Predictive modelling: putting ICH guidelines to work in process validation

By: Jason J. Kamm, Philippe Cini
Through the International Conference on Harmonisation (ICH) process, regulatory bodies in the EU, US and Japan have been moving steadily towards a sciencebased approach to drug development that will revolutionize the way pharmaceutical companies validate processes and ensure product quality. Concepts such as PAT, quality by design (QbD) and design space (DS), which figure prominently in ICH Q8 and ICH Q9,^1,2 encourage greater scientific understanding of processes and products, and hold out the promise of a lighter regulatory burden for companies that adopt such principles. Although the regulatory agencies have provided some helpful direction about how to put these principles into practice, they have not laid out a stepbystep guide. In the absence of such guidance, many companies have been slow to take advantage of the opportunities that these changes offer. That hesitation is understandable. Consider the uncertainty surrounding validation of manufacturing processes — a key milestone in the drug approval process. The industry knows that the accepted approach to validation — the successful processing of three consecutive batches — is antiquated. Further, FDA, for example, now says that it never meant the threebatch guideline as a hard and fast rule. As recently as 2003, in the pages of this publication, a review of the literature on validation uncovered wide variations among experts in their understanding of the term and the regulatory requirements associated with it.³ However, by following a proven approach to sciencebased validation, forwardlooking companies can cut through the uncertainty, move past outdated methods of validation and begin to realize the potential of recent ICH guidelines:

• reduced compliance risk
• greater regulatory flexibility
• more robust processes
• significant financial benefits.

The goal: managing variability
As anyone who has been involved in validating a pharmaceutical process knows, the apparently simple equation 'fixed raw materials + fixed process = quality' entails a high degree of complexity. Raw materials usually vary from batch to batch and those variations interact in complex ways with the many variable aspects of the manufacturing process. Current approaches to validation are premised on the notion of holding steady by trying to eliminate variability entirely — an endless and ultimately hopeless task. A more realistic and rewarding approach to validation recognizes the inescapable fact of variability. Instead of seeking to stamp out variability, such an approach seeks to manage it by developing a process that can accommodate the range of variables while still maintaining product quality. Such a process would operate within the DS, 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" (ICH, 2005). With an understanding of the DS, the manufacturing processes within that DS could be continuously improved without further regulatory review. The manufacturer would gain more regulatory room to operate, and regulators could be more flexible, using, for example, risk-based approaches to reviews and inspections set forth in ICH Q9 'Quality Risk Management'.
An effective tool: predictive modelling
An effective and practical way to achieve and demonstrate the requisite level of process understanding lies in developing predictive models of the form Y=f(X). Y is the process output that measures the performance of the process and the Xs are process inputs, controlled process variables and uncontrolled process variables.
In pharmaceutical manufacturing, the process output (Y) will be a function of raw material properties and process parameters (Xs). These models should identify critical raw material and process parameters, and reliably predict the behaviour of the process with the wide range of complex multivariate relations among those critical parameters and the outputs they generate.
Although we understand the first principles of kinetics, thermodynamics, heat and mass transfer, we don't have data about the possible behaviours of all the compounds we deal with. Our predictive models for the behaviour of any novel formulation must, therefore, be developed empirically. While validation has always entailed at least some basic empirical techniques, such as simply testing whether a given set of process parameters produces an in-specification result, the application of sophisticated statistical modelling has often lagged. Used with other techniques and bodies of knowledge — raw material science, formulation science and engineering — statistical modelling can help realize the potential of ICH Q8 and Q9.  
Two examples illustrate the power and value of statistical modelling for validation.

Figure 1: Modelling historical data.

