Lara
☆    

Mexico,
2016-02-10 01:46
(2969 d 16:55 ago)

Posting: # 15966
Views: 24,648
 

 Sample Size estimation for Replicate Cross over studies FDA or EMA criteria [Power / Sample Size]

Hi,

I have review the paper:

Someswara Roa. K et al. "Sample Size Estimation for Highly variable drugs using reference scaled average bioequivalence criteria". International Journal of Recent Scientific Research Vol. 6 Issue 7, pp.5040-5045, July, 2015.

They are determined the optimal sample size for reference scaled average bioequivalence under criteria of FDA and EMA. The paper has several tables with the sample size according with FDA and EMA criteria for the designs partial replicate 2x2x3 and full replicate 2x2x4 and use the BE limit. However, they are using the estimation of the sample size for a simple crossover. I am not sure that this is correct starting with df=2*n-2 and I do not know when they are using the characteristics of the designs. Can somebody help me with that?

Thanks

Lara


Edit: Category changed. [Helmut]
d_labes
★★★

Berlin, Germany,
2016-02-10 09:34
(2969 d 09:08 ago)

@ Lara
Posting: # 15967
Views: 22,493
 

 Garbage

Dear Lara,

❝ Someswara Roa. K et al. "Sample Size Estimation for Highly variable drugs using reference scaled average bioequivalence criteria". International Journal of Recent Scientific Research Vol. 6 Issue 7, pp.5040-5045, July, 2015.


forget this "gem" of science literature. It's simply crap. Don't believe even in comma of this paper. Throw it in a high arc into your waste bin.

Use instead:
Laszlo Tothfalusi and Laszlo Endrenyi
"Sample Sizes for Designing Bioequivalence Studies for Highly Variable Drugs"
J Pharm Pharmaceut Sci (www.cspsCanada.org) 15(1) 73 - 84, 2011
Online here.

The described simulation approach to get power/sample size is implemented in the R-package PowerTOST.

Regards,

Detlew
nobody
nothing

2016-02-10 12:17
(2969 d 06:24 ago)

@ d_labes
Posting: # 15968
Views: 22,467
 

 Garbage

WOW, what a journal! Horticulture and Bioequivalence!

And they even invested the 50 bugs to obtain an "SJIF impact factor" of five point somethink... :-D

Kindest regards, nobody
d_labes
★★★

Berlin, Germany,
2016-02-10 14:28
(2969 d 04:14 ago)

@ nobody
Posting: # 15971
Views: 22,291
 

 Garbage

Nobody!

❝ And they even invested the 50 bugs to obtain an "SJIF impact factor" of five point somethink... :-D


IMHO bugs is the totally correct term :-D!

Regards,

Detlew
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2016-02-10 21:42
(2968 d 21:00 ago)

@ Lara
Posting: # 15972
Views: 22,535
 

 Ponzi scheme

Hi Lara,

the

❝ International Journal of Recent “Scientific Research”


is predatory. Pay the publishers and any copypasted bullshit gets online within days. Read this story.

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
Lara
☆    

Mexico,
2016-02-11 21:35
(2967 d 21:07 ago)

@ Helmut
Posting: # 15977
Views: 22,176
 

 Ponzi scheme

Thank you for your advise everyone. I appreciate a lot ;-)


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! [Helmut]
jag009
★★★

NJ,
2016-02-16 20:55
(2962 d 21:46 ago)

@ Helmut
Posting: # 16000
Views: 21,952
 

 Ponzi scheme

Helmut,

You think my mentor Laszlo E has read this? :-D:-D:-D

Just out of extreme curiosity, I got a copy and read it.. Am I missing something or I got a bad copy? The last sentence of the journal was not even finished... whats after "the"? All I see bibliography...

"To obtain the within subject standard deviation of reference product (Swr) or within subject variability of reference product the"

John
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2016-02-17 13:29
(2962 d 05:12 ago)

@ jag009
Posting: # 16002
Views: 21,961
 

 Ponzi scheme

Hi John,

❝ You think my mentor Laszlo E has read this? :-D:-D:-D


No. Would send him straight through the ceiling.

❝ Just out of extreme curiosity, I got a copy and read it.. Am I missing something or I got a bad copy? The last sentence of the journal was not even finished... whats after "the"? All I see bibliography...


I received this masterpiece already last October from Detlew. Looked the same. The cross-references to formulas are wrong as well. The text is truncated at the sentence you mentioned. Hence, references 6–8 are not mentioned anywhere. In predatory journals no serious review takes place…

As Wolfgang Pauli once said:

That is not only not right, it is not even wrong!
(Das ist nicht nur nicht richtig, es ist nicht einmal falsch!)


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
sanketh.gupta
☆    

India,
2016-03-07 07:21
(2943 d 11:21 ago)

(edited by sanketh.gupta on 2016-03-07 09:49)
@ Helmut
Posting: # 16061
Views: 21,225
 

 Ponzi scheme

Hi John, Lara, Detlew and Helmut

Please do not underestimate the article by looking at journal name. You all are requested to see the formulas mentioned to arrive at sample size estimations, since it is pure Mathematics. Please be noted that Mathematics is same throughout the globe.

