Sereng
☆    

USA,
2023-03-13 18:08
(350 d 01:46 ago)

Posting: # 23494
Views: 2,730
 

 Definition of Full Re­pli­cate [Design Issues]

Dear colleagues, my associate is using the term "full replicate" to describe a four-way crossover BE study of Test (Fed), Test (Fasted), Reference (Fed) and Reference (Fasted), with each treatment given only once. In my view, this is not an appropriate us of the term "full replicate" but I want to be certain. Additionally, what is the appropriate term, "Full Replicate" or "Fully Replicate"? Regards

Biostatistically Challenged CEO
Mahmoud
★    

Jordan,
2023-03-13 21:27
(349 d 22:27 ago)

@ Sereng
Posting: # 23495
Views: 2,520
 

 Definition of Full Replicate

Dear Sir
=========
You can use the following design


1 B-A-C-D-A-C

2 A-D-B-C-C-A

3 D-C-A-A-B-C

4 C-A-D-C-A-B

5 A-C-C-B-D-A

6 C-B-A-A-C-D


SABE for fast and food accepted by FDA

Six period with six sequences

B= Test under fast

A= Ref under fast

D= Test under fed

C= Ref under fed
Sereng
☆    

USA,
2023-03-14 00:38
(349 d 19:16 ago)

@ Mahmoud
Posting: # 23496
Views: 2,546
 

 Definition of Full Replicate

Hi Mahmoud, thank you for your response. I think you may have misunderstood my question. I have always understood a "full replicate" design to be one where each treatment condition is repeated twice. Therefore, if the Test was being compared with Reference in the fasted state, a full replicate would involve giving both the Test and Reference on two occasions. Is this correct? Furthermore, calling a four-way fed fasted crossover study (Test Fed & Fasted, and Ref Fed & fasted), each given once a Full Replicate study would be wrong. is this correct?

Biostatistically Challenged CEO
Mahmoud
★    

Jordan,
2023-03-14 09:44
(349 d 10:10 ago)

@ Sereng
Posting: # 23497
Views: 2,464
 

 Definition of Full Replicate

Dear Sir
=======
If you neeed to genrate two studies fed and fast separately
you can use the following design

sequence 1 TRRT
sequence 2 RTTR

but the best to do only one study for fed and fast with six sequence and six period
Achievwin
★    

US,
2023-03-27 21:08
(335 d 23:46 ago)

@ Mahmoud
Posting: # 23513
Views: 2,097
 

 Definition of Full Replicate

Dear Experts

Can TRT, RTR be called two sequence full replicate in three way cross over? can it be a substitute for RTRT or TRTR, or TTRR or RRTT (in four period fully replicated design)?

The goal is to do a 3-period cross-over yet to capture ISCV for Test and Reference separately. Can this type of design address RSABE for warfarin by FDA?

(Nothing) Achievwin:-D:confused:
mittyri
★★  

Russia,
2023-03-28 00:16
(335 d 20:38 ago)

@ Achievwin
Posting: # 23514
Views: 2,102
 

 Full Replicate with 3 periods

Dear Achievwin!

❝ Can TRT, RTR be called two sequence full replicate in three way cross over?


I would say in three periods. For me 'a way' is closer to the sequence (subject's way :-D). But that's just my feeling, I see many people are using it, so your wording is rather correct.

❝ can it be a substitute for RTRT or TRTR, or TTRR or RRTT (in four period fully replicated design)?

Hey! Look at the fresh FDA draft
Bloody partial replicate goes away and is being substituted to that design in GL. I would opt for it.

Kind regards,
Mittyri
Achievwin
★    

US,
2023-03-28 05:06
(335 d 15:48 ago)

@ mittyri
Posting: # 23516
Views: 2,058
 

 Full Replicate with 3 periods

Fantastic advise thank you

❝ Bloody partial replicate goes away and is being substituted to that design in GL. I would opt for it.