Historical data. The first example focuses on the statistical modelling of historical data (existing information about a process) undertaken by a pharmaceutical manufacturer that was having trouble with tablet water content, which was varying unpredictably. The manufacturing process included blending, granulation, milling and tablet compression. As the water content was measured following compression, the problem showed up there, but the project team suspected that the root cause of the variability lay elsewhere in the process. So, working with historical data, the team performed a regression analysis to identify and quantify the drivers of variation of tablet water content. The results are displayed in a table of effects (Figure 1). The table of effects shows the moisture problem related to the blend water and the presence of coarse particles after milling (screen 20). In other words, the variability in particle size was the driver of variability in tablet water content. Therefore, it was those two critical parameters that needed to be monitored and controlled to keep the variability of the tablet moisture within specification. In considering how to control particle size, the team noted that the mill was being fed by hand rather than in a uniform way. They installed an auger to solve the problem, thereby doing exactly what FDA is now encouraging: understanding the drivers of variability in the process and putting good controls in place.
New data. Experiments can also be designed to generate new data regarding process parameters for regression analysis. For example, the maker of a pain management product had encountered wide variations in its dissolution rate.⁴ Dissolution sometimes occurred too rapidly, which could be lethal to patients, and sometimes too slowly, which could cause patients to suffer unnecessarily. A project team reviewed the available historical data and interviewed a large cross-section of relevant personnel. Using a systematic approach for root cause analysis, the team narrowed the range of possible causes of the unacceptable dissolution rate to nine potential variables — four properties of the raw material (RM), including three related to the API and one related to an excipient, and five process variables (PV) such as temperature and feed rate. To test each of the nine variables at three different values, the team undertook a design of experiments (DoE) to screen out irrelevant variables and find the proper values for critical variables. The team created an L27 orthogonal array design constructed in nine blocks (one for each different raw material blend) and within each block, the different process parameters.

Figure 2: Quality of fit.

Multiple regression analysis capabilities of a software application were used to provide estimates of the effects of the variables on dissolution, their interactions and their statistical significance. The number of significant variables having an impact on dissolution was narrowed to six, with one of those — a process variable — dominating. The final regression model for dissolution was able to explain approximately 95% of the variation in dissolution (Figure 2).

Figure 3: Determining the design space.

Although that one process variable was found to have the greatest influence on dissolution, other process and raw material variables, and their interactions clearly played a role. The various permutations of the settings for all of these variables that still result in an inspecification rate of dissolution (and other product properties) constitute the DS for the manufacture of the product (Figure 3).

This picture of the DS was created using the optimum settings for each of the six significant variables. It brings together 15 3D response surface plots, each of which was originally created in the modelling software. The X and Y axes are made up of the DoE variables, and the Z axis (the contour curves) represents dissolution (the response variable). In the red regions, dissolution is out of specification. In the green regions, the dissolution rate is within specification. By finding the DS in which PV2 — by far the most influential variable — interacting with other variables can produce inspecification dissolution, it is possible to optimize the path to achieve the desired rate of dissolution. In this case, the PV2 variable needs to be kept at a high level while the RM2 variable should be maximized. The RM1 variable should be kept in the centre of the range used, thereby maximizing the 'green space'.
The optimum conditions were entered into the model, two confirmation batches were processed and the results conformed to those predicted by the model. Ultimate confirmation of the power of the technique came with the successful validation and launch of the product.
The results: scientific rigour
When carefully sequenced and applied, predictive modelling can provide the rigorous, scientificallybased understanding of products and processes envisioned by ICH guidelines. As the examples illustrate, using predictive models:

• Acknowledges that raw materials and processes entail some inherent variability.
• Allows for the management of that variability.
• Recognizes that not all variables are equally important, which, in turn, allows risk managementbased approaches to regulation.

On the go...

