|
|
|
|
|
|
|
|
Home > QbD Viewpoint > Categories
|
4/29/2009Central to the QbD approach is the notion of a Design Space comprised of the "multidimensional combination and interaction of input variables (e.g., material attributes) and process parameters that have been demonstrated to provide assurance of quality." However, limited prescriptive information is available regarding to how to construct such a Design Space and to demonstrate that operation within it "…provides assurance of quality."
Development and calibration of a Design Space typically involves construction of multiple predictive response surface models corresponding to critical drug product attributes. Such a Design Space is constraint by requirements of meeting multiple response criteria. Optimizations are typically approached using overlapping mean response or by a desirability function. However, these approaches fail to account for the uncertainty in model parameters and the correlation structure of the data, which can lead to dramatically misguided conclusions. As shown by Peterson (Journal of Biopharmaceutical Statistics, 18, 959-975, 2008), a Bayesian approach employing posterior predictive distributions addresses both of these limitations.
At the useR! 2009 conference, Blue Reference will present several case studies illustrating the use of Inference for R in conjunction with an assortment of R packages towards the construction of design spaces for representative pharmaceutical products using Bayesian Methods. An abstract of the presentation can be obtained at its linked title below:
Bayesian Approach to the Specification of Design Space in Quality by Design
Also to be presented at the useR! 2009 conference, is a paper illustrating the use of Inference to construct dynamic applications, as in Inference for QbD. An abstract of this presentation can be obtained at its linked title below:
Microsoft Office Dynamic Documents as R Applications
Despite a bountiful enumerated list of benefits, implementation of Quality by Design in pharmaceutical development is moving at glacial speed. Needed are cheaper, easier and faster methods for achieving QbD goals. The source of the problem and its potential solution, featuring Inference for QbD, are outlined in an article by Paul Thomas, Senior Editor of pharmaQbD, entitled:
Can Bayesians Get QbD Past Tipping Point?
The article is bases on a persistent refrain heard in pharmaceutical development comprised of the following:
- Clear business benefits of QbD need to be demonstrated before large-scale implementation. However there is a reluctance to initiate demonstration programs because current QbD approaches require a large investment of resources (people, time and materials) to execute the requisite scope of experimental work.
- There is a plethora of prior data from past development studies. However, using classical risk assessment, DOE and multivariate analysis, there is no way to integrate prior data with new studies. Huge benefits would be gained if one could make effective use of the abundance of prior data for QbD planning, execution, decision making and CMC filings.
10/6/2008
Companies shifting to science-based manufacturing are taking their first steps by transitioning from quality-by-inspection to quality-by-design (QbD). From the FDA's viewpoint, the principles of QbD in pharmaceutical development are fairly straightforward. It requires establishing a clear linkage between the safety and efficacy of the drug product in the patient with its quality as defined by the attributes of the drug product and then linking it all the way back to the process for preparing the drug product. Specifically, QbD requires achievement of understanding at two levels as illustrated below:
- Clinical Understanding, which establishes a link between the attributes of the drug product and safety and efficacy in humans; and
- Process Understanding, which establishes a link between the attributes of the drug product and process parameters, process attributes and material attributes of the active pharmaceutical ingredient (API) and excipients that go into the drug product.
From a practical standpoint, process understanding, and the associated design space, entails:
- identifying and explaining all critical sources of variability;
- managing variability by the process via measurement and control of critical process variables; and
- reliably and accurately predicting and controlling product attributes within specifications (achieve quality).
Sounds simple enough, but arriving at process understanding and putting it to work remains a tall order. Towards that end, the FDA suggests that modeling and simulation must play an increasingly important role in science-based manufacturing-that is, from its current supportive role in empirical-based pharmaceutical development to a central role in QbD-based pharmaceutical development. For modeling and simulation to fulfill this promise will require development and deployment of three critical elements:
- A QbD Modeling and Simulation Infrastructure. An easy-to-use system for assembling, transforming, exploring, analyzing and reporting on QbD data.