You all believe following is the formula for sample size estiamtion? Please be noted that the same formula used world wide for sample size estimation.

SAMPLE SIZE ESTIMATION (The formula obtained from the book “design and analysis of BA and BE Studies, SHEIN-Chung CHOW, JEN-PEI LIU”), Refer Page no: 153.

ne = [t(α, 2n-2) + t(β, 2n-2)]2 * [CV/(V-δ)]2

Where

n- is sample size as per pilot study.
CV- is Coefficient of Variation (intra-subject variability) of that PK parameter which has maximum variability.
V- is Log of Bioequivalence limit (0.8 or 1.25) i.e. 20% difference.
(In general 10% difference is considered. In case of highly variable drugs go for 5% difference of Generic Vs Innovator)
δ- is Absolute value of expected/ observed difference.
α- is level of significance, generally 5%.
β- is probability of Type II error.
Note: The minimum number of subjects that would be acceptable is 12. Care should be taken to avoid any bitter any experience.

The formula correlating intra-subject variability of reference and within subject standard deviation of reference obtained from the EU directives of Doc. Ref.: CPMP/EWP/QWP/1401/98 Rev. 1/ Corr ** (Refer page no: 17 of 27).

CV(%)=100*Sqrt((e^(S[2]/WR])-1)

We noted that the incomplete sentence in the article and suggested to delete the same before publishing, but unfortunately it was not deleted by the Editor. Please refer to detailed communication with the Editor.

Dear Editor

We have given comments to delete the below mentioned incomplete sentence (page no: 5045) from the manuscript, but unfortunately that was not done.

We appreciate, if you could able to delete this.

Thanking you.

Best Regards,

Someswar Korla

From: Recentscientific Recentscientific [mailto:[email protected]]
Sent: Wednesday, July 29, 2015 11:20 AM
To: Someswar.Korla
Subject: Article has been published

Dear author,

The Editor-in-Chief is pleased to inform you that, submitted your research article has been published in the International Journal of Recent Scientific Research through

recentscientific.com/sample-size-estimation-highly-variable-drugs-using-reference-scaled-average-bioequivalence-criteria

You can now download your published paper.

Hope to get more papers from you and your colleagues.

With Regards,
Editorial Office
International Journal of Recent Scientific Research
www.recentscientific.com
Email: [email protected]

Research Article are Invited for Publication
www.recentscientific.com

For any query in this regard feel free to contact me at any time
with regards
--------------------
Editorial Office
International Journal of Recent Scientific Research
www.recentscientific.com
email: [email protected]


Moreover, we have validated obtained sample size estimations against the published article “Sample Sizes for Designing Bioequivalence Studies for Highly Variable Drugs, J Pharm Pharmaceut Sci (www.cspsCanada.org) 15(1) 73 - 84, 2011”. The same article did not specify the sample size estimations beyond 60% ISCV and above. Our intention to provide sample size estimations with 60% ISCV and above as pharmaceutical scientists are frequently using scaled average bioequivalence studies.

If anything not understand please send your comments to me, I will make you understand.

If you all are “gems” please provide sample size estimation formula for scaled average bioequivalence studies from your end and educate the pharmaceutical world.

Best Regards,
Someswara Rao
d_labes
★★★

Berlin, Germany,
2016-03-07 12:20
(2943 d 06:21 ago)

@ sanketh.gupta
Posting: # 16063
Views: 20,925
 

 Education necessary

Dear Someswara Rao,

I'm not in the mood to comment on any of your arguments.

Sorry, but I only can repeat Helmut's quote of Wolfgang Pauli:

That is not only not right, it is not even wrong!
(Das ist nicht nur nicht richtig, es ist nicht einmal falsch!)


You may find out the reasons why I'm repeating this if you study the posts in the forums category [Power / Sample size] and additionally if you study the literature on this topic in depth. Especially that about power/sample size for replicate crossover designs and for HVD(P)s.

Moreover I recommend Helmuts lectures / presentations. There are some about power / sample size.

Regards,

Detlew
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2016-03-07 15:21
(2943 d 03:20 ago)

@ sanketh.gupta
Posting: # 16068
Views: 21,347
 

 Dangerous nonsense

Hi Sanketh Gupta (or Someswara Rao?),

first of all: Who are you? From the article and the registration e-mail of Sanketh Gupta I suppose that both of you are employees of the same company. Are you sharing an account? This is not acceptable and may lead to blocking Sanketh’s account (see the Forum’s Policy).
If you are in fact Someswara Rao please register to the forum and post with your own account.

❝ Please do not underestimate the article by looking at journal name.


I didn’t do so when I became aware of the article last October. I judged it based on its – rather doubtful – content.

❝ You all are requested to see the formulas mentioned to arrive at sample size estimations, since it is pure Mathematics. Please be noted that Mathematics is same throughout the globe.


THX for reminding me about the spatial validity of mathematics. Please note that a correct (though approximative) formula applied wrongly (e.g., to a false model) gives still a wrong result.

❝ You all believe following is the formula for sample size estiamtion?


