Bioequivalence and Bioavailability Forum

Dear Martin & Harish!

❝ Källén A. (2008). Computational Pharmacokinetics. Chapman and Hall / CRC

❝ Biostatistics Series.

Yes, this is a nice one!

Quoting pp 66 from Chapter 4:
This reaction [the formation of the drug-protein complex] is usually very fast (milliseconds) and we can therefore assume that it is always in equilibrium [...]
The reasons for such fast binding/release are twofold:

1. the surplus of protein compared to drug, and mainly
   
2. the weak binding van der Waals forces.

So if you are dealing with any kind of extraction (SPE, LLE), the drug will be released from the protein; we are always measuring total (OK, unless we perform ultrafiltration on every sample - what a mess).
There are a few exceptions where molecules are trapped in precipitated protein - but this would mainly affect recovery and both accuracy/precision of the method.

There was a lengthy debate about measuring total vs. free drug in BE studies more than 15 years ago which ended in the consensus of measuring total even for highly bound drugs whithout any correction factors whatsoever.
BTW your example should read like this:
  measured: 50ng/mL (this is total!)
  free:     50 x  2% =  1ng/mL
  bound:    50 x 98% = 49ng/mL
Your BE assessment wil not be affected whether you apply such a factor or not.
Introduction of a factor from the literature does not make any sense. If you really want to start fiddeling around with ultrafiltration consider sneeking around at Millipore, but keep in mind:

• in your case you would have to develop a method which is 50times more sensitive
  
• nobody asks you to do so...