In a recent addition to USP 37 GENERAL NOTICES AND REQUIREMENTS (link), USP describes that:
“Compliance with any of the [dissolution] tests does not assure bioequivalence or bioavailability”.
The above-mentioned statement clearly indicates or establishes the fact that compendial (USP) tests and, by extension, hundreds and hundreds of similar tests reported in the literature, cannot provide in vivo or bio-relevant results.
It is also important and/or critical to note that in most cases, there are no differences between methods suggested in compendia and what the manufacturers use during product development stages. In fact, it is commonly suggested and recommended that one should preferably follow the compendial suggested methods as a first choice. The point being, in reality, there are no differences between methods/testers whether they are used for compendial or bio-assessment purposes. Therefore, if compendial methods cannot assure BE/BA, then the same methods/testers cannot assure BE/BA in other situations, such as product development, bio-waiver, IVIVC etc.
It appears that people do not realize a potentially profound implication of this statement or view. Some even believe, and are promoting, that it is simply a clarification or “stating the obvious or of already known”. However, in my view, such an interpretation is incorrect. By stating and establishing that compendial dissolution tests do not assure BE/BA, it has clearly challenged all the regulatory guidelines and requirements for recommending dissolution tests for assessing BE/BA. In fact, in my view again, the statement and view have practically made related guidances useless.
Therefore, it is not a simple “stating the obvious or of already known” scenario. It is a profound development that people have not yet realized. However, I believe it is good and positive news concerning dissolution testing, i.e., by acknowledging the flaws of current practices, USP has opened the door for discussions in seeking modifications and improvements for dissolution testing.