I cannot speak for the others, but myself for 2×2×2 crossovers and ABE (!) as a last resort, yes. Note that the formula is based on the shifted central t-distribution which is an approximation of the noncentral t-distribution which itself approximates the exact method (Owen’s Q-function). Why not use a better method? But we are not discussing 2×2×2 crossovers here, right?

❝ Please be noted that the same formula used world wide for sample size estimation.


Hopefully not for reference-scaling! Due to the conditions of the frameworks (scaling applicable only if CVwR >30%, GMR restriction of 0.8–1.25, different σ0 for the FDA and the EMA, upper cap at CVwR of 50% for the EMA) you must not simply plug the parameters you mentioned into any (‼) formula. Power and, therefore, sample sizes are not directly accessible – we need to perform Monte Carlo simulations! I suspect you did not understand the paper by Tóthfalusi and Endrényi. Quote:

Overall, the statistical properties of the methods proposed by EMA and FDA are rather complex as a result of the additional conditions and requirements (mixed procedure, GMR constraint, and (for EMA) a cap on the limits). Furthermore, the tests required by both EMA and FDA are dependent on each other which makes the theoretical treatment very complicated. Therefore, the required sample sizes were obtained by simulations.

Due to the slow convergence (example) one needs to simulate ≥105 BE studies for any given GMR, CV, and target power. This should be taken into account if comparing results with the article of the two Lászlós (only 10,000 simulations; hence, power might have been not sufficiently stable).

❝ (In general 10% difference is considered. In case of highly variable drugs go for 5% difference of Generic Vs Innovator)


Exactly the opposite. In ABE and low to moderate CVs ≥5% can reasonably be considered. In the case of HVDs/HVDPs Tóthfalusi and Endrényi recommend 10%. A larger deviation than the one commonly used with drugs with low to moderate variability should be assumed regardless whether scaling is intended or not. It is an intrinsic property of HVD(P)s that the GMR varies between studies.

❝ We noted that the incomplete sentence in the article and suggested to delete the same before publishing, but unfortunately it was not deleted by the Editor.


Why the heck did you want to delete the truncated sentence? It should have been the other way ’round: Complete it. References 6–8 are not linked to anything in the article. So likely more text is missing.

❝ Moreover, we have validated obtained sample size estimations against the published article “Sample Sizes for Designing Bioequivalence Studies for Highly Variable Drugs, J Pharm Pharmaceut Sci (www.cspsCanada.org) 15(1) 73 - 84, 2011”.


You can’t be serious!
Example: Four-period full replicate design, GMR 0.90, target power ≥80%.
FDA Reference scaling according to progestrone guidance.
EMA Average Bioequivalence with Expanding Limits, Method A according to the Q&A document.
1 Rao et al. (2015)
2 R package PowerTOST (2016). Function sampleN.RSABE() for the FDA’s RSABE and sampleN.scABEL() for the EMA’s ABEL.
3 Tóthfálusi & Endrényi (2012)

[image]         FDA          EMA   
     ──────────  ───────────
CV%   1   2   3    1   2   3
 30  41  32  30   41  34  35
 32  22  30   –   38  36   –
 34  21  28   –   35  34   –
 35  21  28  26   34  34  34
 36  20  26   –   33  34   –
 38  19  26   –   31  32   –
 40  19  24  24   29  30  31
 42  18  24   –   28  30   –
 44  18  24   –   27  28   –
 45  17  24  23   27  28  29
 46  17  24   –   27  28   –
 48  17  22   –   26  28   –
 50  17  22  22   25  28  28
 52  16  22   –   27  28   –
 54  16  22   –   29  28   –
 55  16  22  22   30  30  30
 56  16  22   –   31  30   –
 58  16  24   –   34  30   –
 60  16  24  23   36  32  32
 62  16  24   –   38  34   –
 64  16  24   –   41  36   –
 65  16  24  24   42  36  37
 66  16  24   –   43  36   –
 68  16  24   –   46  38   –
 70  16  26  24   48  40  40
[image] 72  16  26   –   51  42   –
 74  16  26   –   54  44   –
 75  16  26  26   55  44  45
 76  16  28   –   57  46   –
 78  16  28   –   60  48   –
 80  16  28  29   63  50  50
 82  17  30   –   66  52   –
 84  16  30   –   69  54   –
 85  16  32   –   71  54   –
 86  16  32   –   72  54   –
 88  16  32   –   76  56   –
 90  16  34   –   79  58   –
 92  16  34   –   83  60   –
 94  16  36   –   86  62   –
 95  16  36   –   88  64   –
 96  16  36   –   90  64   –
 98  16  38   –   94  66   –
100  16  40   –   97  68   –
105  17  42   –  107  74   –
110  17  44   –  118  78   –


Note that PowerTOST always rounds up if necessary to give balanced sequences (formatted in red). n–1 would already achieve at least the target power.
Correlations of PowerTOST’s sample sizes with the ones of the two Lászlós are high (R2 FDA 0.977, EMA 0.998) and close to the identity line. Can you please specify what you mean by “validated”?

