In those days, it was very clear who were the regulators and who were the regulated. At that time, validation was still in its infancy and we argued incessantly over the terminology, definitions, and the ever increasing list of acronyms. Ken was working at Pfizer and was also on the program. I can still remember how well he argued his points with the FDA speakers. Ken was always kind in his rebuttal, but never backed down from his position. He didn't so much argue as he explored the issues in a way that was always respectful of the other speakers. I think that is why he was so effective.
When you spoke with Ken, you always knew that he heard every single word you said. He gave you the feeling that he had the utmost respect for you and your point of view and was truly interested in what you had to say. Although he could have been a masterful politician, he truly enjoyed his position in the pharmaceutical industry. To Ken, it was not work, but rather his hobby that he got paid for pursuing.
Ken always remembered my name and went out of his way to say hello and ask how my family was. He would always greet me and say "What's new in cleaning validation, Bill?" With Ken, these weren't just idle words--he really wanted to know. He was truly a kind person and will be greatly missed by all of us who knew and loved him, as well as by an industry that will be forever indebted to him. As colleagues, friends, and beneficiaries of his contributions to industry; we all know that Ken Chapman was a class act.
THE ANCIENT DARK AGES OF CLEANING--BEFORE GMP REGULATIONS
Prior to the issuance of the Good Manufacturing Practice (GMP) Regulations in 978 (Reference 1), cleaning was mostly an "orphan" responsibility, meaning it as not considered very important. It was often relegated to the last activity of the day and sometimes performed on third shift. It was usually assigned to the newest employees, i.e., those with the least experience. In addition, the procedures were often very brief, ambiguous, and very much open to different interpretations. In fact, the procedures were often one sentence, "Clean with hot soapy water."
The cleaning agent often was not specified and it was usually perfectly acceptable and understood that ordinary tap water would be used. Most companies cleaned to a "visually clean" standard that was certainly not defined quantitatively. In most cases, a second person did not verify the cleaning. The documentation of the cleaning was not generally recognized as of critical importance. If the equipment was discovered to be insufficiently clean in the light of the next day, then it was simply recleaned--no big deal and no investigation. It was certainly not a widespread practice to determine the consistency of the cleaning process, i.e., process capability studies were not generally performed.
In all probability, there were significant amounts of cross contamination of one product into another since multiple products were manufactured in the same equipment. Facilities were used that were not designed to be readily cleanable. For example, drop ceilings of cellulose type materials were often used that were not "water friendly" and therefore not regularly cleaned since they were considered as non-product contact in nature. Many of the older facilities had not been designed using good engineering practices such as dust control, anterooms, and adequate principles of operator protection. Many of the piping systems had low spots and "deadlegs" that were virtually impossible to clean. Air flow was not controlled and was often insufficiently filtered to remove micron-sized particles.
Many companies had not developed analytical methods sufficiently sensitive to detect the relatively small amounts of cross contamination that we now know can be medically significant. Most sterile products were manufactured using terminal sterilization whereby any microbial contamination could be controlled and eliminated in the final step of the manufacturing process. The use of barrier isolators in laboratory testing had not matured; therefore, there were false positive micro tests that caused considerable additional confusion.
Looking back at that time period, one might conclude that we had some serious issues in light of today's standards. However, it should be stated that many companies probably did a decent job of cleaning. There were no regulations which required the scientific proof of the adequacy of the cleaning. Therefore, since the proof was absent, the worst case assumption must be made that the equipment was not adequately cleaned.
THE IMPACT OF GMP REGULATIONS AND FDA GUIDELINES
One might assume that with the issuance of the Good Manufacturing Regulations in 1978, the Food and Drug Administration (FDA) drew the "line in the sand" and that cleaning improved dramatically and immediately. Unfortunately, this did not happen. The regulations proved to be somewhat interpretative and many companies were not at all sure what they were being required to do. Many considered that the requirements were unfair and too restrictive and that full compliance would be costly.
