pash413
★    

India,
2017-10-13 15:10
(2358 d 18:17 ago)

Posting: # 17894
Views: 4,014
 

 Ambulatory missing blood samples in EMEA study [Regulatives / Guidelines]

Dear All
In BE study of modified release drug product for EMEA submission can we dropped the subject in clinical phase if he/she was not reported for continuous three or more ambulatory blood sample collections.

If we dropped the subject we need to analyse the renaming blood samples as per EMEA GL. However we should not included in final statastical analysis.


As per EMEA GL "Subjects should not be excluded from the statistical analysis if AUC(0-t) covers less than 80% of AUC(0-∞), but if the percentage is less than 80% in more than 20% of the observations then the validity of the study may need to be discussed. This does not apply if the sampling period is 72 h or more and AUC(0-72h) is used instead of AUC(0-t)".

Is there any regulatory concern if we have not included the drop out subject(s) due to ambulatory missing samples in the statastical analysis.
ElMaestro
★★★

Denmark,
2017-10-15 18:56
(2356 d 14:31 ago)

@ pash413
Posting: # 17897
Views: 3,315
 

 Ambulatory missing blood samples in EMEA study

Hello Pash413,

❝ Is there any regulatory concern if we have not included the drop out subject(s) due to ambulatory missing samples in the statastical analysis.



as I see it there is not necessarily a need to exclude a subject just because some values are missing; there may be a need for some deeree of exclusion if the data do not add to your understanding of comparative PK (/safety/efficacy). If you do not have an SOP in place then I think you could do a primary analysis on those subjects for which you have adequate characterisation however many blood samples are missing. NCA is per se not ruined by missing values. Typically missing values are in the ambulatory (and thus terminal) phase, so it would sometimes mean you can do AUCtlast, but not AUCinf.
At the end of the day if this dossier is submitted in EU or US you can be quite sure that someone will check one way or another if the ones you chose to exclude seemed to affect the conclusion of BE in the subjective way you excluded them. So whatever you do, exclude or include on basis of something very objective with zero regard to the outcome, or you could get flagged and flogged.
Excluding data is always a regulatory concern for principal reasons.

Pass or fail!
ElMaestro
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2017-10-16 12:41
(2355 d 20:46 ago)

@ pash413
Posting: # 17898
Views: 3,445
 

 Excluding PK-metric(s) instead of subjects

Hi Pash,

❝ In BE study of modified release drug product for EMEA submission …


Nitpicking: EMEA changed its name to EMA back in 2009. ;-)

❝ Is there any regulatory concern if we have not included the drop out subject(s) due to ambulatory missing samples in the statastical analysis.


I agree with ElMaestro. Furthermore keep in mind that

Guidelines
are guidelines
are guidelines.
   Henrike Potthast (BfArM, member of the CHMP’s PKWP)


Though the GL speaks about excluding subjects only, you should not take that literally. My procedure:
  • If no reliable estimate of λz possible → keep Cmax and AUC0–t (exclude AUC0–∞ of the subject from the comparison).
  • If no clearly defined Cmax (e.g., ~two increasing concentrations before and two to three decreasing concentrations afterwards) → exclude the subject from the comparison.
Since in most cases the variability follows the order Cmax > AUC0–∞ > AUC0–t and likely the study is powered for the PK-metric with the highest variability the impact on power is low (exception: reference-scaling for Cmax together with highly variable AUC).
An SOP is fine but I suggest to state the rules for exclusion already in the protocol. Referring in the report to an SOP (which is not available to the assessor) may rise questions and delay the approval.

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;
89 visitors (0 registered, 89 guests [including 4 identified bots]).
Forum time: 08:28 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