Apart from the general disagreement, starting with a CV of 45% your sample sizes for RSABE become some kind of a “natural constant” (16–17). How do you explain that? With increasing CVs the likelihood of a PE with a large deviation from unity increases (pure chance!) and therefore, the GMR restriction to 0.8–1.25 will prevent demonstration of BE. Practically beyond 50% the GMR restriction is leading the decision; scaling itself becomes less important. You don’t even need a pocket calculator to check that. Pure reasoning. Tons of papers. This effect explains why with proper sample size estimation you will see a minimum at ~50% and increasing sample sizes beyond.

If companies follow your tables, studies would be
  • either only unethical (too low sample sizes lacking sufficient power) or
  • both unethical and economic questionable (too high sample sizes).
Scylla and Charybdis…

❝ The same article did not specify the sample size estimations beyond 60% ISCV and above.


Wrong again. All tables give sample sizes for up to 80% CV.

❝ […] I will make you understand.


How; what do you suggest? Brainwashing, waterboarding, hire a contract killer?

❝ […] please provide sample size estimation formula for scaled average bioequivalence studies from your end …


There is none.
If you hope that miraculously you will be able to derive an analytical solution, fantastic. Go ahead and submit it to a real journal. Become famous.

❝ … and educate the pharmaceutical world.


Mission accomplished.
@Someswara: I hope you have the guts to retract this article – if that’s possible at a predatory journal at all.

Since recentscientific.com and this site have the same PageRank (4) I hope that colleagues interested in sample size estimation for reference-scaling will find not only your article but this thread as well.


Keywords: Highly Variable Drugs (HVDs) • Highly Variable Drug Products (HVDPs) • Reference-scaled Average Bioequivalence (RSABE) • Average Bioequivalence with Expanding Limits (ABEL) • Sample Size Estimation • Monte Carlo Simulation • Food and Drug Administration (FDA) • European Medicines Agency (EMA) • Progesterone Guidance • Power

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
somu_korla
☆    

India,
2016-03-09 09:17
(2941 d 09:24 ago)

@ Helmut
Posting: # 16076
Views: 21,328
 

 Clarifications to the Dangerous nonsense

❝ If you are in fact Someswara Rao please register to the forum and post with your own account.


My account is created, response is provided.

❝ ❝ Please do not underestimate the article by looking at journal name.

❝ […] I judged it based on its – rather doubtful – content.


Detlew commented as mentioned below and hence I responded.

❝ ❝ ❝ ❝ ❝ “Forget this "gem" of science literature. It's simply crap. Don't believe even in comma of this paper. Throw it in a high arc into your waste bin”.


Based on the formula given in “design and analysis of BA and BE Studies, SHEIN-Chung CHOW, JEN-PEI LIU”, Refer Page no: 153)”, the only “log of bioequivalence limit” need to be substituted to obtain the required sample size. Rest all other parameters in the formula can be filled based on the pilot/historical data.

❝ ❝ You all are requested to see the formulas mentioned to arrive at sample size estimations, since it is pure Mathematics.

❝ Please note that a correct (though approximate) formula applied wrongly (e.g., to a false model) gives still a wrong result.


The correct formula never applied wrongly here; instead it was used as base for the sample size estimation. The reference for this is “design and analysis of BA and BE Studies, SHEIN-Chung CHOW, JEN-PEI LIU”), Refer Page no: 153.

❝ ❝ You all believe following is the formula for sample size estimation?

❝ I cannot speak for the others, but myself for 2×2×2 crossovers and ABE (!) as a last resort, yes. Note that the formula is based on the shifted central t-distribution which is an approximation of the noncentral] t-distribution which itself approximates the exact method (Owen’s Q-function). Why not use a better method? But we are not discussing 2×2×2 crossovers here, right?


I don’t have to respond for this.

❝ ❝ Please be noted that the same formula used worldwide for sample size estimation.

Hopefully not for reference-scaling! Due to the conditions of the frameworks (scaling applicable only if CVwR >30%, GMR restriction of 0.8–1.25, different σ0 for the FDA and the EMA, upper cap at CVwR of 50% for the EMA) you must not simply plug the parameters you mentioned into any (‼) formula.


We do understand that the requirements of HVDs for both FDA and EMA are different. Hence to obtain the log of bioequivalence with respective to FDA and EMA separately, which need to be substituted in the formula, we have explained all the calculations in the paper.

❝ Power and, therefore, sample sizes are not directly accessible – we need to perform Monte Carlo simulations! I suspect you did not understand the paper by Tóthfálusi and Endrényi. […] Due to the slow convergence (example) one needs to simulate ≥105 BE studies for any given GMR, CV, and target power. This should be taken into account if comparing results with the article of the two Lászlós (only 10,000 simulations; hence, power might have been not sufficiently stable).


Simulation with ≥105 BE studies is not possible with us, instead it is pure formula based one. In general sample size calculations are done using 80% power requirements; however, the post hoc power is not at all important, as long as you bioequivalence limits are within the acceptance range.

