American Pharmaceutical Review: A Russell Publishing Publication

Mallincrodt BakerFoss NIRSystems

Functionality and Performance of Excipients in a Quality-by-Design World Part 2: Excipient Variability, QbD and Robust Formulations

Excipients

Chris Moreton FinnBrit Consulting

As mentioned in the inaugural column, excipients will play a major role in QbD development programs. Excipients are a major part of any formulation development; the other two parts being the active pharmaceutical ingredient (API) and the processing. All three together make the medicinal product, and all are equally important in a functional sense, because if we remove any one of the three, we do not have a product.

One of the issues that we have to come to terms with is variability. There is variability in almost everything. We are bound by the statistical distribution, be it normal, log-normal, or something else. There is really no way to escape variability, so we had better comes to terms with it. Excipients, APIs and medicines are no exception to this general rule, but variability is where the confusion starts. The discussion in this column will focus on excipients, but many of the same concerns apply to APIs; after all, from a materials science perspective, they are simply components of the formulation. Traditionally, we have controlled the chemical composition of our APIs, but sometimes their physical characteristics have not been as well controlled as the chemical process development people would have us believe.

Pharmaceutical excipients are typically manufactured at large scale; sometimes using batch manufacturing, but very often using some form of continuous processing operating 24/7 and 50 weeks per year. These large plants are designed to produce material at the center of the in-process specification, regardless of how they are processed. Because there is inevitably some variation, the manufactured excipient will also show some variation. The question for the formulation scientist and pharmaceutical engineer is, "How much variability will the product and processing cope with, and still produce a medicinal product that conforms to specification?" In other words, how robust is their formulation and process? This is the essence of the Design Space concept that is the vital part of QbD. One of the key questions in investigating Design Space is how to incorporate excipient variability into the Design of Experiments (DoE) that will be used to establish the Design Space and possibly the Edge of Failure.

In order to begin to cope with variability, we had better understand how it arises. If we think of the medicinal product, there are three main components, or groups of components (see above) namely: API, excipients, and process (equipment and unit operations). They can all show variability, but that is not the whole story. Our model is lacking; it is not describing all the components; the missing pieces are how the three main components interact during manufacture, and the impact of the operator(s). Without recognizing these additional sources of variability, we are not going to be able to solve our problem. If we take the variance as an indication of variability, and there is an advantage in using the variance since variances are additive, then we can describe the total variance of the product as follows:

1

Interactions can arise in many ways, for example; powder-powder, powder-liquid, powder-liquid-process, operator-process, etc. In any given application not all interactions will be important, but some most likely will be, and we have to find which ones are important for our particular project.

In the past, some have tried to go for tighter and tighter excipient specifications. That is not the way! The manufacturer cannot always control the variability in ways that the excipient users demand. One approach that has been tried is to select particular lots of excipient
based on a particular parameter and narrow specification. My advice is DON'T! It is a disaster waiting to happen. For lot selection to work, you will probably need better than 50% of the excipient lots to comply. I say this because although the variability is probably cyclical, it is also random, and continuity of supply will be an issue for both excipient manufacturers and users. Try explaining to senior executives that the latest blockbuster product is on restricted supply because the formulation relies on a particularly tight excipient specification that cannot easily be met!

...

To view the complete article, please login or subscribe if you don't already have an account.

Site Search:

Bruker AXS