from the CI

CI2CV

Beyond the abstract

Beyond the abstract

Beyond the abstract & Kinetica

Beyond the abstract & Kinetica

Partial credit

Partial credit

Partial credit

Partial credit