Classifying Manufacture

Why do I think we need a manufacturing classification system (MCS) for pharmaceutical development of oral solid dosage forms? Simple: the timing is right. There is an increasing focus on simplifying development, identifying risk and using knowledge to design better processes that will increase robustness, improve speed of delivery to patient and reduce costly failures.

The pharmaceutical industry already has a relevant example: the Biopharmaceutics Classification System (BCS), a scientific framework for classifying the in vivo absorption risk of drugs based on their solubility and permeability. The BCS allows rapid assessment of project risk and directs effort towards the appropriate areas. Given the success of the BCS in the biopharmaceutical area, a team within the Academy of Pharmaceutical Sciences (APS) decided to explore whether an MCS to aid drug product manufacture would be useful. The driver was the awareness that many aspects of the current situation are not ideal. There is no definition of what the “right particles” are and what the “best process” is, which leads to a lack of clarity in the goal for particle engineering efforts. By providing a common understanding of risk, an MCS could predict from prior experience and reduce the potential for issues across the key interface between R&D and manufacturing.

At a subsequent seminar to consider this initiative, there was a large degree of enthusiasm and great input from the delegates present – and even a pleasing degree of agreement on the basic structure: an MCS based on processing routes divided into four classes:

1. direct compression
2. dry granulation
3. wet granulation
4. other technologies.

The structure assumes that there is a preference for simpler manufacturing routes. Moving down through the classes, process complexity increases, increasing cost of goods and the risk of unwanted changes to the API. In the case of wet granulation, for example, addition of water along with drying and milling steps could lead to undesirable form changes, API attrition, and degradation.

Having established the structure, the next step is to determine what API properties are important when selecting or modifying materials to enable an efficient and robust pharmaceutical manufacturing process. I believe that most formulators already have an informal MCS in their own heads. It’s what comes to mind when someone approaches you for advice on a new project. What questions do you ask to rapidly assess risk? BCS Class? Dose? Likely drug loading? I would certainly ask all those questions, and, in addition, I always like to examine the size and shape of a new API under a microscope. I have found that higher–risk APIs tend to be smaller and more needle-like. What’s in your mental MCS?

I will not give a detailed technical overview of our progress so far here, as it can be found in our recently published white paper (1), which also gives examples of an ideal direct compression material, properties necessary for dry and wet granulation, along with examples of when other technologies may be needed. Once you have read the white paper, we would be interested in hearing your opinions and suggestions. You can contact us by email ([email protected]), by participating in our conference roundtable session at FIP 2015 or by completing our online survey (http://tmm.txp.to/0215/MCS).

A working group has now been established with the aim of publishing a second, more detailed white paper that could involve gathering more data on input API, possibly in a centralized database. It will also consider the use of target material profiles, which would inform API optimization. Identification of surrogate materials that could act as model materials for each MCS class could also be a promising route forward along with the development of modeling tools for predicting formulation performance. If you think you can contribute, then please get involved.