OR WAIT null SECS
Andrew G. Edwards is a senior consultant at EduQuest.
FDA is modernizing and streamlining current good manufacturing practices. The author examines FDA's evolving approach to quality systems and how a manufacturer can implement a quality system framework.
In 2002, the US Food and Drug Administration publicly recognized a need for change with its current good manufacturing practices (CGMP) for the 21st century initiative (1). This new approach to GMP compliance and enforcement recognized that FDA and the drug and biologics industries were not where they needed to be in terms of quality, manufacturing science, and risk management. This changing world of GMP applied to human and veterinary drugs and biological drug products. Key to FDA's approach has been to request a holistic cradle-to-grave management of quality issues.
Quality system: What is it?
The quality system should be an integrated framework within which the design, manufacture, packaging, labeling, and distribution take place. Commitment from management is crucial; without it there is no quality system. The underlying principles are:
In the changing, evolving world of GMP regulations, quality concepts are changing. The original GMP regulations started with a product-focus and quality control. Next was a shift to a process-focus and quality assurance. Now drug and biologics manufacturers face an enormous paradigm shift with the new model of quality systems (see Figure 1).
Figure 1: Models of quality.
However, this new evolutionary step is not new to FDA because the medical device industry has already adopted the quality-system approach in 1996 (2). Other industries also have adopted this approach (e.g., the automotive industry). Guidance is available from American National Standards Institute and International Organization for Standardization (ISO) on what is required in a quality system that parallels the medical device regulation (3, 4). FDA's approach has been to focus on quality by design and the control and reduction of variability by encouraging the use of sophisticated process analytical technologies (PAT) (5). Following the revised GMP guidance and the adoption of new technological advances, FDA hopes manufacturers will consistently maintain high-quality products and improve their manufacturing efficiency. FDA believes this should help lower costs and prevent shortages of critical medicines owing to failures that can result in product recalls. The estimated potential worldwide cost savings from efficiency improvements in the pharmaceutical industry is as high as $90 billion (6).
In October 2006, FDA issued a final guidance on quality systems intended for use in the manufacturing of human and veterinary drugs, including biological drug products (7). This guidance provides manufacturers with the ability to make technological improvements more readily, with appropriate regulatory oversight, and it offers guidance for defining management responsibilities, allocation of quality resources, dividing manufacturing duties between the quality unit and the production staff, and reviewing records and evaluating data.
The quality systems approach describes the responsibilities of the "quality unit," which combines the duties of quality control and quality assurance, "ensuring that the various operations associated with all systems are appropriately planned, approved, conducted, and monitored." The quality unit also is responsible for ensuring that controls are implemented and used; seeing that procedures and specifications are adhered to (at the manufacturer and at any contractors); approving or rejecting incoming materials, in-process materials, and drug products; and reviewing production records and investigating unexplained discrepancies.
FDA is applying a "six-system inspection model" in which the overarching quality system embodies five overlapping subsystems: production, facilities and equipment; laboratory controls, materials, and packaging-and-labeling (8). FDA will continue to monitor manufacturing plants through its inspection program and will continue to advance the training of its investigators in the latest technologies. FDA will focus the detailed inspection of a system so that the findings reflect the state of control in that system for every product (profile) class. A system is considered out of control based on GMP deficiencies that suggest lack of quality assurance. If one of the six systems is out of control, then the firm is considered out of control.
FDA is being much more explicit about the industries' scientific and regulatory responsibilities by making product design and process development part of the quality system and focusing on measurement and control (9). The agency has clearly placed the burden of evaluating and demonstrating adequacy and acceptability on the manufacturer. This shift is going to take a substantial effort and will affect research and development more than in the past. An ongoing dedication of more resources will be needed on the "front end" of process design and development. A huge increase is needed in the early and very detailed understanding of the process and product formulation. This will probably mean more time, effort, people, and money than the 1980s process validation movement and may take years to realize the benefits on the "back end." GMP requirements will have much more impact on pharmaceutical development.
Figure 2: Today's model.
