khaoula
★    

Algeria,
2014-11-19 13:00
(3417 d 23:26 ago)

Posting: # 13876
Views: 9,074
 

 pilot replicate design [RSABE / ABEL]

Hi everybody !!!

please, If we suspect that gastro-resistant capsules of omeprazole are highly variable drug product, can we do a PILOTE replicate study? to determine the within subject variability of the reference formulation and widening the bioequivalence limits if CV CR > 30%
thank you
ElMaestro
★★★

Denmark,
2014-11-19 13:38
(3417 d 22:48 ago)

@ khaoula
Posting: # 13877
Views: 8,029
 

 pilot replicate design

Hi Khaoula,

❝ please, If we suspect that gastro-resistant capsules of omeprazole are highly variable drug product, can we do a PILOTE replicate study? to determine the within subject variability of the reference formulation and widening the bioequivalence limits if CV CR > 30%



Yes, you can do a pilot study to learn if the drug (drug product) is potentially hihgly variable. But usually in a pivotal trial you can only widen the acceptance limits if the high intra-subject variability has been shown in that same pivotal trial.

There might be local variations to this principle which I wouldn't know. Would be happy to learn.

Pass or fail!
ElMaestro
khaoula
★    

Algeria,
2014-11-19 15:11
(3417 d 21:15 ago)

@ ElMaestro
Posting: # 13881
Views: 8,001
 

 pilot replicate design

Hi El Maestro,
thank you for your answer, so to do a pilot replicate design, IS the number of subject needed 24 ? because in pilot cross over design the number is 12, but for a replicate design the minimum number required is 24
ElMaestro
★★★

Denmark,
2014-11-19 15:29
(3417 d 20:57 ago)

@ khaoula
Posting: # 13882
Views: 8,018
 

 N=24??

Hi Khaoula,

I do not know of such a rule.
In a pilot study as far as I know you can use whatever sample size you like. More subjects increase your confidence in the figures you wish to estimate.

In a pivotal N=12 minimum is the golden rule, but I thought this is regardless of the number of replicates for any treatment.

Have a good day.

Pass or fail!
ElMaestro
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2014-11-19 16:18
(3417 d 20:08 ago)

@ khaoula
Posting: # 13883
Views: 8,024
 

 pilot replicate design

Hi Khaoula,

I agree with ElMaestro. The FDA requires 24 dosed subjects in replicate studies intended for RSABE, but only for the pivotal studies. The number of eligible subjects can be lower.

As usual the impact of the ratio on the estimated sample size of the pivotal study is higher than the one of the CV (which is taken care in RSABE anyway). Therefore, I would not design the pilot in a small number of subjects. For HVDs the GMR jumps around like crazy between studies. Don’t be tempted to assume a GMR too close to 1 even if you observed it in a pilot. That’s why the two Lászlós suggested a GMR of 0.9 in study planning.

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
khaoula
★    

Algeria,
2014-11-19 16:57
(3417 d 19:29 ago)

@ Helmut
Posting: # 13885
Views: 8,031
 

 pilot replicate design

Hi Helmut, thank you for your answer,
If you remember I'm doing my post graduate disertation in pharmacology and the subject is bioequivalence I have already requested your help to understand the results of our omeprazole's bioequivalence study, so I studied statistic of bioequivalence from "Design and Analysis of Bioavailability and Bioequivalence Studies, of Shein-Chung Chow, in chapman edition", and I redid all the calculation:
CVw= 18- 23 % for AUCs and Cmax, GMR= 1,16- 1,17 for AUC and Cmax, treatment effect: it's justified because 100% is 'nt include in CI for all parameters
no sequence and no period effect
subject effect: justified: subjects are'nt multiple homozygote
CI C max : (102,15% 135,62%)
AUC 0-12 : (103,43% 131,02%)
AUC 0-∞: (104,08% 131,25%)

GMR is very high, in bibliography it's the case but in fed condition, so I suspect subjects that's did't respect the 10h fasting condition before study (volunteers are very big problem in our country), and I read that omeprazole is HDVP specially in fed condition (gastro resistant preparation), EMA and FDA recommends fed study for this preparation.
So to find a solution to our study, I want to do a replicate pilote study in fed condition to evaluate real intra subject variability of omeprazole gastro resistant preparation (reference), and in fed condition to control the subjects
what do you think?
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2014-11-20 15:39
(3416 d 20:47 ago)

@ khaoula
Posting: # 13894
Views: 7,905
 

 Futile…

Hi Khaoula,

according to both FDA’s and EMA’s requirements you have to perform studies both in fasting and fed state. With a PE of 117.7% for Cmax and significant differences of all PK metrics IMHO any further studies with this formulation are futile.

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
khaoula
★    

Algeria,
2014-11-21 13:19
(3415 d 23:07 ago)

@ Helmut
Posting: # 13906
Views: 7,889
 

 new information

Hi Helmut,
thank you for your answer,
I have new information (which I did'nt now before), before the study, the quality control of chemical dosage was realised:
the ratio of the chemical dosage of active ingredient in drug product T/R was 106 %, it's > 105%, so, this could perhaps explain the results...
can we introduce a corrective factor in the result ?


Edit: Follow-up post deleted according to the Forum’s Policy. [Helmut]
khaoula
★    

Algeria,
2014-11-22 11:38
(3415 d 00:48 ago)

@ Helmut
Posting: # 13914
Views: 7,806
 

 kinetica

Hi Helmut,
I read yesterday your article "Reference Datasets for 2-Treatment, 2-Sequence, 2-Period Bioequivalence Studies, The AAPS Journal, Vol. 16, No. 6, November 2014", about unbalanced cross over and calculation of confidence interval, but what about sequence effect which is calculated MS carry/ MS intra and not MS carry /MS inter?
thank you, I will never cease to say; I have learned a lot of things, thanks to you
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2014-11-22 13:43
(3414 d 22:43 ago)

@ khaoula
Posting: # 13915
Views: 7,810
 

 kinetica

Hi Khaoula,

❝ […] what about sequence effect which is calculated MS carry/ MS intra and not MS carry /MS inter?


Yes, it’s wrong in all versions (see this post). I’m not optimistic whether Thermo “Scientific” will correct it this time.

❝ […] I have learned a lot of things, thanks to you


Welcome!

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, 1,636 registered users;
108 visitors (0 registered, 108 guests [including 8 identified bots]).
Forum time: 12:27 CET (Europe/Vienna)

With four parameters I can fit an elephant,
and with five I can make him wiggle his trunk.    John von Neumann

The Bioequivalence and Bioavailability Forum is hosted by
BEBAC Ing. Helmut Schütz
HTML5