Helmut
★★★
avatar
Homepage
Vienna, Austria,
2023-03-28 11:03
(335 d 09:51 ago)

@ mittyri
Posting: # 23517
Views: 2,108
 

 Full Replicate with 3 periods

Hi mittyri & Achievwin,

❝ ❝ can [TRT, RTR] be a substitute for RTRT or TRTR, or TTRR or RRTT (in four period fully replicated design)?

❝ Hey! Look at the fresh FDA draft

❝ Bloody partial replicate goes away and is being substituted to that design in GL.


Yes and no. The partial replicate was nowhere mentioned in the previous (2001) guidance. On page 8 (after the 2-sequence 4-period full replicate) we read:

Other replicated crossover designs are possible. For example, a three-period design, as shown below, could be used.

$$\small{\begin{array}{rcccc}
& & & \text{Period}\\\\
& & 1 & 2 & 3\\\\
& 1 & \textbf{T} & \textbf{R} & \textbf{T}\\
\text{Sequence}\\
& 2 & \textbf{R} & \textbf{T} & \textbf{R}
\end{array}}$$ However, the partial replicate (and not the TRT|RTR) is still stated in the ANDA draft

❝ I would opt for it.


So would I. Not at least because of the nasty issues of the partial replicate in A.

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
Brus
★    

Spain,
2023-03-14 11:10
(349 d 08:44 ago)

@ Sereng
Posting: # 23498
Views: 2,421
 

 Definition of Full Replicate

Dear Sereng,

From my understanding, this is not a replicate study. 4-ways is not always a replicate. There can be 4 periods but with 4 different formulations, or in different administration conditions, and in those cases it would not be a "full replicate". In your case, according to my humble opinion it is not a "full replicate" since they are administered in different conditions (fast VS fed). In your case, I think it would be more accurate to call it "4-ways food effects". At least in our own previous experiences, this is the way we have approached it.

Best regards,
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2023-03-14 11:37
(349 d 08:17 ago)

@ Sereng
Posting: # 23499
Views: 2,477
 

 Replicate Design ≠ Higher-Order Cross­over

Hi Sereng,

❝ … my associate is using the term "full replicate" to describe a four-way crossover BE study of Test (Fed), Test (Fasted), Reference (Fed) and Reference (Fasted), with each treatment given only once. In my view, this is not an appropriate us of the term "full replicate" but I want to be certain.


You are right. For some of the designs see the vignette of the R package replicateBE.
For the named food-interaction study you have three options. Primarily you are interested to demonstrate BE in fasted and fed state. The assessment of food effects of test and reference is just ‘nice to know’.
  1. Two separate 2×2×2 studies AB|BA and CD|DC, where A = Tfed, B = Rfed, C = Tfasting, D = Rfasting. Esp. for controlled release products BE in fed state is more difficult due to potentially different food effects of T and R. Hence, I suggest to perform this study first.
  2. A higher-order crossover in a Williams’s design. You can use any of the six designs ADBC|BACD|CBDA|DCAB, ADCB|BCDA|CABD|DBAC, ACDB|BDCA|CBAD|DABC, ACBD|BADC|CDAB|DBCA, ABDC|BCAD|CDBA|DACB, or ABCD|BDAC|CADB|DCBA. What you must not use is the Latin Square ABCD|BCDA|CDAB|DABC. Only from a Williams’ design you can extract balanced pairwise comparisons (see there).
  3. Like #1 but in the same subjects and the second following immediately after the first. Essentially it’s one study with two parts.
#1 is most simple and allows to perform both studies in either fixed-sample or adaptive two stage designs. However, assessment of food effects is done in parallel groups and therefore, has low power.
Whilst #2 is popular, there are drawbacks. If variabilities in fasted and fed state are different (commonly fed > fasted), you have to base the sample size on the worst case.
#3 is for the courageous ones (though all of my studies were accepted by European agencies :-D). Both parts (fed and fasting) are TSDs like in #1. You start with a best guess sample size for the condition with lower variability. Food effects can by assessed as paired comparisons (assuming identical period effects), which is much more powerful than the parallel comparison in #1.

