The most commonly reported results from dissolution testing for establishing the performance of oral products (e.g. tablets/capsules), in particular for immediate-release (IR) products, is that products should meet a criterion of not less than 80% drug dissolution within less than 60 minutes, mostly 30 minutes.
Let us assume that a dissolution test (n=6) provides the following results (% drug dissolved) at 30 minutes (96, 88, 65, 110, 66, 65; Ave = 82; RSD = 23%). Obviously, this test/product, with at Q of 80%, would not meet the USP tolerance at the first stage. However, with a bit of luck and a second round of testing of 6 units it may meet/pass the USP Tolerance criteria. For details, please see the link, in particular the data set in row one.
Let us discuss the interpretation of these observed results.
(1) Recognition of high variability (RSD=23%) in the USP Tolerances is built-in. If one gets high variability then this should be accepted as expected and normal for testing. Often people, who desire a typical outcome of lower %RSD values of analytical testing (spectroscopic or chromatographic), become uncomfortable with such high variability in dissolution testing. Indeed, dissolution testing should provide lower variability just like any other typical analytical test, but that is not the case. A common reaction to such highly variable and unpredictable results is often directed towards the poor performance of the analytical laboratory and/or manufacturing of the product. However, it is a well-known fact, established with experimental and computer simulated studies, that indeed testing, using paddle or basket, provides such high, or higher, variability in dissolutions (link).
Therefore, no matter how good (reproducible) a manufactured product would be, it will show variability as high as of the testing. It is simply unfair and inaccurate to focus or blame the analytical laboratories and/or the products for the observation of high variability in results. If one would like to see better quality (reproducible) test results, one has to use an improved dissolution method or approach. Conclusion: Dissolution results using currently recommended dissolution apparatuses do not, or cannot, reflect the true quality of products.
(2) Concerning the relevancy of results: In principle these tests are conducted to establish potential drug dissolution characteristics of a product in vivo that is why these tests are conducted simulating GI tract environment. However, it is a very well established fact, as per a recent statement from USP as well (link), that current dissolution tests, in particular using paddle and basket apparatuses, are not linked to in vivo behaviours of a product.
Unfortunately, there is a made-up reason often presented in support of the current practices, with a strong marketing element, which is that dissolution testing provides “help” in monitoring consistency in manufacturing. It is to be noted that dissolution testing has no link to represent manufacturing in this respect. It is a test to evaluate a drug release characteristic of product under a simulated GI tract environment. Conclusion: The drug dissolution results as reported at present are not relevant to product behaviour in humans, thus cannot reflect quality of drug products for human use.
(3) In addition, a serious difficulty in evaluating the relevancy and reproducibility of the results and products is that the apparatuses used or recommended by pharmacopeias are not validated for dissolution testing purpose. Conclusion: It is not possible to draw any accurate conclusions regarding quality and/or relevancy of dissolution results, or the products, using apparatuses which are not qualified or validated for the purpose.
The reason for lack of reproducibility and relevancy of dissolution is because of poor mixing/stirring (hydrodynamics) environment within dissolution vessels. These deficiencies have clearly been established based on experimental and computer simulation studies. It would be important to note that the flaws of paddle and basket apparatuses are such that they cannot be corrected without changing their current setup/design. One simply requires a new or modified stirrer or stirring mechanism to address the issue of dissolution testing.
A new spindle known as the crescent shape spindle has been proposed for this purpose, which not only addresses the deficiencies of paddle and basket apparatuses, but also provides a very simple approach for generating relevant and reproducible data (link).
