Researcher101
☆    

Egypt,
2020-11-20 16:59
(6 d 18:21 ago)

Posting: # 22072
Views: 300
 

 Accepted time deviations in Ambulatory PK intervals [Regulatives / Guidelines]

What is the time limit deviations that are accepted (before and after) in the 24, 36, 48 and 72 h interval


Edit: Please follow the Forum’s Policy[Helmut]
ElMaestro
★★★

Belgium?,
2020-11-21 09:08
(6 d 02:13 ago)

@ Researcher101
Posting: # 22073
Views: 253
 

 Accepted time deviations in Ambulatory PK intervals

Hi R101,

» What is the time limit deviations that are accepted (before and after) in the 24, 36, 48 and 72 h interval

I do not know of any specific rules in any guideline.
As I see it, one can make a case for saying that the way to deal with time point deviations should be API-specific as well as time-specific.

Some CROs define a window in terms of time (for each time point); if the sample is taken with the window then the nominal time is used for PK (for AUC calculation), otherwise the actual time. Some CROs will always use the actual time. Both approaches seem to work well across jurisdictions. The option that does not work is to use nominal time regardless of sampling time.
All this requires a rock solid way of having the actual time points conveyed to the people doing PK. Some CROs will have the clinical people calculate the actual time used for PK and then let PK staff do their work on the data; other CROs will let PK-people scan over the sampling time to derive their actual time for AUC-calculation.

Another issue is if or when a deviation of sampling time is a protocol deviation. From the top of my head I am thinking most CROs specifically do not call a time point deviation a PD, but has another category for it. They for instance put those situations in an odd sock drawer called "Other types of deviations".

I could be wrong, but...

Best regards,
ElMaestro

No, of course you do not need to audit your CRO if it was inspected in 1968 by the agency of Crabongostan.
Researcher101
☆    

Egypt,
2020-11-22 10:52
(5 d 00:28 ago)

@ ElMaestro
Posting: # 22075
Views: 215
 

 Accepted time deviations in Ambulatory PK intervals

» Some CROs define a window in terms of time (for each time point); if the sample is taken with the window then the nominal time is used for PK (for AUC calculation), otherwise the actual time. Some CROs will always use the actual time. Both approaches seem to work well across jurisdictions. The option that does not work is to use nominal time regardless of sampling time.

So if I have a time window 12 h and I have a participant come after 18 h so Will I take this sample or not? Please note that the delay will be considered in the statistics calculations


Edit: Full quote removed. Please delete everything from the text of the original poster which is not necessary in understanding your answer; see also this post #5! Please follow the Forum’s Policy[Helmut]
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2020-11-23 10:49
(4 d 00:31 ago)

@ ElMaestro
Posting: # 22079
Views: 175
 

 Accepted time deviations in Ambulatory PK intervals

Hi ElMaestro and Researcher101,

» As I see it, one can make a case for saying that the way to deal with time point deviations should be API-specific as well as time-specific.

Agree. The later in the profile, the less critical deviations are.

» Some CROs define a window in terms of time (for each time point)…

See this post and followings. I don’t comprehend why too early sampling should be performed. PK calculations (regardless whether by NCA or by modeling) should always be performed based on the actual time points – otherwise the estimates will be biased.
Extremely nasty in modeling. Both OLS and maximum likelihood approaches require that the independent variable is free of any error. It’s a misconception that errors in the independent variable only increase the residual error. If one wants to deal with that, it get’s really, really tricky ([image] orthogonal regression, where the selection of appropriate weights is black magic).

Dif-tor heh smusma 🖖
Helmut Schütz
[image]

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes
Activity
 Admin contact
21,213 posts in 4,426 threads, 1,481 registered users;
online 19 (1 registered, 18 guests [including 13 identified bots]).
Forum time: Friday 11:21 UTC (Europe/Vienna)

Biostatistician. One who has neither the intellect for mathematics
nor the commitment for medicine but likes to dabble in both.    Stephen Senn

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