Is quality by design (QbD) in pharma only for gifted scientists?
These days in pharma development circles, hardly a day goes by without mention of quality by design (QbD). Who is doing it properly? Is it too expensive? What is the business case? Can small companies afford it? How do we get executive management interested?
You could be forgiven for believing QbD is some kind of complex hybrid mix of rocket science and nuclear fusion, territory reserved for the scientifically gifted only. This is not the case of course, and in this article I hope to convince you so, especially if you are one of the scientific cognoscenti wrestling with the issues and opportunities of QbD.
What did the regulators Intend?
Many will be familiar with the famous Modernization words of Dr Janet Woodcock, Director CDER at FDA [1], now seemly lost in the mists of time, pleading for:
“A maximally efficient, agile, flexible pharmaceutical manufacturing sector without
extensive regulatory oversight.”
Nearly 20 years on, who would argue we are even close to that goal? Certainly no-one I talk to. The reason to me is clear—the industry has totally missed the intention behind the words—to make your products in the same way as those companies demonstrating outstanding outputs from their production systems. Yes, production systems; and this is where we are stuck.
Science doesn’t understand the notion of making products for customer markets. It is, of course, a vitally important part of the mix, but it is not the cake itself . . . the production system makes the cake and delivers it to customers. So if you are wrapped up in design space, process validation, process analytical technology, design of experiments, etc, spare a thought for the rest of it. The regulators intended it that way.
Importance of the production system (PS)
A word about a Production System (PS). The Toyota Production System (and others similar) was spawned from the work of Shewhart, Deming and Juran et al in the late 50s and 60s, helping those companies introduce increasingly effective ways to produce products for customer markets that were “fit for purpose.” When the NUMMI study in the 1980s investigated what they had achieved, they discovered a completely different way of working compared with the hitherto mass production paradigm.
These findings subsequently became reported in the western world as Lean Production and were recorded in the now famous book “The Machine That Changed the World” [2]. There were several absolutely fundamental differences between this way of working and what the lean fraternity dubbed the process village—producing to a “batch and queue” mentality. Firstly, they were fanatical about understanding what their customers “value for money” needs were—what they were willing to pay for, and importantly, what they were not.
Then they aligned the PS to deliver that value and relentlessly eradicated the non-value adding activities; and this is where the critical lesson for Pharma and QbD lay—they designed an end-to-end value stream for a defined product family in early development; this was operated in close synchronization across the value stream and processes/equipment were joined up. By joining their processes, it took away the cosy blanket of being able to run a machine to bulk, oblivious to the need of the downstream process.
They had to solve the problems where, for example, the components fed from an upstream activity (and that could be a third party supplier) contained defects that could not be accommodated by the downstream operation. That meant also, involving and developing their suppliers, and even their suppliers, so that the PS improved continuously over time, resulting in the superior quality, cost and delivery leadtime performance we have come to expect from the best companies in the field.
In summary, they created a tight system of people, processes and technology working in unison and joined up from beginning to end. Interestingly, much a Deming’s work was on the subject of systems [3] and yet we see little of it in the Pharma world.
What Does This Mean for QbD?
It means that it is not enough to attempt to optimize the existing paradigm of mass production that exists in this sector; it needs a fundamental shift in mindset. For example, if a product has been developed using materials that have some basic incompatibility with the downstream process being PAT’d (PAT = process analytical technology), the answer does not necessarily reside in better
technology on the line; it may lay in involving the supplier at the development stage to identify a better characterized material. This is a supplier relationship issue and is an example of how “the system” is such an important element. If we now take the example above, the critical need for a systems approach to QbD begins to emerge. Because of the interconnected, interdependent nature of Production Systems, cause and effect are rarely in the same place [4]. The effect of high defect rates, for example, can often be traced back to issues locked-in very early on in development.
When we trace back to those early days in development, we often see thinking processes devoid of downstream considerations.
What Needs to Change?
The systems world tells us that effective change for the better (Kaizen in Lean parlance) can only be achieved by attacking cause rather than effect. So, what is the cause? In my opinion it is the fact that, historically, Pharma has had a deeprooted mind-set that aims to answer the question: “How can I invent a miraculous new treatment for an unmet medical need?” To redress the effect of
this (total ignorance of the process of bringing products to customer markets) the industry needs to answer the question: “How can I make a difference to this person’s life so that they will come to me rather than my competitors?” When asking that question, you instantly focus on the source of your long-term success, satisfying patient needs; and then you do what every other successful
sector does to stay in business—align the entire organization with delivering the customer value position. How many drug development organizations are thus focused? Not many, if any, I suspect.
This is where the change has to take place, in early stage drug development, where all the manufacturing data begins to become hard coded—processes, suppliers, specifications, analytical methods, etc., and registered in the CMC package (CTD Module 3). Those downstream of this can only ever be grappling with the “foliage” of the problem. In the famous words of Henry David Thoreau:
“For every thousand people hacking at the leaves of evil, there is one striking at the roots.”
If you haven’t already done so, why not join me, make your way down to the long-term source of QbD success?
References
1. Pharmaceutical cGMP’s for the 21st Century – A Risk-Based Approach. FDA
website
2. James P Womack, Daniel T Jones and Daniel Roos, “The Machine That Changed the World, Simon & Shuster, New York, 2007.
3. W Edwards Deming, The New Economics of Industry, Government, Education, 2nd Ed
4. Hedley Rees “Supply Chain Management in the Drug Industry: Delivering Patient Value for Pharmaceuticals and Biologics, J. Wiley, NJ, 2011 PP 357 – 362.