Predictive modelling also helps enable continuous process verification, defined by ICH Q8 as "an alternative approach to process validation in which manufacturing process performance is continuously monitored and evaluated" (ICH, 2005) and recognized by PAT guidelines and ICH Q8 and Q9 as more effective than the traditional threeconsecutivelots exercise. By identifying critical parameters and interactions, predictive modelling points to those areas that should be monitored — with PAT and other techniques — and controlled. Through continuous monitoring, a manufacturer can accumulate an understanding of the process, refine the characterization of it, and make adjustments within the DS that don't require refiling with regulators. It is important to reiterate that although predictive modelling is an indispensable element of achieving the desired process knowledge, process characterization should not rely solely on quantitative statistical models. It should also employ scientific and engineering knowledge, and univariate experiments. Further, all of these scientific and quantitative techniques should be complemented by basic scientific knowledge and the insight, wisdom and experience of people in the organization who work with the process.
Finally, although recent guidelines encourage frequent communication with regulatory authorities, some organizations, fearing they may become prematurely entangled in regulatory red tape, have been wary of such open communication. However, when armed with increased process understanding, and the powerful data derived from predictive modelling, manufacturers can undertake such communication with confidence. By demonstrating the scientificallybased understanding of processes that regulators are encouraging, manufacturers can establish the mutual trust that is necessary for a productive working relationship, and for realizing the full benefits of the revolution in validation.
Jason J. Kamm is a Managing Consultant with Tunnell Consulting (PA, USA).
Philippe Cini is a Vice President with Tunnell Consulting (PA, USA). http://www.tunnellconsulting.com
References
1. ICH Harmonised Tripartite Guideline: Pharmaceutical Development, Q8, 2005. http://www.ich.org
2. ICH Harmonised Tripartite Guideline: Quality Risk Management, 2005. http://www.ich.org
3. M. Helle, J. Yliruusi and J. Mannermaa, Pharm. Technol. Eur.,15(3), 52–57 (2003).
4. J. Kamm, Pharmaceutical Manufacturing, May (2007). http://www.pharmamanufacturing.com

Validation by Numbers

By: Russell E. Madsen Jr.
How many samples should I take? Is the assay method validated? Is this result really out of specification? Should I adjust the tablet weight? Is this process under control? Should I reject the batch? The pharmaceutical industry has always faced questions such as these, and they may be answered best from a statistical perspective. Unfortunately, many quality-assurance workers and production managers charged with answering these types of questions do not have an adequate working knowledge of statistics. They find it difficult to answer these questions and to understand answers that statisticians provide.
The book Validation By Design: The Statistical Handbook for Pharmaceutical Process Validation, by Lynn Torbeck, a member of Pharmaceutical Technology's Editorial Advisory Board, contains information useful to people who are new to statistics and to employees responsible for implementing statistical techniques that monitor and control pharmaceutical production processes and quality-assurance activities. The book was written specifically to address the statistical issues contained in the US Food and Drug Administration's November 2008 draft guidance for industry titled Process Validation: General Principles and Practices, which is reproduced in its entirety in an appendix. Torbeck's book can form the basis for interdepartmental discussions and for an understanding of the statistical techniques that are consistent with the intent of the guidance.
An important feature of the book is its interpretation of the guidance's statistical implications. The author rewrote the guidance's statistical content as a series of self-audit questions that cite specific lines in the guidance document. For example, lines 27–29 of the guidance state, "The lifecycle concept links the product and process development, qualification of the commercial manufacturing process, and the maintenance of the process in a state of control during routine commercial production." The associated self-audit question asks, "Is the process in a state of control during routine commercial production?" The question is followed by a brief explanation of the term "state of control" and a reference to the chapter of the book that describes the concept. This self-audit format leads the reader through the various statistical techniques that enable compliance with various sections of the guidance.
A second important feature of the book is the chapters that elucidate statistical methods and concepts. Each of these chapters is written in a standard format that contains subtopics such as "Other Names," "Acronyms," "Definition," "Related Topics," "Calculation," "Illustration," "Cautions," "Advice," and "References." This consistent categorization enables the reader to understand the statistical concepts and decide whether their use is appropriate in a particular situation. The chapters explain simple statistical concepts such as average and relative standard deviation, as well as complicated ones such as control charts, root-cause analysis, process mapping, interquartile range, and Plackett–Burman designs.
The book should prove useful to employees charged with developing a self-audit program to measure the company's level of compliance with the process-validation guidance. It also would be a solid basis for writing validation protocols.
The book cannot be considered an introductory text about statistics because the topics are not presented in depth. Still, its references provide necessary information for readers who wish to delve further into any of the subjects. According to the author's preface, the book is for "those engaged in meeting the requirements of the FDA process-validation guidance." The questions that interpret the guidance, the author's explanations, and the chapters about statistical methods and techniques should do much to help personnel meet those requirements. 

