H.Patel
☆    

2011-06-15 18:28
(4670 d 00:22 ago)

Posting: # 7131
Views: 10,510
 

 Redosing study… [Study As­sess­ment]

Hello All,

We did a Pivotal study… sample size: 26.
It failed in cmax due to one subject whose T/R=300% due to reference formulation.

we are planning a redosing study to see if the data of this subject was
abberant or not…
In redosing study we are planning to go with 6 subjects. equal TR and RT sequence

Ques1) Can we mention in protocol that we will take sub no 1, 4, 8, 9, 14, 22 in the redosing study or it will stand as a selection bias. This is bcoz the ratio of these subs are near to unity. This will help in making our control arm in study more strong.

Ques 2) What shall be the range of data wherein if the redosing data of postulated outlier falls, then it can be considered as abberant observation.


Edit: Category changed. [Helmut]
swapnil.kuche
★    

2011-06-16 10:35
(4669 d 08:15 ago)

@ H.Patel
Posting: # 7134
Views: 9,590
 

 Redosing study…

Dear Mr. Patel,

Have you confirm whether redosing approach is acceptable? If redosing approach is acceptable can please mentioned for which regulatory body it is acceptable. If possible Please confirm the same with suitable references or examples.

I am eagarly waiting for your reference as we have also same case enumerated by you.


Regards,
Swapnil D. Kuche
Dr_Dan
★★  

Germany,
2011-06-16 14:37
(4669 d 04:14 ago)

@ swapnil.kuche
Posting: # 7136
Views: 9,837
 

 Redosing study…

Dear both
A redosing study is acceptable for the FDA but not for the EMEA
Based on data from the pivotal study, an acceptance range for the ratios - mean ±3 standard deviations - should be calculated for AUC and Cmax using the log transformed data. For each subject, a ratio (test/reference) should be calculated for AUCt, and Cmax based on data obtained in the re-dosing study. The ratios should be compared to the acceptance range.
The following scenarios are possible:
• If the ratios of AUCt, and Cmax of the suspected outlier lie outside the corresponding acceptance range and all the ratios of AUCt and Cmax for the control subjects fall within the acceptance range, then the outlying subject will be confirmed as an outlier. The pharmacokinetic and statistical analysis for the pivotal study will then be performed without data for the outlying subject.
• If the ratios of AUCt and Cmax for the suspected outlier lie inside the corresponding acceptance range and all the ratios of AUCt and Cmax for the control subjects fall within the acceptance range, then the outlying subject from the pivotal study will not be confirmed as an outlier.
• If the ratios of AUCt and Cmax for one or more of the control subjects lie outside the corresponding acceptance range, then this re-dosing study will be inconclusive with regard to the outlying subject.
All subjects to be included in the redosing study have to have participated in the pivotal study. In the study protocol of the redosing study you have to describe how to select these subjects. A special rule how to select does not exist.
I hope this helps
Kind regards
Dan

Kind regards and have a nice day
Dr_Dan
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2011-06-16 14:52
(4669 d 03:58 ago)

@ Dr_Dan
Posting: # 7138
Views: 9,842
 

 Redosing study…

Dear Dan!

❝ A redosing study is acceptable for the FDA […]



Should read ‘A redosing study may be acceptable for the FDA – but still they hate it.’

❝ Based on data from the pivotal study, an acceptance range for the ratios - mean ±3 standard deviations - should be calculated for AUC and Cmax using the log transformed data.


I’ve heard about this ‘method’ – do you have a reference? Assuming a normal distribution (after transformation) mean ±3SD covers 99.73% of the data. In other words, the sample size should be 370 in order to be able to ‘catch’ a single outlier.

Canada’s approaches are more reasonable, IMHO (cave: e.g., nonparametrics!).

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
GSTATS
☆    
Homepage
India,
2011-06-16 18:12
(4669 d 00:38 ago)

@ Dr_Dan
Posting: # 7141
Views: 9,560
 

 Redosing study…

Dear Dr Dan,

❝ The following scenarios are possible:

❝ • If the ratios of AUCt, and Cmax of the suspected outlier lie outside the corresponding acceptance range and all the ratios of AUCt and Cmax for the control subjects fall within the acceptance range, then the outlying subject will be confirmed as an outlier. The pharmacokinetic and statistical analysis for the pivotal study will then be performed without data for the outlying subject.


I have doubt in the scenario mentioned above. If the suspected subject repeats the behavior as in pivotal study (i.e very low value of PK parameter for test as compared to Reference or vice-versa) then the behavior of the subject is not by chance and it may repeat for other subjects too (in larger pool of subjects), with same biological conditions. And for these particular type of subjects, two products are not bio-equivalent.

So, in my opinion, if suspected and control subjects doesn't show an outlying behavior in re-dosing study then we can exclude that subject from the pivotal study.

Regards,
GSTATS

Let Noble Thoughts come from Every Side: RIG VEDA
luvblooms
★★  

India,
2011-06-17 08:34
(4668 d 10:16 ago)

@ Dr_Dan
Posting: # 7144
Views: 9,513
 

 Redosing study…

Dear Dr. Dan

Now I will give you 2 new scenario if

1. The identified Volunteer shows totally opposite behavior then the main study?
E.g.: In main study T/R ratio was lets say 1500% and in re-dosing study it becomes 15%. In that condition what can one do?? :angry:

this a case of abnormal volunteer behavior only but how to justify it?

