Problems of failures of calibration/ Performance Verification Testing (PVT) are not new. These have been with us since their introduction in the USP. Presently, the issue is not of the failures of calibration/PVT, which is very well established, but how it should be addressed.
To address the issue, let us breakdown the testing, calibration/PVT or drug dissolution testing in general, in different components.
(1) Calibration/PVT Tablets
(2) Dissolution testing procedure and/or analyst training/experience
(3) Apparatuses (paddle/basket)
(1) Calibration/PVT Tablets: At present there is no independent mechanism available or used to establish quality or reproducibility of the tablets. The quality and reproducibility of calibration/PVT are established using the paddle/basket apparatuses themselves. Therefore, quality and reproducibility of the tablets, hence failures, are dependent on the characteristics of the apparatuses used. Using statistical analyses, studies from the USP clearly demonstrated that PVT tablets do not significantly contribute to the failures (1, 2).
(2) Dissolution testing procedure and/or analyst training/experience: In this regard, one should note that dissolution testing means dropping a tablet/capsule into the vessel containing a medium maintained at 37ºC and starting the stirrer (spindle) at a pre-set rpm, followed by withdrawing a sample from the vessel. It is extremely important to note that an associated or required analytical technique such as spectrophotometry or chromatography is not part of the dissolution testing or testers. An analyst can conduct a dissolution test independent to the analytical (quantitation) technique and may send the samples to another analyst/analytical laboratory physically located hundreds of miles away to determine drug levels in the samples from dissolution testing using techniques of their choice.
Therefore, the training and validation aspects of a dissolution testing or a tester are limited to the steps as noted above i.e. dropping a tablet/capsule in a container (round bottom beaker) containing water or buffer maintained at 37ºC and pressing the button to start the spindle at a pre-set rpm. Now, one should ask which of these steps require training that the analyst might not have already been taught in his/her undergraduate degree program as well as in some cases post-graduate programs. The point being, it is highly unlikely that an analyst’s training or experience, or lack of it, could possibly be a reason for the failures. In addition, based on statistical analyses, studies from the USP have also clearly shown that analysts do not contribute significantly to the failures (link).
(3) Apparatuses (paddle/basket): For all practical purposes, a dissolution tester is a mechanical stirrer and comes with most, if not all, pre-settings to provide a reproducible placement of a vessel, a stirring element and a few buttons for operating the testers.
The question is, are these testers capable of providing reproducible results with an acceptable %RSD (or %CV) e.g. 10% or less? Before, calling or promoting these as dissolution testers, suppliers should provide independent evidence that these testers are capable of providing dissolution results with an RSD of 10% or less. The problem is that no supplier provides evidence of such capability, which, by the way, should be the most fundamental and basic information available for dissolution apparatuses.
The reason such information (expected %RSD) is not available is that these apparatuses cannot provide reproducibility of 10% RSD or less. A number of studies are reported in literature showing that considering the flow dynamic within dissolution vessels, apparatuses inherently have to provide very high variability (RSD) in results e.g. 30%+, for example see links, 1, 2, and 3.
The tolerances are commonly set narrower which, in reality, causes, or have to cause, the failures.
The suppliers rather than acknowledging this problem (deficiency) of their apparatuses suggest that the PVT Tablets are the reasons of the problems/failures, which is not accurate. Furthermore, they also often suggest that lack of adequate control or care (de-aeration, vibration free environment, alignments, etc.) on an analysts’ part may cause such failures. However, in reality, the failures of calibration/PVT Tablets just reflect the expected high variability, and unpredictability aspects of the apparatuses.
Just like people expect RSD values from USP for PVT tablets, people should expect, and demand, %RSD values for the testers from the suppliers. This should be a normal practice or requirement for any instrument/apparatus to show its reproducibility in terms of %RSD. Why can’t suppliers of dissolution testers provide this (%RSD) value? Why do they always refer to the USP and/or the users/analysts? In fact, the USP itself should be requiring the suppliers to demonstrate capability of their stirrers, known as dissolution apparatuses/testers.
The suppliers often use the excuse that they are providing apparatuses/stirrers which meet the specifications as described in the pharmacopeias (e.g. USP). The excuse is not valid. The suppliers have to demonstrate that these apparatuses are capable of providing reproducible results with acceptable %RSD. Then, these apparatuses should be used for PVT to show whether PVT Tablets behave as expected or not.
Pharmacopeias (such as USP) should consider removing the requirements and specifications of the testers from Dissolution Chapter (such as USP <711>) as it is providing an excuse to the vendors for supplying flawed and non-validated stirrers and restricting improvements.
In short, the suppliers have to acknowledge that the apparatuses are causing the PVT failures and to demonstrate that dissolution apparatuses are indeed capable of providing dissolution results with an acceptable %RSD. So, the next time a PVT, or a product, fails to meet tolerances of a dissolution test, consideration should be given to using alternate testers which would be free from the flaws of paddle/basket apparatuses.
