auditor
★    

India,
2013-11-14 11:38
(3787 d 06:15 ago)

Posting: # 11884
Views: 10,976
 

 Outlier in a BE study [Outliers]

Hello to all,

We have conducted the pivotal study. In that one subject is showing different profile from rest of all. We have identified that subject as an outlier from Lund's test. By including that subject in statistical analysis study is not meeting BE criteria and excluding that subject study is meeting all BE criteria.

Now how to present this data to regulatory (including and excluding outlier or just excluding)?
whether regulatory accepts the such these type of data with outliers? Please explain the about these as early as possible.

Regards,
Auditor
Dr_Dan
★★  

Germany,
2013-11-14 15:41
(3787 d 02:12 ago)

@ auditor
Posting: # 11887
Views: 9,841
 

 Outlier in a BE study

Dear Auditor
According to CPMP/QWP/EWP/1401/98 Rev. 1 unbiased assessment of results from randomised studies requires that all subjects are observed and treated according to the same rules. These rules should be independent from treatment or outcome. In consequence, the decision to exclude a subject from the statistical analysis must be specified in the protocol and made before bioanalysis. Exclusion of data cannot be accepted on the basis of statistical analysis or for pharmacokinetic reasons alone, because it is impossible to distinguish the formulation effects from other effects influencing the pharmacokinetics. Bad luck :-(
However, you can try to present both study results (with and without outlier) to the authorities but in this case you need a reasonable explanation why the subject under discussion is an oulier (cave: Exclusion of data cannot be accepted on the basis of statistical analysis or for pharmacokinetic reasons!). You need to find evidence in the subject´s medical history, behaviour, state of health etc. which allows to assume that this subject behaves different.
If the subject shows a different profile for the reference product you might have a chance. If you want to exclude the subject due to a different profile under test, no way! :no:
Hope this helps
Kind regards
Dan

Kind regards and have a nice day
Dr_Dan
Erkin
●    

2014-01-21 20:21
(3718 d 21:33 ago)

@ Dr_Dan
Posting: # 12222
Views: 9,317
 

 Outlier in a BE study

Dear Dr_Dan and Auditor,

I agree with Dr_Dan.

❝ According to CPMP/QWP/EWP/1401/98 Rev. 1 unbiased assessment of results from randomised studies requires that all subjects are observed and treated according to the same rules. These rules should be independent from treatment or outcome. In consequence, the decision to exclude a subject from the statistical analysis must be specified in the protocol and made before bioanalysis. Exclusion of data cannot be accepted on the basis of statistical analysis or for pharmacokinetic reasons alone, because it is impossible to distinguish the formulation effects from other effects influencing the pharmacokinetics. Bad luck :-(


One of our study, we've the same situation that auditor have. We've decided to give both study results (with and without outlier). But, we've a note to file from investigator in the subject's CRF that there has been a suspicion about vomitting.

I might be wrong but in my opinion giving both results (with and without outlier) are the best choice if you have other evidence other than statistics.

Kind Regards,

Erkin
mittyri
★★  

Russia,
2014-01-30 21:14
(3709 d 20:39 ago)

@ Erkin
Posting: # 12298
Views: 9,090
 

 Outliers: reasons to not analyse?

Dear Auditor, Dr_Dan, Erkin and all members,

Could you explain your practice and guideline interpretation:

If we state in the protocol the reasons to exclude the results of subjects from analysis (according to the EMA guide), does it mean that the analysis of these samples should be performed and presented anyway?
e.g. vomiting was begun in 1 hour after dosing. After that the volunteer was excluded from the study. The plasma concentration-time profile will be evedently unreliable. should we perform the sample analysis of this subject?

Could you indicate any reasons/situations when the samples analysis can be not performed for some of subjects?

PS I met very strange opinion to this issue. I'm in doubt at this moment :confused:

Kind regards,
Mittyri
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2014-02-06 03:35
(3703 d 14:19 ago)

@ mittyri
Posting: # 12352
Views: 9,010
 

 I wouldn’t do that

Hi Mittyri,

❝ Could you explain your practice and guideline interpretation:


I don’t base my decisions primarily on guidelines. ;-) You never know why a subject drops out. What if vomiting was drug-related or even a SUSAR? The sponsor has the budget to analyse all samples, so do it – as far as they are available. Some background:
PS: EMA’s GL is ambiguous: “Drop-out and withdrawal of subjects should be fully documented. If available, concentration data and pharmacokinetic parameters from such subjects should be presented in the individual listings, but should not be included in the summary statistics.”
I would say: Make them available.

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
jag009
★★★

NJ,
2013-11-15 23:06
(3785 d 18:47 ago)

@ auditor
Posting: # 11894
Views: 9,663
 

 Outlier in a BE study

Hi,

Take a look at previous posts. The topic has been discussed before.

❝ Please explain the about these as early as possible.


Please refrain from using such a (strong) statement. This is a free forum and we are always willing to help but we do have our own things to take care of. :-)

Thanks
John
riteshsrmc
☆    

India,
2014-01-16 13:44
(3724 d 04:09 ago)

@ auditor
Posting: # 12177
Views: 9,389
 

 Outlier in a BE study

Hi Auditor,

By the way to which agency you are filing the product? Agreed with Dan! EMEA has clearly defined the situation in guidance. I would suggest you to look at FDA perspective as well, if you are filing to FDA. FDA does not encourage redosing to handle outliers subjects incase of non-replicate design study wherein you may carry out redosing incase of replicate design. You should be clear they way agency look at the data with or without outliers.
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2014-01-16 15:54
(3724 d 02:00 ago)

@ riteshsrmc
Posting: # 12181
Views: 9,307
 

 Outlier in a BE study

Hi Kumar,

❝ FDA does not encourage redosing to handle outliers subjects incase of non-replicate design study…


Correct.

❝ … wherein you may carry out redosing incase of replicate design.


I don’t think so – at least in a fully replicated design (see this post). In a partial replicate – if the “outlier” occurred after T – I would not perform a re-dosing study without contacting FDA OGD’s review stuff beforehand.

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
luvblooms
★★  

India,
2014-01-27 10:11
(3713 d 07:42 ago)

@ auditor
Posting: # 12273
Views: 9,379
 

 Outlier in a BE study

Dear Auditor

❝ Now how to present this data to regulatory (including and excluding outlier or just excluding)?

❝ whether regulatory accepts the such these type of data with outliers? Please explain the about these as early as possible.


I came across a paper “For-Cause” Inspections in Bioequivalence Studies Submitted to the Food and Drug Administration" written by FDA hot-shots ;-) and published in January 2013 in AAPS

In this article, they have not directly mentioned that they do accept re dosing study or not; but have mentioned the redosing study as in example or reason of For-Cause Inspections in Bioequivalence Studies.

Please look in the section Improper study design and conduct examples were it is mentioned as Insufficient number of control subjects in re-dosing study, and lack of SOP in effect at the time of the study related to the conditions that warrant the performance of an outlier test.


So IMHO, if you have a proper justification and procedures in place to justify the outlier test, you may approach to FDA with re-dosing study or contact them through CC for their opinion.


I may be wrong, but this is worth trying...


Regards

~A happy Soul~
UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,638 registered users;
85 visitors (1 registered, 84 guests [including 7 identified bots]).
Forum time: 17:54 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