With the quality system approach, FDA hopes manufacturers will apply an effective, knowledge-based scientific management of the entire product life cycle, from research to nonclinical (in vitro and GLP) studies to pharmaceutical development, clinical manufacture, clinical development (human trials), approval, commercial manufacture, and the postmarketing life of the product. This GMP paradigm shift will change the basic premise of the manufacturing process, completely shifting the point of regulatory and scientific emphasis by moving the regulatory anchor from the fixed process to fixed (constant) outputs (see Figures 2 and 3).
Figure 3: Tomorrow's promise.
For the industry to evolve to the new model, companies must be open to change. The change will be more complex than just reengineering manufacturing processes where FDA will play a substantial role in the change. The current product life cycle is very inflexible by the regulatory approach where the process is fixed within the operating ranges established in regulatory filings and approvals, and change is difficult and often requires preapproval. Because of this, companies are reluctant to submit supplemental applications and "reopen" product approvals. This reluctance is a result of substantial delay in implementing changes because of regulatory review. There is usually a limited ability to fully understand the potential clinical impact of changes. Currently, the process is not fully understood when a product is newly developed and standards and expectations change with time (e.g., analytical methods, impurity profiles, limits of detection). Companies have chosen to live with poorly understood, inefficient, low-yield processes instead of going through another round of regulatory review and approval and run the risk that additional clinical trials and/or preapproval inspections might be required.
FDA's vision is for industry to focus on reducing variability through process understanding (e.g., application of knowledge throughout the product life cycle). It is a "cradle-to-grave" systematic approach where product quality and performance are ensured through the following:
FDA encourages a risk-based regulatory approach where relevant regulatory policies and procedures are modified to accommodate the most current level of scientific knowledge. The change would focus on increasing the level of scientific understanding of how formulation and manufacturing process factors affect product quality and performance. More focus in the development stage is needed to design the capability of process-control strategies to prevent or mitigate the risk of producing a poor quality product.
This step is much more about a new philosophy and operational culture than it is about new standard operating procedures and a new quality manual. There may be a need to vastly improve and retool cross-functional and cross-departmental communication and coordination. A silo-based approach simply will not work. Every part of the organization that touches on the product life cycle will need to operate in a much more integrated way. Most of the advances that have occurred, and are anticipated to occur, are bringing the development, manufacturing, quality assurance, and information and knowledge-management functions so closely together that these four areas should be coordinated in an integrated manner. There may be a need to also include some of the not-so-obvious players such as regulatory, marketing, product surveillance, senior management, and so forth. Management responsibility and involvement is absolutely critical to the quality system. The current business model based on speed-to-market must strike a different balance and approach with more emphasis on doing it right.
Process design, process controls, and resource allocations will need to be based on risk assessment and risk management (10). Long-standing concepts of hazard control that use failure mode and effects analysis (FMEA) as well as hazard analysis and critical control point (HACCP) principles may not be familiar to the drug and biologics industries. FDA believes there will be significant regulatory benefits by improved ability to understand and implement changes with less FDA oversight and preapproval, faster and more predictable FDA review, analysis, and approval of new products and CMC changes, and fewer preapproval and shorter GMP inspections. The real business benefits (in cost, time, and efforts) are:
There is a practical relationship between quality and today's business realities (profit and cost-of-goods) where business objectives can be used to drive quality-system improvements. An understanding is also needed of the basic operational and scientific principles behind quality systems, quality by design, and PAT. There are core principles common to them all:
These principles boil down to one common theme: Know what you're doing, why, when, and how. The same core principles drove the development of FDA's GMP regulations as well as the standards by the ISO and International Conference on Harmonization and serve as the fundamental basis for the various operational excellence methods (Six Sigma, lean manufacturing, etc.).
Table I: Hazard control concepts.
The overarching context for the core principles are the application of hazard control—an umbrella term that broadly describes the various risk-based concepts that FDA has been discussing (see Figure 4). Hazard control includes:
The concepts of hazard control should be applied to the product and the process. These concepts are described in Table I.
Figure 4: Hazard control.
It is important to look at these core principles through the lens of PAT. PAT is really another way to describe advanced process control and is better described as "process analytical control technology." There is a need to deal first with the process and then the analytical pieces before the technology can have any meaningful use. Both FDA and the industries are looking at new technologies as the silver bullet to cure all of our quality ills, but it will be more important to build the gun first and know where to point it to be able to fire it and hit the target.
