It is often argued that validation of dissolution tests/testers can be, or have been, achieved using IVIVC based on convolution/deconvolution (CON/DECON) methods, and more recently applying PBPK (Physiologically Based Pharmacokinetic) modelling/simulation techniques. Unfortunately, this is scientifically incorrect and an invalid view and practice.
In reality, CON/DECON and PBPK techniques are mathematical/statistical techniques which use in vitro and in vivo results obtained experimentally. While applying the mathematical techniques, mathematicians/statisticians make a fundamental assumption that the data/results provided to them are relevant and valid, obviously obtained using qualified and/or validated tests and testers. It is the analytical laboratory which ensures that the results were obtained using qualified and validated tests and testers. The important thing to note here is that CON/DECON and PBPK are applications, mostly software, used to transform in vitro results into in vivo. They are not to show or prove that testers/tests used were validated or qualified, but work on the assumption that testers/tests were appropriately validated. Stating that CON/DECON or PBPK techniques establish that dissolution results were obtained using validated/qualified tests/testers is scientifically invalid view and unfortunately reflects misunderstood concept of the mathematical/statistical techniques applied.
It is of utmost importance to note that before using dissolution results for any purpose – eyeballing, similarity factor, tolerances setting, QC/QA, CON/DECON, PBPK or any other sophisticated statistical analyses – one has to establish that results obtained were from qualified or validated tests/testers. This qualification/validation has to be done PRIOR and INDEPENDENTLY, and not using products under development or assessment. If dissolution testers provide irrelevant and erratic results, which they do especially paddle and basket, then any modelling or statistical technique will not help in getting relevant results or correcting the problem. It will be complete waste of time and resources. One must first focus on developing and using qualified and validated dissolution testers.
In short, as the currently recommended testers have never been qualified and validated independently, thus use of techniques mentioned such as CON/DECON, PBPK, simulation/modelling, and other statistical analyses, becomes void and are of no practical use. Therefore, one must first focus on establishing or developing qualified and validated dissolution testers, if valid and successful use of IVIVC or PBPK is desired