2. If one/more that one control subjects behave as an outlier in re-dosing study??? :no:

Luv

P.S.: These conditions are normal (real life instances) for re-dosing study .

~A happy Soul~
ElMaestro
★★★

Denmark,
2011-06-17 11:16
(4668 d 07:34 ago)

@ luvblooms
Posting: # 7145
Views: 9,488
 

 Redosing study…

Hello lb,

❝ 1. The identified Volunteer shows totally opposite behavior then the main study?

E.g.: In main study T/R ratio was lets say 1500% and in re-dosing study it becomes 15%. In that condition what can one do?? :angry:


Reformulate :-D

Pass or fail!
ElMaestro
luvblooms
★★  

India,
2011-06-17 14:50
(4668 d 04:00 ago)

@ ElMaestro
Posting: # 7149
Views: 9,374
 

 Redosing study…

Dear EM

Well

❝ ❝ 1. The identified Volunteer shows totally opposite behavior then the main study?

❝ ❝ E.g.: In main study T/R ratio was lets say 1500% and in re-dosing study it becomes 15%. In that condition what can one do?? :angry:


❝ Reformulate :-D


But that is not the problem of formulation as all others (65 volunteer out of 66) gave normal results in the range of 80-125% TR ratio (with min of 34% and max of 244%) and only one was acting weird.

If such is the case, even after reformulation same volunteer or someone of the same category :crying: could be there to kill your study!!

Then What??

Again reformulation ;-)

Luv

~A happy Soul~
ElMaestro
★★★

Denmark,
2011-06-17 15:09
(4668 d 03:42 ago)

@ luvblooms
Posting: # 7150
Views: 9,412
 

 Redosing study…

Hello LB,

❝ But that is not the problem of formulation as all others (65 volunteer out of 66) gave normal results in the range of 80-125% TR ratio (with min of 34% and max of 244%) and only one was acting weird.


❝ If such is the case, even after reformulation same volunteer or someone of the same category :crying: could be there to kill your study!!


Yes, this spells trouble. Highly variable drug 'in some patients'. Or highly variable patients among your study population. I think this is where recruitment principles could help you in the future - no guarantees though. I can't imagine a solution in the concrete situation, especially not when non-parametrics are discouraged.

Pass or fail!
ElMaestro
luvblooms
★★  

India,
2011-06-18 13:47
(4667 d 05:04 ago)

@ ElMaestro
Posting: # 7153
Views: 9,313
 

 Redosing study…

Hello EM,

❝ Yes, this spells trouble. Highly variable drug 'in some patients'. Or highly variable patients among your study population.


Yes!!!! But again issue is how to identify the highly variable volunteer.

❝ I can't imagine a solution in the concrete situation, especially not when non-parametrics are discouraged.


Now planning to put a day housing before the study to minimize the physiological based variation. Let's see.

Luv

~A happy Soul~
GSTATS
☆    
Homepage
India,
2011-06-17 18:57
(4667 d 23:54 ago)

@ luvblooms
Posting: # 7152
Views: 9,393
 

 Redosing study…

Dear LB

❝ ❝ ❝ 1. The identified Volunteer shows totally opposite behavior then the main study?

❝ ❝ ❝ E.g.: In main study T/R ratio was lets say 1500% and in re-dosing study it becomes 15%. In that condition what can one do?? :angry:


I will say theoretically it is not possible and practically it is possible only when the subject was in specific condition (vomit, under stress or anything) for one product in pivotal study and under same condition for another product in re-dosing study which means poor conduct of study.

❝ But that is not the problem of formulation as all others (65 volunteer out of 66) gave normal results in the range of 80-125% TR ratio (with min of 34% and max of 244%) and only one was acting weird.


❝ If such is the case, even after reformulation same volunteer or someone of the same category :crying: could be there to kill your study!!


❝ Then What??


Even if in a re-dosing study, subject repeats its anomalous behavior then no one can save your study. OR In place of going for re-dosing study, one can do an investigation and give solid clinical justification for such an outlying behavior with "statistical outlier test" confirming the value as an outlier.

Regards,
GSTATS
www.gstatsolutions.com

Let Noble Thoughts come from Every Side: RIG VEDA
luvblooms
★★  

India,
2011-06-18 13:53
(4667 d 04:58 ago)

@ GSTATS
Posting: # 7154
Views: 9,414
 

 Redosing study…

Dear GS

❝ I will say theoretically it is not possible and practically it is possible only when the subject was in specific condition (vomit, under stress or anything) for one product in pivotal study and under same condition for another product in re-dosing study which means poor conduct of study.


Well, the gap between the study was more than 4 months and I am still not sure that in 4 months also his worries has not been solved :-D (there were no vomiting) and also I am unable to understand the type of worry which completely change the GI behaviour in 4 months and that too in a formulation specific way.

❝ In place of going for re-dosing study, one can do an investigation and give solid clinical justification for such an outlying behavior with "statistical outlier test" confirming the value as an outlier.


But how to justify different behaviour at different time.

Luv

~A happy Soul~
UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,638 registered users;
77 visitors (0 registered, 77 guests [including 7 identified bots]).
Forum time: 17:51 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