For PAT to be successful, the effort should be applied first to the development of a deep scientific understanding of the process and designing that process and its controls appropriately based on the target of constant output. This requires an understanding and application of the concept of critical control points and a full understanding of how the process works and how the process and its output are affected by variations in materials and control conditions. Next, one must understand how and where the process can be measured. One should only measure and analyze the things that really matter, based on process knowledge. There's no rational point in measuring things simply because we can, so this also must be tied into the concept of critical control points and how the process and its output are affected by variations. Finally, a successful PAT strategy requires a true understanding of how and where the process can be measured and effectively adjusted and controlled during production to constantly ensure consistent output.
Fancy technology doesn't mean a thing without the necessary process and analytical knowledge. FDA is pushing for quality system approaches that have been used for a long time in other industries (i.e., the electronics, defense, oil and gas, and telecommunications industries).
Inefficiency in the pharmaceutical industry could be wasting more than $50 billion per year in manufacturing costs alone—costs that could translate into lower prices or greater research and development—according to findings of the largest empirical study ever performed of pharmaceutical manufacturing and FDA monitoring policies (11).
Quality and compliance are not the same
Quality done right will result in compliance. Compliance, on the other hand, will not necessarily create or ensure the quality of processes or products. An organization must have both to succeed. The quality of a product is based upon the quality of the processes used in its manufacture. If one has poorly developed processes and procedures, continued compliance with the existing quality system will just ensure that one has consistently have poor product quality and integrity.
Use business drivers to change your quality system
Two key drivers of management's behavior and direction are regulatory compliance and operating costs–profit margins. A management team must focus on both. The consequences of noncompliance (business threat) as increased ongoing costs should be emphasized because focusing on the regulatory threat usually is not the most effective way to change management's views and actions. Use the company's own business realities as a driver. Understand how quality is related to business objectives (profits and cost-of-goods) and how they can be used to drive quality system improvements.
Measure and monitor the cost of poor quality
Philip Crosby, an expert in quality concepts, had these reflections on quality: "It is erroneous to think that quality is intangible, and therefore unmeasurable. It is precisely measurable by the oldest and most respected of means—cold, hard cash." It is also called "the price of nonconformance" but this is "an oversimplified definition . . . it is the costs associated with avoiding, finding, making, and repairing defects and errors"(12). The cost of nonconformance is part of the ongoing, day-to-day operating costs and is usually hidden in the operations (not measured). Examples are out-of-specification results, investigations, rejects, rework, expired components, unused labeling, production delays, waste, and so forth. The cost of nonconformance is a powerful tool for quantifying and reducing operating costs based on solid science and improved quality. The concept of "do more with less" can be translated into "make more product and profit with less waste and rework." It can be effective even if narrowly focused on just one part of the process. Very few companies are effectively looking at this. "[F]inancial measurements generally aren't used to validate quality's impact on profitability and costs." Most companies have absolutely no idea how their day-to-day operating costs are affected by poor quality.
Measuring and monitoring the cost of nonconformance is going to be increasingly important for our industries. Profit margins are facing growing global pressures from government price controls, larger purchasing power (both public and private), cross-border pricing, Rx-to-OTC switches, loss of patents, and so forth. It is already important to identify opportunities for improvement and hidden value and convince management to do more. Finding the cost of compliance can stop the debate because the returns on investment can be enormous. According to Crosby, "most organizations spend 25–40% of operating cost in nonvalued activities (the price of nonconformance) because requirements are not clear, communicated and consistently met" (13).
A company's cost of nonconformance can increase because:
How to establish an appropriate quality system
Existing products and processes must be the focus for implementing the quality system. However, it will be years before we approach the ideal cradle-to-grave quality system model. The industry must do what it can, as a practical matter, to effectively reengineer the moving train while the new tracks are being put in place. Dealing with existing products and processes boils down to dealing with and improving the current reality of our products and manufacturing processes.
To start, apply a hazard-control approach to select an existing product and process for improvement. Define the key practical steps toward improvement while proving value to build management support. Then apply quality-system thinking to problems found during the improvement steps.
A quality system should be appropriate for the degree of risk presented by the product, the complexity of the product in the manufacturing process, the intended use of the product, and the size and complexity of the manufacturer.
