Modeling using Dissolution Data
- Posted by: Ken Phelps
- Published on: August 22, 2008
Not only pharmacokineticists get to have fun in the modeling sandbox. Chemists and formulators get to have their fun synthesizing data. Let’s use an example of how dissolution data can be used for modeling. The example is taken from a project to develop a oral modified release drug where the RLD is an immediate release drug; IR -> ER. There are some twists to the desired ER profile, but bear with me on the generalization, please.
Before doing expensive and time consuming pk studies, we use dissolution testing as a surrogate for how the drug will dissolve in the GI tract (an assumption is being made here – that the absorption and resulting plasma levels will correlate well with dissolution data, is something we should know from the RLD information). Hopefully we have some idea of the pH profile, solubility, etc. to help select dissolution media, but that is out of the scope of this example. We’ll assume we have a good dissolution model.
We develop a couple of candidate formulations, probably at the ends and middle of the spectrum of desired behavior:
The chart above is the usual way to present dissolution data. Another view is instructive to determine how much is released at selected time intervals:
Using this information, we can figure out how much should be released at each time interval to obtain the desired behavior (below). Using this information, we can then formulate two candidates for a comparative pharmacokinetic study. Note that a parallel modeling is conducted with the pharmacokinetic and dissolution information from the RLD.