Following is response to a question/concern I posted on LinkedIn Forum. I hope you as visitor to this blog may find it useful as well. Saeed
Question:
You are always asking for references – so here is a recent example linking in vitro to in vivo that was accepted by the regulatory agencies. If the testers are not validated and non qualifed then this work would not work – but it does. Hence your arguments are nonsense as I can link in vitro to PK using the testers plus PBPK modelling. http://pubs.acs.org/doi/abs/10.1021/acs.molpharmaceut.6b00497
My response:
James: I am sorry to say that you are not following the question/concern or the science of dissolution testing. You are just trying to be bullying. Let me explain: (my comments are based on the content of the abstract, as I do not have access to complete paper. I am assuming the tests were conducted using paddle apparatus. If you desire a detailed analysis or response then please provide me a copy of the publication. Furthermore, I do see some specific flaws in the interpretation/conclusions; however, my comments are of general nature describing publications such as the one you cited.)
What is described in the publication is a typical example of matching in vitro and in vivo outcomes, which are not that uncommon in literature. Note I am using the word “matching” not correlation or validation. What people are doing and fooling others, including you, is calling “matching” and “association” of results as “correlation” or “prediction”, which is incorrect and unfortunate. There is no need to have such results from a dissolution tester, one can use anything from lab stirring bar to kitchen food processor to obtain some dissolution results (of course by adjusting experimental conditions) and match them with any of the outcomes, maybe drug levels in blood, air, space or Mars. I am almost certain that you would not follow/understand the mathematical “rituals” people use to come up with expected or desired results to impress others like you and peers. I am not impressed.
To address the question/concern, to obtain some relevancy or predictability of exercise, one has to start with a question, is the dissolution tester used to generate dissolution results is capable of providing “relevant” and “reproducible” results. Once, it is established, and then one has to ask the question, is the tester, after showing the capability relevancy and reproducibility, capable of providing relevant “in vivo” dissolution results. That will be called the “qualification” and “validation” of the dissolution testers. Now you should take this data and start relating to blood/plasma levels, using again scientifically and mathematically valid models (note that even the mathematical modelling will require its own independent validation before applying for such correlation or prediction). I will be happy to discuss this in further details if so desired. There is no qualification or validation is available here, in particular for dissolution testing, so the question/concern remains.
In short, anyone can obtain some numbers and mix them with mathematical equations to get any results one wants or likes. However, to be valid, they require prior validation, which is missing, in particular for dissolution testers and results. No question about that. You, and others, like to call this exercise modelling, simulation and validation and you are free to call anything you like. To me, it is just jugglery with numbers, thus has no scientific merit, credibility or use. Sorry!
