Log in
Register|
beman ☆ 2014-04-04 15:48 (4047 d 08:59 ago) Posting: # 12771 Views: 7,668 |
|
|
Dear all, we plan to perform a pilot study as 4-period BE study fed/fasting conditions. Some issues were discusssed in: http://forum.bebac.at/mix_entry.php?id=3820#p3820 Approach 1: Williams Design as 4x4, 4 Sequences - Rfe Rfa Tfa Tfe - Tfa Rfe Tfe Rfa - Tfe Tfa Rfa Rfe - Rfa Tfe Tfe Tfa T vs. R (BE): Statistical Data evaluation (ANOVA + CI) will be done separately for each conditions without the data of the other condition. fed vs. fast (Food effect): Statistical Data evaluation (ANOVA + CI) will be done separately for each formulation withour the data of the other formulation. pros: - balanced design cons: - statistical data evaluation using only 50 % of the whole data, the influence (effects, ...) of the other 50 % of the data is lost. Approach 2: 2x2-Design for feating conditions in Periode 1 and 2, 2x2-Design for fed conditions in Periode 1 and 2 leads to 2 sequences - Rfa Tfa Rfe Tfe - Tfa Rfa Tfe Rfe T vs. R (BE): Statistical Data evaluation (ANOVA + CI) will be performed as 2x2x2-Design for each condition. fed vs. fast: Statistical Data evaluation (ANOVA + CI) will be done separately for each formulation without the data of the other formulation. fed vs. fast (Food effect): Statistical Data evaluation (ANOVA + CI) will be done separately for each condition withour the data of the other condition without the factor period in the ANOVA. pros: - 2x2-Design for each condition cons: - period effect could not be captured in food effect analysis Because the pivotal studies will be perfomed in two seperate 2x2-crossover studies, one under fasting and one under fed conditions, Approach 2 would be the best for our purpose. In the current EMA-Guideline is stated, it is acceptable to conduct either two seperate 2x2-crossover studies or a four way crosssover study. My question is, which approach is the best for a 4x4-Pivotal study in your oppinion? Thank you in advance for your answers. Best regards. beman |
|
Helmut ★★★   Vienna, Austria, 2014-04-04 16:00 (4047 d 08:46 ago) @ beman Posting: # 12772 Views: 6,533 |
|
|
Dear beman, since I suggested #2 in a comment to the BE-GL, you may guess which one is my favorite. 
For an example of beman’s #2 see:
— Dif-tor heh smusma 🖖🏼 Довге життя Україна! ![[image]](https://static.bebac.at/pics/Blue_and_yellow_ribbon_UA.png) Helmut Schütz ![[image]](https://static.bebac.at/img/CC by.png) The quality of responses received is directly proportional to the quality of the question asked. 🚮 Science Quotes |
|
Tina ★ India, 2014-05-16 12:35 (4005 d 12:12 ago) @ Helmut Posting: # 12954 Views: 6,099 |
|
|
Dear Helmut, Is it possible to have a two-stage design in a 4-way crossover study (involving fasting and fed of both R and T) if variability observed between the two conditions differ pointing to 2 different sample sizes for fasting and fed. Suppose fasting reuires 60 subjects and fed requires 100 subjects, do we have to dose 100 subjects intoto or to separate the sample sizes? Kind regards, Tina |
|
Helmut ★★★ Vienna, Austria, 2014-05-16 14:03 (4005 d 10:43 ago) @ Tina Posting: # 12956 Views: 6,116 |
|
|
Hi Tina, ❝ Is it possible to have a two-stage design in a 4-way crossover study (involving fasting and fed of both R and T) if variability observed between the two conditions differ pointing to 2 different sample sizes for fasting and fed.
❝ Suppose fasting reuires 60 subjects and fed requires 100 subjects, do we have to dose 100 subjects intoto or to separate the sample sizes? Example for #2: First stage 48 subjects, CVfasting 34%, CVfed 45%, Method B (T/R 0.95, 80% power). Perform the second stage in 60–48=12 subjects fasting and 100–48=52 subjects fed. — 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 |
|
Dr_Dan ★★ Germany, 2014-04-04 21:44 (4047 d 03:03 ago) @ beman Posting: # 12774 Views: 6,421 |
|
|
Dear beman, to demonstrate bioequivalence you need to show that the 90% CI are within the acceptance range in both states, under fasting as well as under fed conditions. Do you really need to know the size of the food effect? IMHO not since the food effect of the originator formulation is known and as long as your formulation is bioequivalent in both conditions it is more or less the same with your formulation. Therefore I would keep the design as simple as possible. The current EMA-Guideline states, that it is acceptable to conduct either two seperate 2x2-crossover studies or a four way crosssover study. Using the four way crossover study design would give you the possibility to determine the size of the food effect but you will have a higher risk of drop outs, you can not adapt sample sizes in case the two conditions differ in their variability, you can not conduct the evaluation in sequence which reduces the risk of additional costs in case of failure etc..... Therefore I prefer two 2x2 crossover studies. I hope this helps Kind regards Dr_Dan — Kind regards and have a nice day Dr_Dan |
|
Tina ★ India, 2014-05-16 14:31 (4005 d 10:16 ago) @ Dr_Dan Posting: # 12958 Views: 6,164 |
|
|
Dear Dr_Dan, Thanks for sharing the 'common-sense' approach. I would also plan 2 seperate studies. This plan would be perfect for EU submission. However, FDA requires studies for both fasting and fed and if literature dosent say much interference with meal conditions and invitro equivalence is confirmed, Beman approach appears appealing Kind regards, Tina |
Ing. Helmut Schütz