GE Healthcare — efficient facility validation

GE Healthcare's Victor Bornsztejn offers insight to help companies achieve more efficient validation, which ultimately leads to a wide range of benefits including cost containment, reduced approval times and assurance of traceability.
Q1: What are the main issues with how manufacturers are currently validating their facilities and what impact do they have?
Manufacturers must achieve regulatory compliance of facilities, laboratories and equipment more efficiently (i.e., faster, with standardized documentation, elimination of unnecessary work, reduction of errors, etc.). Current facility validation processes are based on traditional models where validation tends to start during or after facility design and ends with the handover of the facility. In our opinion, this validation effort begins too late and can lead to significant issues such as lack of transparency and standardization, and validation projects running over-budget and over-time. Many manufacturers start by putting together a very large and detailed site validation master plan (VMP) that contains a plethora of information such as:

• system lists and descriptions
• impact assessments
• validation rationales
• process descriptions
• equipment, process and analytical validation.

Such a bulky VMP is difficult to write, review and maintain, making it difficult to control and, subsequently, close-out (complete) projects. Critically, this style of VMP is not a flexible document and usually needs constant revisions and rewrites. The impact can be incredibly significant as new facilities suffering delays in the go-live date can lead to delays in product manufacture, resulting in additional costs and loss of revenue.
Q2: What facility validation services do GE Healthcare offer?
We believe there is a disconnect between drug/process development, validation of facilities and ongoing facility compliance management. This is why we have launched a global service for complete facility validation for new or existing builds, including all facilities, utilities, equipment, computer systems etc.
We understand that many manufacturers work against a background of high activity with a proportional level of supporting documentation. Therefore, we have introduced a Modular Validation Platform (MPV) that allows manufacturers to benefit from the development of a workable and economic validation policy that is compliant with all current and foreseeable initiatives. It minimizes exposure of surplus information, while optimizing the level of control and ease of inspection. It also provides a springboard for future validation programmes.
Our approach is for planning to start before the facility has been designed. It involves thinking about validation holistically and not developing detailed documents too early — invariably they will need modifying and adapting as the validation programme progresses. We start putting together a modularized document/master plan that begins with the high-level detail (e.g., setting out intent and broad observations) and then building in other details as the facility design and construction progresses (Figure 1). We use automated document generation software to make the process faster and more accurate, and all validation decisions are supported by a unique regulatory database; together these provide an efficient compliance status capability with full traceability.

Q3: What benefits are provided by using your Facility Validation approach?
There are a number of advantages to the MVP compared with traditional facility validation strategies, including:

cost containment

no duplication of information

minimized impact of change

minimum exposure during inspections

timely generation of protocols

no repeat generation of test detail

optimized pre-approval of test detail

minimal discrepancies

reduced document review/approval times

proven history of acceptability

assurance of traceability to agreed user requirements.

As part of overall project management, we also apply operational excellence methodologies such as Lean and Six Sigma where appropriate.
The main point to stress here is that our approach focuses on building long-term relationships with customers through sharing our experience and expertise across the entire product and facility life cycle. This can help reduce approval times, which means facilities can go on-line earlier and products can get onto the market quicker. It also enables customers to become more self sufficient for the future. Q4: Do you believe your customers and potential customers have enough industry knowledge about facility validation?
We believe that industry knowledge is highly variable and dependent on one critical factor — resources. Validating a facility is expensive in terms of costs, personnel and time. Many people in the industry do not understand how poorly validation can be managed, and how much time and money could be saved if facility validation was effectively planned during facility design, initiated BEFORE the facility is being built and efficiently executed during and after facility construction to ensure timely completion of validation reporting and, ultimately, facility approval by regulators.
Q5: What are the main issues faced by your customers?
The biggest issue is regulatory compliance. Poorly or incorrectly validated facilities, processes and life cycles cost manufacturers millions of dollars in lost product and time. Customers are regularly receiving citations and warnings that are linked to their validation, or lack thereof, usually for reasons that can be avoided by use of a holistic validation strategy.
Q6: At which point in the life cycle do you get involved?
For new facilities, we ideally need to be involved right from the start — at least when customers are planning their new facility. We can start in an advisory capacity even before a decision has been made on a new facility, which allows an ideal risk-based approach to be built from the beginning. Otherwise, we can come in at any stage as we can also work with existing facilities that need to update/review their validation programmes.
Q7: How do you see facility validation developing in the future?
There is a lot of discussion regarding ‘risk-based’ validation and regulatory authorities are advocating the use of this type of approach. However, as an industry, we are still some way from fully realizing that. Inevitably, the industry will develop strategies to use risk-based approaches to manufacture and validation. We’re also going to see early process development become more important (e.g., improved process knowledge enabling PAT-driven processes) and more leveraging of automation to drive document generation and project management. The onus is on companies, such as ourselves, to help educate manufacturers in how to validate early and validate correctly — which can ultimately make a huge difference to their business.
www.gelifesciences.com/service
Victor.Bornsztejn@ge.com

