ssussu
☆    

China,
2017-08-16 13:45
(2416 d 09:51 ago)

Posting: # 17690
Views: 9,240
 

 long half life drug bioequivalence study design [Design Issues]

Dear all,
I have questions on the BE study design of long half life drug, please kindly give me some advice. According to FDA GL:
You can use Cmax and a suitably truncated AUC to characterize peak and total drug exposure, respectively. For drugs that demonstrate low intrasubject variability in distribution and clearance, you can use an AUC truncated at 72 hours (AUC0-72 hr) in place of AUC0–t or AUC0–inf.
The question is
Should I just need to design the sampling time to 72hr or need to longer than 72hr?
If I just sampling for 72hr, how can I prove the drug is a long half life drug?

Best regards!
ElMaestro
★★★

Denmark,
2017-08-16 14:53
(2416 d 08:43 ago)

@ ssussu
Posting: # 17691
Views: 8,440
 

 long half life drug bioequivalence study design

Hi ssussu,

❝ Should I just need to design the sampling time to 72hr or need to longer than 72hr?

❝ If I just sampling for 72hr, how can I prove the drug is a long half life drug?


In my humble opinion and experience:
  • Up to 72hrs
  • You will perhaps see some subjects do not display much of an elimination phase within the sampling time span. It is actually not easy to define quantitatively how you deliver the proof of subjects not entering the elimination phase or not doing it much. The thinking is of course that you compare AUC72 to AUCinf but since AUCinf is not always estimated due to missing elimination constants direct numerical gymnastics on AUCinf and AUC72 isn't going to be possible in all cases.
    There are generally two explanations for not having an elimination constant: Too short sampling time or a lot of scatter.
    So, depending on how long the elimination phase is you may or not have have AUCinf for some subjects. As long as you can argue it isn't a scatter issue you will then be fine for those, and for the rest it may come down to a plain comparison of the AUCs.
It is not much of a deal, actually.

Pass or fail!
ElMaestro
ssussu
☆    

China,
2017-08-17 12:18
(2415 d 11:18 ago)

@ ElMaestro
Posting: # 17703
Views: 8,264
 

 long half life drug bioequivalence study design

Thank you, ElMaestro!
I have another question describe below(reply to Helmut), but I don't how to quote two people(you and Helmet) at one post at the same time. Sorry!
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2017-08-16 16:18
(2416 d 07:18 ago)

@ ssussu
Posting: # 17693
Views: 8,446
 

 Truncted AUC, low variability

Hi ssussu,

which is your “target” agency? If you want to submit the study to the (US)FDA you have to provide the protocol to the OGD for review anyway.
BTW, I never understood why the FDA allows AUC0–72 for drugs with “low intrasubject variability in distribution and clearance only. The latter might be possible to show based on published data (though difficult because one needs the subjects’ λz). The former is almost impossible without raw data and PK modeling. In BE we are interested in absorption (property of both the drug and the formulation). Distribution/elimination are solely properties of the drug. I don’t get the FDA’s rationale.

❝ If I just sampling for 72hr, how can I prove the drug is a long half life drug?


I agree with what ElMaestro wrote. You posted in the right category – that’s a design issue. Think about planing the washout. It is based on an expected half-life (literature, previous studies). No agency* asks for a demonstration that f.i. the washout was 5–10×t½ of subjects (only that there are no residual concentrations >5% of Cmax in higher periods).

It is an open question what a “long half-life drug” is. An numerous conferences both the FDA and the EMA refused to give a definition. Only the Canadian HPFB formerly stated t½ ≥24 hours.


  • In Taiwan one has to demonstrate that the saturation phase in multiple dose studies was “sufficient”. Hence these poor guys have to sample beyond τ

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
ssussu
☆    

China,
2017-08-17 12:15
(2415 d 11:21 ago)

@ Helmut
Posting: # 17702
Views: 8,380
 

 Truncted AUC, or AUCinf, which one

Dear ElMaestro &Helmut,
Thank you two for your reply!Thank you so much!

❝ which is your “target” agency? If you want to submit the study to the (US)FDA you have to provide the protocol to the OGD for review anyway.

Thank you! Our target anency is not FDA but we almost follow the FDA BE GL:-|.

I am still confused for another question:confused:. I don't know which endpoint should be chosen to specify in the protocol, AUC0-72,or instead of AUClast and AUCinf(e.g.if the half life is 18-40hr according to literature report). As Elmaestro wrote,"...You will perhaps see some subjects do not display much of an elimination phase within the sampling time span.", while some subjects do.
Then if I choose truncated AUC0-72 instead of AUClast and AUCinf in the protocol,while most of the subjects half-life is not long enough (i.e 18hr) , actually, in this situation, it is possible the time point of 72hr sample concentration is BQL or can not be detected, then what I shoud do?
Then if I choose AUClast and AUCinf instead of AUC0-72,while most of subjects half life is very long actually, then at this situation AUCinf will be not accurately estimated.
Could I pre specify in the protocol alternatively?

Best regards!
ElMaestro
★★★

Denmark,
2017-08-17 12:47
(2415 d 10:48 ago)

@ ssussu
Posting: # 17705
Views: 8,305
 

 Truncted AUC, or AUCinf, which one

Hi again ssussu,

❝ Then if I choose truncated AUC0-72 instead of AUClast and AUCinf in the protocol,while most of the subjects half-life is not long enough (i.e 18hr) , actually, in this situation, it is possible the time point of 72hr sample concentration is BQL or can not be detected, then what I shoud do?


Well... it would be a bit unusual to this often when the drug generally has an excessive long half-life but OK it might happen. Between-subject variability etc.
You may need to invest at least 90 seconds of your time in figuring this one out with pen and paper. You have some subjects that go below the LLOQ, which means they comfortably display an elimination phase. From those subjects you can routinely derive an elimination constant and extrapolate AUClast (least measurable) to AUCinfinity. This is where you sadly need to invest this obscenely large amount of your precious time: use the same logic and jommetry and extrapolate areas to 72 hrs. in stead of infinity. You'll end up with a full data listing of AUC72 estimates, and Bob's your uncle.
Good luck.

Pass or fail!
ElMaestro
ssussu
☆    

China,
2017-08-18 09:13
(2414 d 14:23 ago)

@ ElMaestro
Posting: # 17708
Views: 8,275
 

 Truncted AUC, or AUCinf, which one

Thank you again, ElMaestro!
I got you. And I will leave this thing to WinNonlin:-D .

Best regards!
jag009
★★★

NJ,
2017-08-18 01:50
(2414 d 21:46 ago)

@ Helmut
Posting: # 17707
Views: 8,275
 

 Truncted AUC, low variability

Hi Helmut,

❝ It is an open question what a “long half-life drug” is. An numerous conferences both the FDA and the EMA refused to give a definition. Only the Canadian HPFB formerly stated t½ ≥24 hours.


So far no problems for me w FDA. I've ran 6 - 10 of these with t1/2=24 and up (okay, 3 drugs w mean 24 hrs.)

John
UA Flag
Activity
 Admin contact
22,957 posts in 4,819 threads, 1,636 registered users;
74 visitors (0 registered, 74 guests [including 5 identified bots]).
Forum time: 22:36 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