❝ ❝ (In general 10% difference is considered. In case of highly variable drugs go for 5% difference of Generic Vs Innovator)

❝ Exactly the opposite. In ABE and low to moderate CVs ≥5% can reasonably be considered. In the case of HVDs/HVDPs Tóthfálusi and Endrényi recommend 10%. A larger deviation than the one commonly used with drugs with low to moderate variability should be assumed regardless whether scaling is intended or not. It is an intrinsic property of HVD(P)s that the GMR varies between studies.


In general, a generic player will always try to match innovator. To reserve the resources and to ensure the probability of success (to pass the study), a 10% difference can be considered low to moderate variable drugs as it demand less sample size and a 5% difference is always appropriate for highly variable drugs as it demands more sample size due high variability.

Kindly note that, as long as the test versus reference difference increases, more sample size is required to prove the bioequivalence. I don’t advice to go with 10% difference as unpredictable behavior will be observed HVDs as well as it requires large sample size. The probability of success will be very high with 5% difference while ensuring the resources conservative instead of 10% test versus reference difference and I don’t want to compare with Tóthfálusi and Endrényi in this case.

❝ ❝ We noted that the incomplete sentence in the article and suggested to delete the same before publishing, but unfortunately it was not deleted by the Editor.

❝ Why the heck did you want to delete the truncated sentence? It should have been the other way ’round: Complete it. References 6–8 are not linked to anything in the article. So likely more text is missing.


We would like to explain that, “to obtain within subject standard deviation of reference product (SWR) or within subject variability of reference product by using following formula” and it is obtained after rearranging the equation mentioned in page no: 17 of 27 in the EU directives of Doc. Ref.: CPMP/EWP/QWP/1401/98 Rev. 1/ Corr **
Swr = Sqrt [ln (CV)2+1]

Reference 6 is used to obtain the above mentioned formula correlating within subject standard deviation of reference product (SWR) or within subject variability of reference product.

Reference 7 is used is just to describe formula given in “design and analysis of BA and BE Studies, SHEIN-Chung CHOW, JEN-PEI LIU”, Refer Page no: 153)”,

Reference 8 is used is just to describe the initial introduction part.

❝ ❝ Moreover, we have validated obtained sample size estimations against the published article “Sample Sizes for Designing Bioequivalence Studies for Highly Variable Drugs, J Pharm Pharmaceut Sci (www.cspsCanada.org) 15(1) 73 - 84, 2011”.

You can’t be serious! […] Can you please specify what you mean by “validated”?


We have compared the obtained sample size by using our method with Tóthfálusi & Endrényi (2012) method and it was identified that the obtained sample size did not differed significantly using EMA approach. However it was differed using FDA method. This difference is expected as our approach is the formula based one as the calculations are different here due to change in the regulatory constants.

❝ ❝ The same article did not specify the sample size estimations beyond 60% ISCV and above.

❝ Wrong again. All tables give sample sizes for up to 80% CV.


Agreed.

❝ […] I will make you understand.

How; what do you suggest? Brainwashing, waterboarding, hire a contract killer?


What I mean is how we have arrived at these calculations.

❝ ❝ […] please provide sample size estimation formula for scaled average bioequivalence studies from your end …

❝ There is none.

❝ If you hope that miraculously you will be able to derive an analytical solution, fantastic. Go ahead and submit it to a real journal. Become famous.


I am confident that the formula used will definitely works, since I did not manipulated anything and the same formula know to everyone in the pharmaceutical industry.

❝ ❝ … and educate the pharmaceutical world.

❝ Mission accomplished.

❝ @Someswara: I hope you have the guts to retract this article – if that’s possible at a predatory journal at all.


I don’t have to retract this article since nothing is manipulated. Of course there are some typo errors I accepted the same. Again what are all questions raised by you all are answered.

❝ Since recentscientific.com and thus site have the same PageRank (4) I hope that colleagues interested in sample size estimation for reference-scaling will find not only your article but this thread as well.


I am not bothered about the ranking of journals; the intention was to reveal the sample size justification formula for SABE studies.

Best Regards
Someswara Rao


Edit: Full quotes removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post! The forum’s standard quote system restored. [Helmut]
d_labes
★★★

Berlin, Germany,
2016-03-09 10:46
(2941 d 07:56 ago)

(edited by d_labes on 2016-03-09 11:12)
@ somu_korla
Posting: # 16077
Views: 20,545
 

 Absurd absurdity

SCNR.

"Forgive me my nonsense as I also forgive the nonsense of those who think they can talk sense."
Robert Frost

Regards,

Detlew
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2016-03-09 17:02
(2941 d 01:40 ago)

@ somu_korla
Posting: # 16080
Views: 20,845
 

 Outright bizarre

Hi Someswara,

❝ Simulation with ≥105 BE studies is not possible with us […]



Too bad that simulations are the only possible approach.
Your article is a slap in the face of the ones devoting a lot of time and efforts into developing fast algorithms and setting up simulations. When in 2010 the FDA’s draft progesterone guidance and the EMA’s BE-GL were published statisticians were desperate because it became immediately evident (though still not to you, obviously) that common methods for sample size estimation cannot be applied.
It was a big relief when the two Lászlós published their paper in 2011 and that since Feb. 2013 package PowerTOST allows simulations for any expected θ0, CVwR, and desired power.
R and package PowerTOST come free of costs and are licensed under GPL-3. What do you mean by “not possible with us”?