A Perspective on Computer Validation 3

Perspective on enforcement and observations
We can understand FDA concerns about computer systems as we look at the following examples. The "Therac-25" medical-device software went awry between 1983 and 1987, overdosing patients with X-rays (21). FDA stopped the use of the device. According to the recollection of the authors, Wyeth Laboratories Inc. is believed to have received the first computer validation–related 483 from Center for Drug Evaluation and Research (CDER) in October 1983 for a lack of documentation for the validation of a computer system used for the statistical analysis of data. In the late 1980s, the generic-drug scandal caused concern because data had been falsified. One of the authors served as a witness in a grand jury investigation of the generic-drug scandal. Interestingly, Phil Piasecki, a former FDA official who later worked in the industry, once told one of the authors of an observation he made on computer systems. He noticed that the red light of a dehumidifier in a data center was on. After inquiring about the dehumidifier's purpose, he cited the company for not having a standard operating procedure (SOP) for its use and maintenance.
These examples illustrate the wide range of enforcement actions and levels of impact on patients' safety (from serious to limited and indirect impact). Since the mid 1990s, we have also seen numerous observations relating to computer-validation issues, ranging from "the system is not validated" to "the SOP is not followed." Since the early 2000s, however, it seems the number of observations has decreased. Between 2000 and 2005, the number of Warning Letters issued decreased by 50% (22), perhaps resulting from a shift in FDA's focus after the 9/11 attack in 2001 to homeland security and food safety (23). At the same time, FDA faces a decrease in budget spending (24). FDA's presentation titled "Data Integrity, Another Looming Crisis," revealed its recent enforcement has focused on record integrity (25). The presentation made it clear that based on recent inspection findings, FDA would refocus on the integrity of e-records and train inspectors in this topic. Despite challenges that FDA faces, it has increased significantly the understanding and knowledge of computer systems in the industry. The agency actively participates and has a Computer System National Expert representing FDA in industry forums (e.g., GAMP), as well as providing computer validation training courses internally and through selected providers. Hence, although we are now seeing fewer computer-validation 483s, partly because FDA is citing predicate rules instead of Part 11, these 483s are more meaningful and should not be interpreted as a reflection of relaxed enforcement. Perspective on computer-validation practices
The Standish Group's "Chaos" reports, indicate that "incomplete requirements and specifications" and "changing requirements and specifications" are two of the top three "project challenged factors" for computer-system projects (26). This assessment seems to hold true in today's computer-validation practices. An informal poll conducted by the authors indicates that user requirements are the main challenge when validating a computer system. User requirements have been a factor since the early years of computer validation. Nonetheless, the general understanding about how to conduct computer validation in the industry has increased since the 1990s, and most companies now have groups or departments dedicated to computer validation. Since the introduction of the Sarbanes–Oxley financial regulations for a publicly traded company, more and more information technology (IT) departments have established compliance groups. It is generally accepted now that IT infrastructure must be qualified to meet growing regulatory requirements across the business. 
Conclusion
FDA has published more than 30 documents related to computer systems and computer validation, and the term computer validation is no longer foreign to the industry. Computer-validation practices and regulations are evolving and reaching the maturity stage of other validation disciplines, even if more recently it seems that the efforts on computer validation and Part 11 compliance are less apparent than before. This complacency might be a result of the wait-and-see attitude toward what the new Part 11 regulation amendment might bring and may also result from the perception that computer validation–related 483 observations have decreased in recent years.
The intent of Part 11 regulations was to allow businesses to be more efficient, to enable automation, and to generate less paper documentation. Yet, these goals have not been fully realized. Most companies addressed how to meet and comply with the Part 11 regulations, but they did not necessarily develop business strategies to take full advantage of what the regulations allow the companies to do.
As technology evolves, computer validation also will change. We will see what the next 10–30 years bring. Could we see programmable drugs based on nanotechnology? For example, one article suggests that nanodiamonds could be useful in biological applications such as carriers for drugs (27) or perhaps as bioerodible implants with programmable drug release (28). Computer validation will need to be continuously simplified, standardized, and automated while reflecting the growing complexity of designed and engineered drugs and delivery systems. Acknowledgment
The authors thank Alan Kusinitz, managing partner of SoftwareCPR, and George R. Smith Jr., FDA consumer safety officer at CDER's Office of Compliance, for their input and review.
Rory Budihandojo* is the computer validation manager at Boehringer-Ingelheim Chemicals and a member of Pharmaceutical Technology's editorial advisory board, rbudihandojo@bichemicals.com
Steve Coates is the director of computer system quality assurance at Wyeth. Ludwig Huber, PhD, is a compliance expert at Agilent Company. Jose E. Matos is manager of manufacturing systems and process automation at Bristol-Myers Squibb. Siegfried Schmitt, PhD, is the quality director at GE Healthcare Global IT. David Stokes is the life sciences manager at Business & Decision. Graham Tinsley is president of THINQ Compliance Ltd. Maribel Rios is senior editor of Pharmaceutical Technology.
*To whom all correspondence should be addressed. The scope of this article is specific to the healthcare industry and to the view of the authors or FDA, which may not necessarily reflect the view of the companies where the authors and reviewers are employed.
Where were you 30 years ago?
Rory: "I was in the UK, going to school."
References
1. E. Kübler-Ross, Five Stages of Grief, http://www.businessballs.com/elisabeth_kubler_ross_five_stages_of_grief.htm.
2. Risk Management, http://www.fda.gov/oc/mcclellan/riskmngt.html
3. IVT Proposed Validation Standard VS-2, Computer System Validation, IVT 3 (2002).
4. Guide to Inspection of Computerized Systems in Drug Processing (1983), http://www.fda.gov/ora/Inspect_ref/igs/csd.html.
5. Timeline of Key FDA Software Documents, http://www.softwarecpr.com/libraryframepage.htm.
6. ITG Subject: The Computer in FDA-Regulated Industries (1976), http://www.fda.gov/ora/inspect_ref/itg/itg23.html.
7. ITG Subject: The Computer in FDA-Regulated Industries Part II Computer Hardware (1977), http://www.fda.gov/ora/inspect_ref/itg/ itg29.html
8. Guidance for Industry: Computerized Systems Used in Clinical Trials (1999), http://www.fda.gov/ora/compliance_ref/bimo/ffinalcct.htm.
9. Guide To Inspections of Computerized Systems in The Food Processing Industry (unclear publication date), http://www.fda.gov/ora/inspect_ref/igs/foodcomp.html.