❝ Additionally, what is the appropriate term, "Full Replicate" or "Fully Replicate"?


You can either say “full replicate design” or “fully replicated design”. In the former term you use an adjective and in the latter adverbs.

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
Sereng
☆    

USA,
2023-03-14 21:08
(348 d 22:46 ago)

@ Helmut
Posting: # 23500
Views: 2,399
 

 Thank You!

Hi Helmut, you are a Godsend! BTW, have you considered monetizing this site, even if it is for a modest one-time or annual subscription fee? This could also help you with resource allocation. Just a thought...... Regards

Biostatistically Challenged CEO
Relaxation
★    

Germany,
2023-03-20 09:56
(343 d 09:58 ago)

@ Helmut
Posting: # 23506
Views: 2,232
 

 Replicate Design ≠ Higher-Order Cross­over

Hi everybody and sorry for posting this additional question late.
But I cannot get to a clear answer by myself.

Helmut, you stated

❝ What you must not use is the Latin Square ABCD|BCDA|CDAB|DABC. Only from a Williams’ design you can extract balanced pairwise comparisons


I somehow fail to see why the extracted "pairs" are not considered balanced.
They are not balanced for carry-over (which was always the most important reason for Williams), but aren't they for period and sequence?
Is this what you mean here?

I have no idea how to make it fancy, so I will try an ASCII-style
Sequence | Treatments | IBD for A vs B 
1        | ABCD       | AB..
2        | DABC       | .AB.
3        | CDAB       | ..AB
4        | BCDA       | B..A

So A is present once in each sequence and period and the same for B.

Best regards,
Relaxation.
mittyri
★★  

Russia,
2023-03-28 01:01
(335 d 19:53 ago)

@ Relaxation
Posting: # 23515
Views: 2,019
 

 imbalance in Latin Square

Dear Relaxation!

let me cite the classical book written by Byron Jones and Michael G. Kenward:

❝ Although this design has maximum efficiency in the absence of carry-over effects, it is not as efficient at estimating the direct treatment effects as it could be. To achieve the highest possible efficiency the design must be balanced. The term balance refers to the combinatorial properties that the design must possess. In a balanced design, not only does each treatment occur once with each subject, but, over the whole design each treatment occurs the same number of times in each period and the number of subjects who receive treatment i in some period followed by treatment j in the next period is the same for all i != j. The design in Table 4.1[Latin square] is not balanced because treatment A is followed three times by treatment B and treatment B is never followed by treatment A.


Kind regards,
Mittyri
Helmut
★★★
avatar
Homepage
Vienna, Austria,
2023-04-21 12:01
(311 d 08:53 ago)

@ Relaxation
Posting: # 23533
Views: 1,388
 

 Balance for all effects

Hi Relaxation,

❝ Hi everybody and sorry for posting this additional question late.


I’m also sorry for replying that late. Somehow I missed your post.

❝ Helmut, you stated

❝ ❝ What you must not use is the Latin Square ABCD|BCDA|CDAB|DABC. Only from a Williams’ design you can extract balanced pairwise comparisons


❝ I somehow fail to see why the extracted "pairs" are not considered balanced.

❝ They are not balanced for carry-over […] Is this what you mean here?


Yes. Three times A → B and once B → A.

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
22,912 posts in 4,806 threads, 1,635 registered users;
27 visitors (0 registered, 27 guests [including 5 identified bots]).
Forum time: 19:54 CET (Europe/Vienna)

[…] an inappropriate study design is incapable of answering
a research question, no matter how careful the subsequent
methodology, conduct, analysis, and interpretation:
Flawless execution of a flawed design achieves nothing worthwhile.    J. Rick Turner

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