Categorized | Expert Insights, Resources

Implementing Quality by Design: Are You Ready, or Not?

    Implementing Quality by Design: Are You Ready, or Not?

    By Bill Schmidt, Principal, Tunnell Consulting

    Quality by Design (QbD) is unquestionably ready to help your organization solve the perennial challenges of drug development and manufacturing. The real question is whether your organization is ready for QbD. Answering that question begins with an understanding of QbD basics, benefits, and barriers to implementation, followed by a readiness assessment that can ensure that you have the right mindset, priorities, and resources aligned for a successful implementation.

    To help you assess your own organization’s readiness for QbD, this article will outline what makes any manufacturer “ready” for QbD. It will then examine a case study, from the fundamental implementation decisions the manufacturer made to the important components that it failed to address properly.

    Basics, Benefits and Barriers

    As long ago as 2004, Janet Woodcock, director of FDA’s Center for Drug Evalution and Research, stated that the basic principles of QbD are almost diametrically opposed to established industry practices. By increasing scientific understanding of products and processes, QbD makes risk-based compliance possible.

    Its goal is not to eradicate variability in processes but to develop a process that can accommodate the range of acceptable variability for maintaining product quality.  Starting from a Target Product Profile (TPP) based on Critical-to-Quality Attributes (CQA) (Figure 1),  you can then use appropriate analytical methods and tools to understand Design Space, defined by ICH Q8 as “the multidimensional combination and interaction of input variables (e.g., material attributes) and process parameters that have been demonstrated to provide assurance of quality.”

    Figure 1. Meeting Product Critical-to-Quality Objectives

    Such methods and tools include high-end statistics like multivariate analysis, modeling tools, and design of experiments (DoE).  They are used to help understand the most critical process parameters (CPPs) and map out the Design Space so that you can create an in-control operating space, preferably near the center of the Design Space where the most process robustness will be found (Figure 2). And with a demonstrable, scientific understanding of Design Space, you can then continuously improve the process without additional regulatory review.

    Figure 2. The Power of Statistical Analysis/Modeling

    Robust, scientifically understood processes that allow for variation without compromising quality generate operating improvements that translate into business benefits, including:

    • Faster time to market and reliable supply, when each day on the market could mean as much as $3 million for a $1 billion blockbuster drug
    • Fewer lost batches, with typical savings of $250,000 to $500,000 per batch
    • Fewer manufacturing deviations, saving hundreds of costly investigative hours and $10-$15,000 per deviation
    • Reduced out-of-specification results, reducing rework
    • Reduced compliance exposure and increased regulatory flexibility, with fewer remediations and the ability to make process changes without re-filing
    • A right-sized quality operation, freeing valuable personnel to concentrate on more value-added activities
    • Many-fold return on investment through cost savings and increased revenue

    These operating successes also help reinforce a “right-first-time” culture, where quality means continuously creating more value, not simply correcting problems.

    Despite these operational and business benefits, many organizations face some significant barriers to successful adoption of QbD. One of those barriers is simply insufficient understanding of the purposes of QbD itself and the benefits that it generates. Some may believe, for example, that QbD delays time-to-market rather than reducing it. There may also be significant organization resistance. Many people simply find change uncomfortable. QbD, which drives objective decisions on the basis of data, may be perceived as a threat in a culture where personal opinions or agendas dominate. Because QbD introduces new ways of thinking and new ways of working, it requires learning, which can also create discomfort.

    Some organizations may simply be in denial: “Our process is in control.” Others may have competing priorities: “We have other fires to fight right now.” Still others may lack the internal resources and expertise to implement QbD.

    These reservations are not entirely misplaced. Implementing QbD does require a significant cultural shift, as well as organizational re-training and learning. Like most transformations, QbD requires time, effort, and commitment, particularly from senior management. That said, benefits emerge relatively early in implementation—in weeks to a few months—and QbD’s ability to lower costs, reduce waste, and eliminate non-value-added activities associated with non-compliance rapidly outweighs the cost of adoption. The key is to begin with a QbD readiness assessment that can tell you what practices and resources you need to put in place to ensure rapid, sustainable, and successful implementation.

    The Elements of Assessment

    Readiness assessment for QbD falls into three major areas: strategy, organization and culture, and operations. Within each of these areas there are numerous elements that must be considered, but to avoid drowning in the details it is helpful to think of this overarching framework.

    The central element of strategic readiness is the Value Proposition for undertaking QbD.  This proposition should not be confused with a mission statement—for example, “we intend to be number one or number two in each of the therapeutic areas our products address.” It should be based on an analysis of the specific benefits the organization wants to get from QbD and how those benefits will be realized.

    Depending on the specific circumstances of the company and the operation, the value proposition for QbD could be based on any number of objectives and metrics. For example, it could be built around reduction of regulatory workload, reduced quality costs through reduced or eliminated deviation investigations, or opportunities to apply Process Analytical Technology (PAT) to enable real-time testing and release.