A Perspective on Computer Validation 2

A discussion about regulations would be incomplete without including 21 CFR Part 11 regulations of electronic records and electronic signatures. In 1991, industry and FDA representatives met to determine how to accommodate paperless record systems under 21 CFR Parts 210 and 211. Specifically, industry requested FDA's official position on substituting 21 CFR Part 211, section 186 "full signature, handwritten" with an electronic signature (12). In response, FDA publshed its progress report Electronic Identification/Signature Working Group in 1992 (13). The report identified seven key issues: legal acceptance, regulatory acceptance, enforcement integrity, validation and reliability, security, standards, and freedom of information. The final regulation was published in 1997, and although the regulation is now 10 years old, discussions and issues still revolve around most of these points. Former FDA Commissioner David Kessler is believed to have said that he was surprised by the number of experts he found when he searched the topic of Part 11 on the Internet, especially because FDA was still trying to address and develop a better understanding of the implementation and enforcement of the regulation. Even more interesting in hindsight is how the initial request for FDA to address a specific section of 21 CFR Part 211 has now evolved and encompassed all other aspects of the GMPs.
Even now, some uncertainty about Part 11 regulations remains. A contributing factor to this confusion may be the fact that earlier FDA guidelines on the regulation were revoked in 2003 (14) and replaced by a single guideline (15) with the intention of adding other guidelines later. FDA's reasoning behind withdrawing these guidelines (Fed. Register, Docket 00D-1540 in Feb. 2003) was "to avoid loss of time spent by industry in their efforts to review and comment on Part 11 issues that may no longer be representative of FDA's approach under the new GMP initiative." Since then, no additional guidelines have been issued, but FDA is working on a Part 11 amendment (16).
Perspective on the industry's approach

Figure 1. "Waterfall" life cycle method.