❝ In general sample size calculations are done using 80% power requirements; however, the post hoc power is not at all important, as long as you bioequivalence limits are within the acceptance range.


Splendid! In the following a comparison of part of your tables with results obtained by simulations.
n1 Sample size for ≥80% power.
n2 Sample size for ≥90% power.
β  Producer’s risk (1 – power). β larger than desired (= underpowered) formatted in red.
  • FDA (RSABE)

                RRT | RTR | TRR        
          Rao    Sim’s    Rao    Sim’s
    CV  n1  β   n1  β   n2  β   n2  β  
    30% 62 11.1 45 19.7 86  4.8 65  9.9
    50% 26 25.1 30 19.8 35 15.3 45  9.9
    80% 24 31.8 41 20.0 32 24.3 88 10.0

                  RTRT | TRTR          
          Rao    Sim’s    Rao    Sim’s
    CV  n1  β   n1  β   n2  β   n2  β  
    30% 41 11.6 31 19.4 57  5.1 44 10.0
    50% 17 30.1 22 19.7 23 21.3 32  9.7
    80% 16 35.1 28 19.9 21 25.9 60  9.9


  • EMA (ABEL)

                RRT | RTR | TRR         
          Rao    Sim’s    Rao     Sim’s
    CV  n1  β   n1  β    n2  β   n2  β  
    30% 62 14.1 52 19.5  86 6.3  73  9.8
    50% 38 20.3 39 19.2  53 9.5  52  9.9
    80% 95 11.3 75 19.7 131 5.3 100 10.0

                  RTRT | TRTR          
          Rao    Sim’s    Rao    Sim’s
    CV  n1  β   n1  β   n2  β   n2  β  
    30% 41 14.1 34 19.7 57 14.4 48 10.0
    50% 25 23.1 27 19.9 35 11.3 37 10.0
    80% 63 11.4 49 19.5 87  5.9 68  9.8

In many cases studies designed according to your tables will be (extremely) underpowered (i.e., producer’s risk larger than desired). Hence, the chances to fail are higher than expected as well. If you are working with a CRO and the sponsor seeks a second opinion by a learned statistician, you will loose a client.

However, sometimes you’ll make a lucky strike (producer’s risk substantially smaller than targeted). Study overpowered, more subjects than required treated. Fortunately HVD(P)s are safe; low chances of AEs. But any intervention possesses risks. As a member of an IEC I would never accept the protocol. Anyhow: Study performed and easily passes BE. Cross you fingers that the sponsor is so happy with the outcome that he does not seek a second opinion. If he does, he would ask himself why he spent more money than necessary. Probably he will hire another CRO the next time.

Almost needless to say that in the simulations β is always as close as possible to the desired level.

❝ In general, a generic player will always try to match innovator. To reserve the resources and to ensure the probability of success (to pass the study), a 10% difference can be considered low to moderate variable drugs as it demand less sample size and a 5% difference is always appropriate for highly variable drugs as it demands more sample size due high variability.


Whereof one cannot speak,
thereof one must be silent.
    Ludwig Wittgenstein

❝ We have compared the obtained sample size by using our method with Tóthfálusi & Endrényi (2012) method and it was identified that the obtained sample size did not differed significantly using EMA approach. However it was differed using FDA method. This difference is expected as our approach is the formula based one as the calculations are different here due to change in the regulatory constants.


Did you bother to look at the table and plots in my previous post? If you don’t see differences likely you are :blind:.

❝ I am confident that the formula used will definitely works, […] and the same formula know to everyone in the pharmaceutical industry.


Old beliefs die hard
even when demonstrably false.
    E. O. Wilson

❝ I am not bothered about the ranking of journals …


In earnest? If you were so confident that you discovered the “philosopher’s stone” – everybody is desperately seeking for so long – why didn’t you submit the manuscript to a reputable journal (free of costs, BTW)? Or were you in private afraid that it will fail to survive any serious peer-review?

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
somu_korla
☆    

India,
2016-03-10 09:44
(2940 d 08:58 ago)

@ Helmut
Posting: # 16083
Views: 20,780
 

 Clarifications to Outright bizarre

Hi,

❝ Your article is a slap in the face of the ones devoting a lot of time and efforts into developing fast algorithms and setting up simulations. When in 2010 the FDA’s draft progesterone guidance and the EMA’s BE-GL were published statisticians were desperate because it became immediately evident (though still not to you, obviously) that common methods for sample size estimation cannot be applied.

❝ It was a big relief when the two Lászlós published their paper in 2011 and that since Feb. 2013 package PowerTOST allows simulations for any expected θ0, CVwR, and desired power..

R and package PowerTOST come free of costs and are licensed under GPL-3. What do you mean by “not possible with us”?


I never questioned anyone’s paper or work what they have done, instead I have learned much out of it. I have done my exercise whatever is possible with me and produced the paper. If you find any limitation you don’t use it instead of criticizing, which is an easy job.

❝ ❝ In general sample size calculations are done using 80% power requirements; however, the post hoc power is not at all important, as long as you bioequivalence limits are within the acceptance range.