    The assessment of organization and culture begins with the question: Do we have the right human resources—the expertise—to implement QbD? This is not simply a question of what combination of current personnel, external assistance (if any), and new hires (if any) might be needed to implement QbD.  It is also about understanding how those resources will be deployed. In assessing those resources and their possible deployment, you should also realistically gauge the degree and kinds of organizational resistance you can expect, so that you can make sure that you have not only technical expertise in QbD but also the change management skills that transformation requires.

    Operational assessment encompasses all of the technical and tactical elements that must be in place and integrated smoothly into a process in order to deliver the promised value.  These elements could include everything from defining the target product profile (TPP) to virtually any process, technology or system in the operation. The list of such elements is long, but by keeping clear the distinctions between strategy, organization, and operations you can be sure that at any point in the assessment you are clear about what, precisely, you are assessing. This clarity is especially helpful when it comes to prioritizing each of the elements by importance, regardless of where each fits in the structure.

    Assessment in Action: A Readiness Case Study

    The experience of a manufacturer of generic and ethical drugs illustrates some, among the many, elements that a QbD readiness assessment encompasses, their prioritization, and some of their potential pitfalls. The company, which I’ll call Manufacturer X, wanted to assess where significant gaps lay in its ability to adopt QbD, and thus undertook an entirely in-house assessment.

    The following review of Manufacturer X’s assessment reviews what the company saw as gaps that could be filled by QbD, and offers some observations on the course of action chosen:

    Value Proposition:
    This should be the number one priority for any manufacturer implementing QbD—understanding precisely what value QbD is expected to deliver. In this case, however, X defined its objective as “the ability to deliver quick success to show the benefits of QbD.” Such a short-term value proposition is both tricky to calculate and difficult to justify, given that QbD is a long-term transformation of the way work gets done. The expected value should be significant, sustainable, and realistic. It should align regulatory benefits with business drivers such as reduction of regulatory workload, decreased cycle times, speed to market, and reduced costs.

    Further, those highly specific benefits should be quantified as rigorously as possible and built into the value proposition. And the proposition in which the benefits are contained must be understood and supported by top management, including the CFO, or it will not get off the ground.

    Timelines and resources to support QbD: This was a top priority and, in Manufacturer X’s view, a significant gap in its readiness.  The company assumed that it needed new people, but what it needed more was to initiate the culture and organizational change required for QbD to be accepted and to work. Both cultural resistance and organizational structure can be addressed through cross-functional teams that span the complete product lifecycle and that go beyond timelines and resources to a coordinated, cross-functional strategy with clearly defined roles and responsibilities (Figure 3). These teams can capture and share lessons learned, develop a common organizational purpose, develop a vehicle for sharing information within and across projects and facilitate robust tech transfer. X did not consider this team approach.

    Figure 3. Building Effective Cross-Functional Project Teams

    In addition, X viewed all sites equally and in need of QbD. It did not consider the fact that one site with a strong workforce and with the skill sets and readiness for change could pilot QbD projects and set a standard for other sites to follow in rolling out QbD.

    Greater understanding and application of common risk management tools to pharmaceutical products: This is also critical because, as FDA and ICH documents have made clear, QbD is intended to enable a risk-based approach to quality which can be extended to the entire product life cycle. Failure Mode and Effects Analysis (FMEA) is a common risk analysis tool used for QbD, one which X took advantage of. However, other risk assessment tools can be added to the QbD and total quality management toolbox to transform the organization’s mindset and culture toward one of continuous improvement.

    Some of these risk assessment tools for use across the life cycle, in addition to FMEA, include Process Mapping, Hazard Analysis and Critical Control Points (HACCP), Hazard Operability Analysis (HAZOP), Fault Tree Analysis (FTA) and Cause and Effect Analysis (CEA).

    Quality Target Product Profile:
    The QTPP is indispensable for defining the critical to quality attributes (CQAs)—the desired outputs of the manufacturing process. It is equally important for mapping the Design Space—that is, understanding the relative impact of input variables (process steps, process parameters, and raw materials) on CQAs. As the organization’s proficiency in risk assessment increases, the QTPP becomes more apparent.

    Statistical software resources:
    Also indispensable. Whether the company builds on its current software or acquires new software, the choice should be based on the ability of the package to perform multi-variate techniques. Manufacturer X did not address this question sufficiently.

    Statistics expertise:
    X concluded that it needed a statistician to more effectively and efficiently design and evaluate experiments intended to define the Design Space. A statistician can certainly jump-start learning, with a contracted statistician sufficing for the short term. But it is equally important to develop internal skills through DoE training and regular use of the statistical software, and beyond hiring outside consultants, Manufacturer X did not have a plan in place to develop its internal statistical expertise.

    Tech transfer process: This extremely important element should be framed around what current tech transfer processes are relative to process understanding, how complex batch records are, and other considerations. In the context of QbD, this means gap and risk analysis of process robustness, which extends beyond API and product processes (the extent of Manufacturer X’s tech transfer) to measurement systems, knowledge management and transfer, and business processes designed to achieve strategic and financial goals.