- A QbD Work Process. Guidelines and work processes for deciding where and how modeling and simulation should be carried out in QbD. Such work processes typically involves risk assessment, experimental planning, prioritization, and data analysis and documentation.
- Organization and Culture. Establishment of integrated, multidisciplinary, multifunctional development teams trained in the use of QbD modeling and simulation for decision making.
Transitioning to QbD is reflective of a larger business trend aimed at fact-based decision making. In Competing On Analytics, Thomas Davenport of Babson College and Jeanne Harris of the Accenture Institute of High Performance Business outline a roadmap to company competitiveness by wielding analytics. Access to data is not enough. "Even if an organization has some quality data available, it must also have executives who are predisposed to fact-based decision-making." And, management support is critical. "A data-allergic management team that prides itself on making gut-based decisions is unlikely to be supporting. Any analytical initiatives in such an organization will be tactical and limited in impact." 1/28/2008
The FDA has recognized that product development is now the weak link in the “critical path” from scientific discovery to commercial product. In response, the FDA has instituted sweeping changes that are beginning to have a tangible impact on the way pharmaceutical developers and manufacturers conduct their business. FDA’s “Pharmaceutical Quality for the 21st Century—A Risk-Based Approach” squarely takes aim at the current state and replaces it by the desired QbD state focused on product and process understanding.
|
Aspect |
Current State |
“Pain” to industry |
Desired QbD State |
|
Pharmaceutical development |
empirical; typically univariate experiments |
unscientific; relies on intuition; high risk of failure |
systematic; multivariate experiments |
|
Manufacturing process |
locked down; validation on 3 batches; focus on reproducibility |
ignores variation; high risk in moving to manufacturing |
adjustable within design space; continuous verification within design space; focus on control strategy |
|
Process control |
in-process testing for go/no-go; offline analysis |
slow response; lost batches; drug development delay |
PAT utilized for feedback and feed forward in real time |
|
Product specification |
primary means of quality control; based on batch data |
difficult to achieve right-first-time production |
part of overall quality control strategy; based on product performance |
|
Control strategy |
mainly by intermediate and end product testing |
delays batch release; low equipment utilization rate |
risk-based; controls shifted upstream; real-time release |
|
Lifecycle management |
reactive to problems & OOS; post-approval changes needed |
inefficient, costly processes; discourages changes |
continual improvement enabled within design space |
Quality by Design (QbD) provides “a framework for allowing regulatory processes to more readily-adopt state-of-the-art technological advances in drug development, production and quality assurance” and shifts focus from “quality by testing” to “quality by design”—that is, build quality into the process rather than rely on resource-intensive quality control systems to prevent defective products from leaving the factory. QbD practices are implemented at four levels:
· QbD Level 1: Process Understanding. Develop end-to-end process understanding based on multivariate analysis of designed experiments and/or historical data. Includes: (a) identification and characterization of critical-to-quality process parameters (CPP) and (b) identification of root causes of variability.
· QbD Level 2: Quality by Design. Design a process defined by a design space that is robust and where the variability is controlled.
· QbD Level 3: Monitor, Predict and Control. Monitor CPPs via off-line, in-line or on-line analyzers. Use real-time monitor feedback in conjunction with prediction to achieve process performance/quality supervision via real-time intervention of CPPs.
· QbD Level 4: Continuous Improvement. Use accumulating manufacturing data as the basis to modify and improve the process within the design space.
At this stage, the underlying concepts and rationale for implementing quality-by-design practices are well understood and accepted. Despite this progress, there remain critical impediments to QbD implementation including the following:
· From a practical standpoint, what comprises and how does one acquire process understanding?
· From a practical standpoint, how does one decide that a process parameter is critical?
· And most importantly, how do we know that the identified design space of the process links to the clinical design space of the patient? After all, the aim is to design a process that meets the needs of safety and efficacy for the patient.
These key questions will be the subject of subsequent posting. |
|
|
|
|
|
|
|
| |
|
© 2008 Blue Reference, Inc. All rights reserved. | 3052 NW Merchant Way, Suite 100, Bend, Oregon 97701 | 541-316-2343 |
Terms of Use
 |
|
 |
|
|
|
|