❝ Splendid! […]


❝ In many cases studies designed according to your tables will be (extremely) underpowered (i.e., producer’s risk larger than desired). Hence, the chances to fail are higher than expected as well. If you are working with a CRO and the sponsor seeks a second opinion by a learned statistician, you will loose a client.

❝ However, sometimes you’ll make a lucky strike (producer’s risk substantially smaller than targeted). Study overpowered, more subjects than required treated. Fortunately HVD(P)s are safe; low chances of AEs. But any intervention possesses risks. As a member of an IEC I would never accept the protocol. Anyhow: Study performed and easily passes BE. Cross you fingers that the sponsor is so happy with the outcome that he does not seek a second opinion. If he does, he would ask himself why he spent more money than necessary. Probably he will hire another CRO the next time.

❝ Almost needless to say that in the simulations β is always as close as possible to the desired level.


I can convince my client and I don’t need your suggestion in this case and it is my headache.

❝ ❝ In general, a generic player will always try to match innovator. To reserve the resources and to ensure the probability of success (to pass the study), a 10% difference can be considered low to moderate variable drugs as it demand less sample size and a 5% difference is always appropriate for highly variable drugs as it demands more sample size due high variability.

Whereof one cannot speak, thereof one must be silent.  Ludwig Wittgenstein


I don’t have to respond for this.

❝ ❝ We have compared the obtained sample size by using our method with Tóthfálusi & Endrényi (2012) method and it was identified that the obtained sample size did not differed significantly using EMA approach. However it was differed using FDA method. This difference is expected as our approach is the formula based one as the calculations are different here due to change in the regulatory constants.

❝ Did you bother to look at the table and plots in my previous post? If you don’t see differences likely you are :blind:.


The paper produced is purely based on the sample size estimation formula and reference is already given to you many times. For this I don’t have to refer what you have suggested.

❝ I am confident that the formula used will definitely works, […] and the same formula know to everyone in the pharmaceutical industry.

Old beliefs die hard even when demonstrably false.  E. O. Wilson


The paper produced is purely based on the sample size estimation formula and reference is already given to you many times.
Moreover as explained earlier I did not manipulated anything in it. The same old formula was used to meet the new regulatory requirement.

❝ ❝ I am not bothered about the ranking of journals …

❝ In earnest? If you were so confident that you discovered the “philosopher’s stone” – everybody is desperately seeking for so long – why didn’t you submit the manuscript to a reputable journal (free of costs, BTW)? Or were you in private afraid that it will fail to survive any serious peer-review?


I don’t have to afraid to any one since the same formula known to everyone and there is no manipulation in it.
You all people helped a lot because for all the questions raised by you all are answered all as well as all the people who are registered in the forum indirectly.
My sincere advice is that; please do not comment on anybody’s work/paper. If you did not find answer you just leave it a side and go ahead.

Best Regards
Someswara Rao


Edit: Full quotes removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post! The forum’s standard quote system restored. Please don’t use your own “private” one. [Helmut]
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2016-03-10 19:58
(2939 d 22:44 ago)

@ somu_korla
Posting: # 16087
Views: 20,591
 

 Science vs. fairy tales

Hi Someswara,

❝ ❝ Your article is a slap in the face of the ones devoting a lot of time and efforts into developing fast algorithm and setting up simulations.

❝ I never questioned anyone’s paper or work what they have done, instead I have learned much out of it. I have done my exercise whatever is possible with me and produced the paper. If you find any limitation you don’t use it instead of criticizing, which is an easy job.


Science does not work that way. All [sic] papers dealing with reference-scaling published so far assessed power through simulations – and for good reasons.

If you tell me that you have a sheep in your garden,
I will believe you.
If you tell me that you have a unicorn in your garden,
I will go and see it myself.

Do you get the analogy?

❝ ❝ In many cases studies designed according to your tables will be (extremely) underpowered (i.e., producer’s risk larger than desired). […] However, sometimes you’ll make a lucky strike (producer’s risk substantially smaller than targeted). Study overpowered, more subjects than required treated.

❝ I can convince my client and I don’t need your suggestion in this case and it is my headache.


Studies have to be sufficiently powered (according to all BE guidelines and ICH E9). This is not the case following your “method”. I strongly suggest that you attend training in medical ethics. Human volunteers are not guinea pigs. If “convincing your client” is your primary aim, selling used cars might be a better field of work. Excuse my French.

❝ ❝ ❝ We have compared the obtained sample size by using our method with Tóthfálusi & Endrényi (2012) method […]


❝ ❝ Did you bother to look at the table and plots in my previous post? If you don’t see differences likely you are :blind:.

❝ The paper produced is purely based on the sample size estimation formula and reference is already given to you many times. For this I don’t have to refer what you have suggested. Moreover as explained earlier I did not manipulated anything in it. The same old formula was used to meet the new regulatory requirement.


I really don’t understand why you are repeating this (non)argument over and over again. You have answered only a tiny fractions of my questions and concerns.