This "resistance to change" led to the situation whereby regulatory inspectors and investigators would cite a company for deficiencies in their cleaning program only to find that the deficiencies had not been corrected by the time of the subsequent inspections. During the years immediately after the issuance of the GMP regulations, most of the citations were DD Form 483 citations, meaning that the company did not have a deadline by which the issues must be addressed and corrected. Problems were noted, but the manufacturing and cleaning procedures were not immediately changed and products were continuing to be manufactured and released to the market.
In defense of the companies, the work involved in validating or proving the adequacy of the cleaning procedures was potentially staggering. Many of the product release assays were "wet chemistry" methods and were not sensitive enough to determine the trace amounts of residues remaining on equipment. Many companies "did the math" and found that tens of thousands of samples might be required because of the hundreds of products coupled with the numerous individual pieces of equipment involved in the manufacture and packaging of a pharmaceutical product. There needed to be some clarification or explanation of how this huge problem could be addressed.
Eventually, FDA became very impatient and took two very proactive steps to hasten the improvement of cleaning in the industry. First, they issued a guidance document in 1993 entitled, "Guide to Inspection of Validation of Cleaning Processes" (Reference 2). Although intended to serve as guidance for inspectors and investigators, this document spelled out for both regulatory and industry the regulatory expectations for cleaning procedures, cleaning approaches, setting of limits, and validation of the cleaning processes.
By this time the number of citations issued for cleaning was beginning to grow in an exponential fashion. Also, the FDA increased the serious nature of cleaning program deficiencies by issuing Warning Letters to many, many companies. The Warning Letter is much more serious than a 483 violation and contains language such as, "Your products are considered misbranded and adulterated." Another important aspect of the Warning Letter is that the company is required to respond with an indicated corrective action in not more than 15 working days.
These events caused cleaning to suddenly become a premiere issue for the industry. FDA, in effect, threw down the gauntlet and said "We consider cleaning a critical process and if you don't have validated cleaning processes, then the worst case assumption will prevail, namely that your cleaning program is inadequate and your products are contaminated." At about the same time, FDA took Barr Labs to court (U.S. versus Barr Labs,) in an historic trial. Although there were many issues in this case, the judge ruled that Barr must not only identify the cleaning agents used in their cleaning procedures, but must also validate or prove that the cleaning agent itself was removed by the final rinsing, that is: that cleaning validation studies must also be performed for the cleaning agent(s) used in the cleaning procedure since this could also represent a possible contaminant of pharmaceutical products.
THE EVOLUTION OF CLEANING FROM ART TO SCIENCE
Because of the events described in the previous section, cleaning began an evolution from an "art" activity to one based not only upon science but upon good science. Whereas many of the older cleaning procedures were not really subjected to the recognized principles involved in cleaning, a new generation of cleaning procedures began to emerge that had been through the same development process as the pharmaceutical products themselves. Cleaning processes were no longer an "afterthought" or some aspect that could be developed later, but rather an "FDA hot button" that should be considered from the very beginning of the product development cycle. Creative thought was put into the cleaning process to determine what makes a residue difficult to remove from equipment and the roles of solubility, solvent, mechanical action, flow rates, temperature, contact time, and many other variables in the removal process. In essence, the scientific principles of the cleaning process began to be more fully explored and understood.
There were also many other positive activities that occurred. Suppliers of cleaning agents began to "partner" with pharmaceutical companies in developing optimal cleaning agents and procedures that were product specific. Companies embarked on truly understanding the adequacy or inadequacy of their current cleaning procedures by means of the validation process. There were many surprises during this investigative process. Some cleaning procedures were inadequate or inconsistent and were revised, often extensively. There were many important, peripheral issues that came to light. For example, cleaning agents were previously often ordered, received as used directly by production units without Quality Assurance (QA) approval. This practice of exposing product-contact manufacturing and packaging equipment surfaces to non-formulation ingredients without QA approval or Quality Control (QC) testing needed to be stopped. Cleaning had to be integrated into the quality systems approach. For example, cleaning procedures needed to be reproducibly performed and placed under change control whereby changes could not be made to a procedure once validated.
