T vs. R1 and T vs. R2 [Design Issues]

posted by Helmut Homepage – Vienna, Austria, 2018-09-14 12:52 (946 d 23:06 ago) – Posting: # 19289
Views: 2,714

Hi Rajasekhar,

» […] study with 3 arms where 1 arm is test product, 2nd arm is US (RLD) & third arm is UK (RLD)


There are no “reference-listed drugs” in Europe.
A product can be either approved via the centralised procedure (directly at the EMA) – then it might by called a “European reference product” (though this term doesn’t officially exist) – or it can be approved via the decentralised procedure (one country acting as the reference member state and others as concerned member states).
In either case you can use any one of the local products marketed in the EEA (EU + Iceland, Norway, Liechtenstein).
I don’t recommend a reference product from the UK unless you are very fast (see this post).


» whether this is permitted & acceptable to both the regulatory authorities.

Sure. For the evaluation I suggest avoiding a pooled analysis (i.e., based on a common variance in ANOVA) but exclude the respective other treatment from the analysis (whilst keeping the codes for period and sequence) – resulting in two incomplete block designs (one for the comparison of T with one of the European originator’s products and the other one for the comparison of T with the US RLD).

     ┌────────────┐  →       ┌────────────┐ and      ┌────────────┐
     │ p1  p2  p3 │          │ p1  p2  p3 │          │ p1  p2  p3 │
┌────┼────────────┤     ┌────┼────────────┤     ┌────┼────────────┤
│ s1 │ T   EU  US │     │ s1 │ T   EU  ·  │     │ s1 │ T   ·   US │
│ s2 │ EU  US  T  │     │ s2 │ EU  ·   T  │     │ s2 │ ·   US  T  │
│ s3 │ US  T   EU │     │ s3 │ ·   T   EU │     │ s3 │ US  T   ·  │
│ s4 │ T   US  EU │     │ s4 │ T   ·   EU │     │ s4 │ T   US  ·  │
│ s5 │ EU  T   US │     │ s5 │ EU  T   ·  │     │ s5 │ ·   T   US │
│ s6 │ US  EU  T  │     │ s6 │ ·   EU  T  │     │ s6 │ US  ·   T  │
└────┴────────────┘     └────┴────────────┘     └────┴────────────┘

See also the EMA’s BE-GL, Section 4.1.8

In studies with more than two treatment arms (e.g. a three period study including two references, one from EU and another from USA […]), the analysis for each comparison should be conducted excluding the data from the treatments that are not relevant for the comparison in question.

… and this post.

Don’t forget that for the FDA the ABE-model treats subjects as a random effect (e.g., by SAS PROC MIXED) whereas the EMA requires all effects fixed (e.g., by SAS PROC GLM). Make that clear in the SAP.

Dif-tor heh smusma 🖖
Helmut Schütz

The quality of responses received is directly proportional to the quality of the question asked. 🚮
Science Quotes

Complete thread:

 Admin contact
21,419 posts in 4,475 threads, 1,509 registered users;
online 9 (0 registered, 9 guests [including 4 identified bots]).
Forum time: Sunday 11:59 UTC (Europe/Vienna)

Nothing fails like success because you do not learn anything from it.
The only thing we ever learn from is failure.
Success only confirms our superstitions.    Kenneth E. Boulding

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