Major sections of FDA's quality system model include the following:
Management responsibility. The quality system should hold senior management responsible for:
Management should review the quality system frequently during implementation and after maturing, as a part of the general management meetings. A periodic review by a qualified outside source could be useful in determining suitability and effectiveness of the quality system. Management's review should include assessments of the process, product, and customer needs. A review should consider the following at minimum:
Review outcomes typically include improvements to the quality system and processes, improvements to manufacturing processes and products, and realignment of resources. Results should be recorded, and planned actions should be implemented using an effective corrective and preventive action and change control procedures and process.
Resources. Management must provide adequate resources for the following:
Management should be held responsible for:
Training is critical to ensure all employees remain proficient in their operational functions and understanding of CGMP regulations. Training should be two-fold: covering the employees' job function, as well as CGMP regulatory requirements. Training programs should include:
Technical experts who understand pharmaceutical science, risk factors, and manufacturing processes related to the product should be responsible for specific facility and equipment requirements. It is important to note that FDA feels that the CGMP regulations require a higher standard for calibration and maintenance than most nonpharmaceutical quality-system models.
Outsourced resources used in operations must be controlled and qualified. Contracts should clearly describe the materials or service, quality specifications, responsibilities, and communication mechanisms. The quality unit is responsible for approving or rejecting products or services provided under a contract.
It is important to qualify the contractor before signing a contract with that firm. Otherwise, the ability to help the contractor become more FDA compliant or choosing another contractor is impaired by a signed contract before audit.
Manufacturing operations. The manufactured product and process should be defined, from design to delivery. Control must be exercised over all changes. Documenting processes, associated controls, and changes to processes will ensure that sources of variability are identified. Documentation should include:
From a regulatory perspective, it is important to have experts in a pharmaceutical environment who understand pharmaceutical science, equipment, facilities, and process types and how variations in materials and processes can ultimately affect the finished product.
As part of design, controls for all processes within the packaging and labeling systems should have written procedures. As part of the design process, before commercial production, the controls for all processes within the packaging and labeling system should be planned and documented.
All inputs (i.e., materials, purchased or manufactured, that go into a final product) to the manufacturing operations must be examined. Materials are components, containers, or closures. Quality systems should ensure quality controls are established prior to the receipt, production, storage and use of all inputs. Suppliers of input materials must be audited on a periodic basis. An essential element of purchasing controls is the data trending for acceptance and rejection of materials for information on supplier performance. The auditing of suppliers should be based on a risk assessment. An audit should determine the reliability of a supplier's certificate of analysis (COA). A quality-system approach ensures the procedures are established to verify that materials are from qualified sources. Equally important is to have a system in place to respond to changes in materials from suppliers to avoid unintended consequences.
Design concepts established during product development typically matures into commercial design after process experimentation and progressive modification where areas of weakness are identified, corrected, and monitored. Critical quality attributes should receive increased scrutiny. Risk management should help identify areas of process weakness or higher risk and factors that may influence critical quality attributes.
A robust manufacturing process should be in place before commercial production. FDA recommends that scale-up studies be used to help demonstrate that a fundamentally sound design has been fully realized. Proper design and reliable technology transfer processes should ensure the ability to validate the manufacturing process. Validation provides initial proof the design produces intended product quality. Validation is not a one-time event.
The quality-system approach calls for manufacturers to develop procedures that monitor, measure, and analyze the operations, including analytical methods and statistical techniques. The quality-system approach indicators that change control is warranted when data analysis or information reveals an area needing improvement. Changes must be controlled and documented. In other words, the entire product life cycle should be addressed by the establishment of continual improvement mechanisms in the quality system.
Under a quality system, there will be a process in place to handle nonconforming material and product so that it is not placed into distribution. This process must be documented and must define responsibilities for halting and resuming operations, investigating discrepancies, recording nonconformity, and taking remedial action. Investigation, conclusion, and follow-up of nonconformities or deviations must be documented. Remedial action may include correction of the nonconformity, thereby allowing the product to proceed with proper authorization and justification of the conclusion regarding the problem's impact and use of the product for another application where the deficiency does not affect the products' quality or reject the product.
Evaluation activities. It is important to analyze manufacturing data on a regular time interval for trends as a means to control the operation. This analysis will detect potential problems early and enable the planning of corrective and preventive actions.
