From the Editor: Defining the Design Space

Feb. 15, 2011
How can you talk your CEO into paying for Quality by Design if you can’t even agree on what its core concepts mean?

A growing number of people are intrigued by the potential of pharmaceutical Quality by Design, especially as its potential impact on product and process development is better understood. Among those paying attention are corporate managers, without whose support QbD is merely just another techie toy. Only change management, supported from the top corporate levels on down, will ensure that pharma QbD efforts succeed.

In 2009, McKinsey & Co. analysts published an article designed to show senior managers how much QbD could save on overall operations. Pharmaceutical product and process development cost the industry up to 20% of its potential net profits each year, they wrote, concluding that QbD could increase profits by $20-to-$30-billion a year.

There is a need for training higher level management on what the goals are and what can be achieved, McKinsey analysts write, so that they ask the right questions. For example, not, “is the project on time,” but “what are the risks associated with this project, and what would be the impact of shifting to an alternative design?” In addition, the analysts note the need for close collaboration between manufacturing, quality, regulatory affairs and R&D.

Well and good. But then why can’t there be agreement on such critical concepts as Design Space? Last month, I attended IFPAC, the International Forum for Process Analytics and Control (IFPAC), in Baltimore. This program for decades was a place where pharma’s process control experts could hear about the state of the art (practiced elsewhere) and bemoan pharma’s primitive level of automation and control.

Today the picture has changed somewhat, as more drug makers harness more powerful analytics and process control. But one presentation brought emotional debate about the Design Space, between those who advocate more aggressive testing of potential process changes and their impacts, without the need to refile a new Design Space, and others advocating a conservative approach. At issue was just how dynamic the Design Space can, or should, be. Should any changes due to unanticipated “real world” situations necessarily trigger the development of a new Design Space, and post-approval change filings and more testing? Or can there be some leeway in testing potential process changes, up front.

Duquesne University professor Carl Anderson had discussed tests that he and his colleagues Benoit Igne, Zhenqi Shi, Sameer Talwar and James Drennan had done with the over-the-counter compound Excedrin, to incorporate raw material changes into a Design Space and test process robustness.

Researchers found that when both API particle size and excipient loading were changed, the resulting material was within specifications. However, if only one of these values was changed at any time, the material would be out of spec.

Changing compression force and speed, it was found, could compensate for the raw material changes and allow the process to make product of sufficient quality. Thus, Anderson concluded, changing critical process parameters, based on raw material characterization, will allow specs to be made.

That’s when the debate began. For more on this discussion, which extended to two different LinkedIn groups, see “Staying Within the Lines?” on Clearly, one must go beyond basic ICH definitions of the Design Space to evaluate what “dynamic” has to mean in a given situation. But there must be some agreement on the basics.

In some small way, we’d like to contribute to clarity by covering controversial areas where QbD may not be entirely clear. We’ll be consulting with experts to develop a glossary for many of the key concepts of QbD on Your suggestions would be essential. Pharmaceutical Quality by Design is too important an initiative to lose ground because of lack of consensus.

About the Author

Agnes Shanley | Editor in Chief