    Definition of a control strategy and control strategy document:
    Manufacturer X clearly defined and communicated its control strategy, but was inconsistent in executing it. Documents articulating a strategy are a necessary but not a sufficient condition of success. Execution is the bottom line, which depends upon organizational readiness to execute the control strategy. The control strategy and execution should be integrated with risk management activities designed to identify sources of deviation, learn from them, and close them.

    A statistical process control (SPC) system for trending:
    X was correct in thinking that understanding and applying the fundamental principles of SPC are important. However, it should have taken greater care in regarding it as a “system”. SPC can be as simple as paper charting for operators to utilize. More sophisticated systems are preferred, but only if the infrastructure exists to support them.

    Pharmaceutical Quality System: Basic elements of an existing quality system can sometimes support QbD, but the key is to evolve them as QbD is established, which X is in the process of doing.

    Knowledge of opportunities to implement PAT and the capital and resources to act on these opportunities: Manufacturer X has spent a lot of time wringing its hands over how, and in which processes, to implement PAT. It didn’t have to. Because QbD does not necessarily require a full-blown PAT revolution, this is not a critical component of QbD readiness. Existing equipment controls and measures may suffice. Linking QbD to PAT can also create perceptions of complexity and barriers that inhibit implementation.

    Need for a marketed product support function within formulation/analytical to support changes to existing products: This is a high priority only if the first QbD candidate is a marketed product. Otherwise, this can be addressed as part of the overall development of roles and responsibilities.

    Neglected Responsibilities

    In short, Manufacturer X had good intentions in wanting to implement QbD, but could have gone about its preparedness much more thoroughly and effectively. In addition, there are several areas which it failed to adequately address. Significantly, the company neglected to assess four important components of success, one of them organizational/cultural and the other operational:

    Strong Management Support: As with any transformational initiative, QbD success requires strong management commitment. Management should not only make its support of the initiative loud and clear but also actively sponsor the effort.

    Knowledge Management System: This key system enables appropriate content to be captured in technical and tech transfer reports as well as team minutes and it facilitates traceability from development through commercial operation. For example, the first step in achieving process understanding is to make sure that you have gathered in one place all of the historical data about the development of the product. The cross-functional team that addresses knowledge management should include IT so that they can understand the business needs of the system and deliver on them.

    Filing Expertise: It is also critical to have people on staff, or outside resources, who know how to manage a QbD filing, especially the new context of risk-based compliance.

    Governance Process: Closely related to the knowledge management system, a rigorous governance process enforces how data is captured and shared with the entire organization. The process should ensure that data isn’t isolated in functional silos, which prevents the valuable, reliable and timely interchanges of mutually helpful data among development, manufacturing, quality and other areas.

    Once Manufacturer X was made aware of these omissions in its assessment, it was in a position to bolster its initial analysis and develop a clearer picture of the true state of its readiness for QbD.

    As this company understood, implementation of QbD does result in additional work, new ways of working, and significant change. But as it also understood, QbD can be significantly less daunting if the organization understands clearly what is most important, where there are critical gaps, and what must be done to close them.

    Ultimately, companies that successfully adopt QbD are rewarded not only with the specific operational and business benefits of the value proposition, but also with the creation of a culture of continuous improvement that keeps paying additional dividends. As Hamlet said, “the readiness is all.”

    About the Author
    Bill Schmidt, Principal of Tunnell Consulting, has more than 25 years’ experience in industry.  Before joining Tunnell, he served as President and CEO of Advanced Gelatin Technologies and Executive Vice President at Midwest Stainless Process Systems. He has held scientific research positions at Dermik Laboratories, Sterling Winthrop and SmithKlineBeecham, where he developed and transferred process and product technology worldwide. Bill has a B.A. from Eastern University and an MBA in Pharmaceutical Business from the University of the Sciences, Philadelphia.

    Related posts:

    1. Transdermal Drug Development Using Quality-by-Design
      Barry Gujral and Peter Amanatides of Noven Pharmaceuticals outline possible approaches for the creation of the Design Space of Transdermal...
    2. QbD Classics: “Quality by Design for Biopharmaceuticals”
      The 2008 article by Anurag Rathore and Helen Winkle is an in-depth presentation of how QbD can be applied in...
    3. Reverse Quality by Design
      If you can design quality into a new product and make it better, why can't you use data to re-engineer...
    4. Pharmaceutical Quality by Design: Is It Time to Clear the Air?
      This PDF report (previously circulated only in bootleg form) speaks very eloquently, of progress, of good intentions, of great and...
    5. The Continually Improving Pharma Quality System
      Erica Brooks and Yasmeen Muhammad provide a broad, current overview of the issues, and explain why the "manager is no...

    One Response to “Implementing Quality by Design: Are You Ready, or Not?”


    Leave a Reply

    You must be logged in to post a comment.