Internal audits should be conducted according to planned intervals for evaluation of implementation, maintenance of the quality system, and determination of whether processes and products meet established parameters and specifications. Although the CGMP regulations require product review on at least an annual basis, a quality-system approach calls for trending on a more frequent basis as determined by risk. Trending analysis can help focus internal audits.
An internal audit procedure should be established. It should call for a planned audit schedule that takes into account the relative risks of the various quality-system activities, the results of previous audits, and corrective actions. There is an obvious need to audit the complete system. Be sure to include a section on how auditors are trained in evidence gathering, their responsibilities, and auditing procedures. It is critical to maintain records of audit findings and assign responsibility for follow up to prevent problems from recurring.
Quality risk management is a reiterative evaluation process and a tool in development of product specifications and critical process parameters. It also helps manage and control change.
Corrective action is a reactive tool to ensure problems do not recur. Effective decision making in a quality system environment is based on an informed understanding of quality issues. The corrective action procedure should be developed and documented to ensure that the need for action is evaluated, the root cause investigated, actions determined, selected action is taken within a defined timeframe and the effectiveness of the action taken is evaluated.
A preventive action process is an essential tool in quality system management. The selected preventative action should be evaluated, recorded, and monitored. Improvement should be promoted, and senior management should be involved. Being proactive is an essential tool in quality system management. Examples are succession planning, training, capturing institutional knowledge, and so forth.
It will take a long time and a lot of work to complete the change. However, companies simply can't afford to sit on the sideline and wait for the new paradigm to be better defined before taking action. FDA's Quality Systems train has left the station and the effects are already being seen in enforcement actions.
1. FDA, CGMP for the 21st Century, available at www.fda.gov/cder/gmp/gmp2004/GMP_finalreport2004.htm , accessed Jan. 7, 2008.
2. FDA, Quality System Regulation for Devices, 21 CFR Part 820 (1996).
3. ANSI/ISO/ASQ Q9001-2000, Quality Management Systems—Requirements (American Society for Quality, 2000).
4. ISO/DIS 13485 (April 1996).
5. FDA, Process Analytical Technology (PAT) Initiative (2002), available at www.fda.gov/Cder/OPS/PAT.htm, accessed Jan. 7, 2008.
6. I. Maes and B. Van Liederkerke, "The Need for a Broader Perspective if Process Analytical Technology Implementation Is to be Successful in the Pharmaceutical Sector," J. Pharm. Innovation, (Sept./Oct., 2006).
7. FDA, Quality Systems Approaches to Pharmaceutical Current Good Manufacturing Practice (CGMP) Regulations (Sept. 2006), available at www.fda.gov/cder/guidance/7260fnl.htm, accessed Jan. 7, 2008.
8. CPGM 7356.002 Compliance Program—Drug Manufacturing Inspections, available at www.fda.gov/cder/dmpq/compliance_guide.htm.
9. FDA, Q8 Pharmaceutical Development (May 2006), available at www.fda.gov/cder/guidance/6746fnl.htm.
10. FDA, Q9 Quality Risk Management (June 2006), www.fda.gov/cder/guidance/7153fnl.htm.
11. Pharmaceutical Manufacturing Research Project—Final Benchmarking Report (Sept. 2006), available at www.olin.wustl.edu/faculty/nickerson/results/.
12. G. Cokins, Quality Progress (Sept. 2006), definition is from the article entitled "Measuring the Cost of Quality For Management", available at www.asq.org/qualityprogress/past-issues/index.html?fromYYYY=2006&fromMM=09&index=1, accessed Jan. 22, 2008.
13. P.B. Crosby, Quality is Free: The Art of Making Quality Certain (McGraw-Hill, New York, NY, 1979 and Mentor Books, Denver, CO, 1992).
Andrew G. Edwards is a senior consultant at EduQuest, 1896 Urbana Pike, Suite 14, Hyattstown, MD 20871, tel. 301.874.6031, fax: 310.874.6033, AndyEdwards@EduQuest.net
What would you do differently? Email your thoughts about this paper to firstname.lastname@example.org and we may post them on pharmtech.com.
For more on this topic, see:
Submitted: Aug. 6, 2007. Accepted: Nov. 26, 2007