While FDA published its official regulations and guidelines, the industry also was actively addressing computer validation. In the late 1980s to the early 1990s, the Pharmaceutical Manufacturers Association (PMA, now Pharmaceutical Research and Manufacturers of America, PhRMA) Computer System Validation Committee (CSVC) led by Ken Chapman was the industry's main forum to discuss computer-validation issues. One series of discussions about system-development life cycle (SDLC) methodology resulted in the selection of the waterfall life cycle model (see Figure 1). In the mid 1990s, a variation of this waterfall model, the V model, became more popular (see Figure 2) and is still the model of choice. Looking back, there is no significant change in the computer validation SDLC. This is rather surprising because in some cases, using other methodologies might be advantageous (e.g., rapid prototyping methodology, which involves configuring how software should operate first, then documenting the final configuration and functional operation, followed by the software's operational verification). 
PMA also published other computer validation related articles, mainly in Pharmaceutical Technology, which has published more than 40 articles about computer validation. Historically, this publication played a key role in shaping and spreading understanding of computer validation in our industry. Even individuals from FDA published articles in the journal back then (17). Sadly, this is no longer the case because, for better or worse, FDA is now stricter about allowing its employees to publish. (A list of computer validation–related articles is online at http://pharmtech.com/.) The PMA CSVC was dissolved in the mid 1990s, and the US Parenteral Drug Association (PDA) became the industry's main forum. Part of PDA's computer-validation committee efforts focused on software-supplier audit and the creation of an audit repository center (ARC) (18). The ARC concept, however, has had varying success. Currently, the repository is managed by Syntegra. Since the late 1990s, the main industry forum for computerized systems validation has been the ISPE good automated manufacturing practice (GAMP ) group. The GAMP guideline is widely considered to be the de facto pharmaceutical-industry standard in computer validation. The current GAMP 4 version introduced the risk-management concept into computer validation in 2001, aligning with FDA's effort on GMPs for the 21st century. Currently, GAMP is working on version 5, which is slated for publication in late 2007 or early 2008. It is important to note that GAMP 4 explicitly excludes 21 CFR Part 11, although a separate guideline has been produced.
The healthcare industry has actively addressed Part 11 regulations. ISPE/GAMP and PDA both have published guidelines on Part 11, hoping to help clarify the implementation and compliance to the regulation. As discussed previously, there was much confusion about the implementation of Part 11 when it was initially introduced. Accordingly, PDA arranged a public FDA conference about Part 11 in June 2000 (19). The conference reinforced the public's concerns about Part 11, and based on industry's reaction, FDA issued the final guidance on Part 11 in August 2003 (15). FDA also scheduled a Part 11 public hearing in June 2004, again allowing the public to voice concerns and suggestions about Part 11. Unfortunately, President Ronald Reagan's funeral fell on the same day as the scheduled public hearing date, and the day was declared a federal holiday. The meeting was therefore cancelled and never rescheduled, prompting industry, as a "21 CFR Part 11 Coalition" to file a public petition letter in September 2004 (20). Since then, there has been a semblance of calm on the topic as FDA considers rewriting Part 11 and industry adjusts to a risk-based approach. The resulting decline in the number of Part 11 483s may be related to the fact that FDA is citing the predicate rule directly, rather than Part 11. Until the new amendment is issued, FDA is indeed showing enforcement discretion regarding the implementation of Part 11. It should be noted, however, that Warning Letters with severe consequences were sent to Able Labs, MDS Pharma, and Ranbaxy, and the warnings related to electronic records, even though Part 11 was not mentioned.