❝ ❝ ❝ I am not bothered about the ranking of journals …

❝ ❝ […] why didn’t you submit the manuscript to a reputable journal […]? Or were you in private afraid that it will fail to survive any serious peer-review?

❝ My sincere advice is that; please do not comment on anybody’s work/paper. If you did not find answer you just leave it a side and go ahead.


Come on, Someswara! Don’t you know what a public forum is? Since you have chosen to publish in a predatory journal the usual path (below) is not possible.
  • Write a letter to the editor.
  • The authors will be given the chance to respond.
  • Both will be peer-reviewed.
  • In the journal’s site of the original article both will be linked and shown in literature data bases.
It’s not about me. Less experienced people will find your paper and perform studies for HVD(P)s with a CVwR of 100% in 18 subjects. This is a 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
somu_korla
☆    

India,
2016-03-11 11:00
(2939 d 07:42 ago)

@ Helmut
Posting: # 16092
Views: 20,513
 

 Clarifications for Science vs. fairy tales

Hi,

❝ ❝ I have done my exercise whatever is possible with me and produced the paper. If you find any limitation you don’t use it instead of criticizing, which is an easy job.

❝ Science does not work that way. All [sic] papers dealing with reference-scaling published so far assessed power through simulations – and for good reasons.

❝ If you tell me that you have a sheep in your garden,

❝ I will believe you.

❝ If you tell me that you have a unicorn in your garden,

❝ I will go and see it myself.


❝ Do you get the analogy?


There is nothing to hide in the paper and every term was explained detail in the presented formula. Hence I don’t have to rely on your logic.

❝ ❝ I can convince my client and I don’t need your suggestion in this case and it is my headache.

❝ Studies have to be sufficiently powered (according to all BE guidelines and ICH E9). This is not the case following your “method”. I strongly suggest that you attend training in medical ethics. Human volunteers are not guinea pigs. If “convincing your client” is your primary aim, selling used cars might be a better field of work. Excuse my French.


Thanks for your suggestion with regards to the medical ethics and I don’t have to learn from you. Please be noted that power is also involved in the computation of sample size.

❝ ❝ The paper produced is purely based on the sample size estimation formula and reference is already given to you many times. For this I don’t have to refer what you have suggested. Moreover as explained earlier I did not manipulated anything in it. The same old formula was used to meet the new regulatory requirement.

❝ I really don’t understand why you are repeating this (non)argument over and over again. You have answered only a tiny fractions of my questions and concerns.


If you don’t understand that formula, please leave it.

❝ ❝ My sincere advice is that; please do not comment on anybody’s work/paper. If you did not find answer you just leave it a side and go ahead.

❝ Come on, Someswara! Don’t you know what a public forum is? Since you have chosen to publish in a predatory journal the usual path (below) is not possible.

❝ – Write a letter to the editor.

❝ – The authors will be given the chance to respond.

❝ – Both will be peer-reviewed.

❝ – In the journal’s site of the original article both will be linked and shown in literature data bases.

❝ It’s not about me. Less experienced people will find your paper and perform studies for HVD(P)s with a CVwR of 100% in 18 subjects. This is a problem!


The formula based computation it is like that and is transparent, finally there is no manipulation. It is pure individual’s decision to go ahead. According to the formula, CVwR increases, the BE limit will also increase. As the BE limit is in the denominator subsequently the sample size is getting decreased.

Best Regards
Someswara Rao


Edit: Full quotes removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post! The forum’s standard quote system restored. [Helmut]
d_labes
★★★

Berlin, Germany,
2016-03-11 17:30
(2939 d 01:12 ago)

(edited by d_labes on 2016-03-11 20:23)
@ Helmut
Posting: # 16097
Views: 20,253
 

 Forlorne hope

Dear Helmut,

obviously every effort to educate Someswara is forlorne hope (in German: Vergebene Liebesmüh).
[image]
Do Not Feed the Troll

Regards,

Detlew
lechia
☆    

C of U,
2016-08-06 00:52
(2791 d 18:49 ago)

@ d_labes
Posting: # 16539
Views: 17,092
 

 Forlorne hope

❝ If you don’t understand that formula, please leave it.


This is arguably the funniest chain on this forum. At least we learned that you are the problem in this exchange, Helmut. :-P
nobody
nothing

2016-08-08 17:49
(2789 d 01:52 ago)

@ lechia
Posting: # 16542
Views: 16,983
 

 Forlorne hope

... I get deeply frustrated by reading this, as the basis of science has eroded in some countries that I in principle would not want to touch any medicine "developed" according to such poor science. But in fact: The market is flooded by such products. And the problem does not start with fancy sample size estimates for BE trials, I guess from my experience/reading the pharma news.

***depressed***

Kindest regards, nobody
d_labes
★★★

Berlin, Germany,
2016-08-09 10:50
(2788 d 08:51 ago)

@ nobody
Posting: # 16544
Views: 16,938
 

 Forlorne hope

Dear nobody!

❝ ***depressed***


The real reason to get depressed are sponsors which order BE studies from such people in order only to save some bucks.

Regards,

Detlew
UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,636 registered users;
71 visitors (0 registered, 71 guests [including 4 identified bots]).
Forum time: 18:42 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