Bioequivalence and Bioavailability Forum

Hello David!

❝ [...] I am very excited to find a whole forum dedicated to bioequivalence.

Welcome to the club!

❝ Is the planning of alternate subjects acceptable for BE studies intended

❝ for EMEA submissions?

Not only acceptable, but general practice.
The expected number of drop-outs depends on

• the number of study periods (2×2, higher order Xover, replicate designs),
  
• the number of treatments (single dose v.s. multiple dose),
  
• the drug / formulation itself.

10%-20% is quite common.

❝ The following guidelines are provided in the EMEA general BE guidance:

❝ "All results should be clearly presented and should include data from

❝ subjects who eventually dropped-out. Drop-out and withdrawal of subjects

❝ should be fully documented and accounted for."

❝ Source: (section 3.8)

❝ http://www.emea.europa.eu/pdfs/human/ewp/140198en.pdf

❝ I have seen different companies interpret this statement in different

❝ ways, and I was wondering if there was a general consensus.

There's no consensus, because - according to my knowledge - there was never a public discussion in the EU.

If you want to apply directly at the EMEA (the centralised procedure) it's quite easy: ask them.
If you want to go with an MRP (the mutual recognition procedure), i.e., national market authorisation in one member state, followed by any number of other members states it will be a little more difficult.
Following the implementation of Directive 2001/20/EC in summer 2004 all study protocols have to be approved by both the local ethic committee and the competent authority. In my experience the majority of IECs have the following point of view:
'It's unethical to treat healthy subjects and not to use potentially available data.'
Therefore samples of all dosed subjects have to be analysed and used in the BE assessment even risking to overpower the study - which may be present a problem in some member states (e.g., Denmark, Belgium, Sweden) if the confidence interval does not include 100% (statistically significant treatment difference).
On the other hand I've seen studies following exactly your procedure. If you opt for not analysing samples of alternates, they should be kept frozen and long-term stability of the analytical method regularily updated - just in case you may need them later on.
As you see in the EU with it's 27 member states - although given the same regulations and guidelines - current (or even future) practice may be quite different.