Since testing of equipment surfaces is a major part of the validation of cleaning, a virtual explosion of technology occurred in which techniques and technologies were developed and applied explicitly to the cleaning process. It was no longer acceptable to use a relatively insensitive colorimetric and ultraviolet test for cleaning samples. Likewise, it was not completely acceptable to "test until clean" a process whereby samples were repeatedly tested until a certain acceptable level was achieved. More testing was done by powerful, sensitive, and sophisticated instrumental methods of analysis such as high performance liquid chromatography (HPLC), total organic carbon (TOC), mass spectrometry, enzyme linked immunosorbant assay (ELISA), near-infrared spectroscopy (NIR), grazing angle FTIR spectroscopy, and Ion Mobility Spectrometry (IMS).
There were also many refinements made in the sampling technology and sampling technique. It was generally recognized by the scientists in the laboratories that the power of the analytical technique could not overcome poor sampling technology and inadequate sampling techniques. Specific companies emerged that provided high quality swab samples which were guaranteed to not interfere with the analysis of the samples. For example, first generation swabs often contained adhesives, binders, plasticizers, and other chemicals that interfered with the new sensitive analytical methods, thus giving false positive and negative results when used in combination with the new sensitive analytical techniques such as TOC which may have detection limits of 50 parts per billion and lower. Also, the variability in results achieved in early swabbing studies indicated that swabbing did not give a true indication of the amount of residues on the equipment. Due to the nature of the surfaces (roughness, porosity, scratches), the swabbing process did not remove the entire residue from the equipment, i.e., it is recognized as being an incomplete process. Thus, it became necessary to know the efficiency of the swabbing process in order to determine the true amount of residue remaining after cleaning. These so-called recovery studies were used to "correct" the data and calculate the true amount of residue remaining. Recovery factor determination is now both a regulatory expectation and an industry standard for cleaning studies and currently all 'state-of-the-art' laboratories are very familiar with the recovery concept.
Potential microbial contamination of equipment assumed a prominent role in the cleaning validation program. This posed an entirely different type of challenge since microbial residues are not visible at the level that can be medically significant. Although microbial analysis had been used for many years for sterile products and aseptic processing, it had not been adequately considered as a potential quality problem for non-sterile manufacturing facilities. Only in the last few years have scientists begun to apply the principles of microbiology to non-sterile facilities. Early indications are that this will be a product-specific situation where certain products and manufacturing facilities will require more microbial scrutiny because of the nature of the product and how it is used by the patient or consumer.
Each company will need to assess the potential for microbial contamination to compromise either the safety of the patient or the performance of the product. In some cases, microbial issues will not be significant because of the nature of the product, how it is used, or the current controls that are already in place to prevent adverse microbial events. For example, a company that manufactures products that do not support microbial growth and are used topically on intact skin may not require microbial studies. In other cases, because of the components in the product or how the products are used, microbial studies may be required in order to verify that the current cleaning procedures are adequate to prevent adverse events in the patient or product quality problems. For example, if the pharmaceutical products contain natural materials such as proteins and sugars which can support microbial growth, then microbial studies may be entirely appropriate. Another contributing factor is the equipment itself and how effectively the equipment is disassembled, cleaned, and dried. This is another important source of potential microbial contamination of products. The important point is that each company will need to determine the potential risk of microbial contamination and its possible consequences for their particular products. Much work has already been done by scientists to provide a high assurance that certain classes of products are probably highly unlikely to be affected by pathogenic organisms (reference to work of Jose Fernandez) either because of the inherent hostile environment for microbial growth created by the product itself or because of a lack of moisture to sustain or support microbial growth.
CLEANING PROGRAMS OF THE FUTURE
Having considered the past and current cleaning issues, it is appropriate to look to the future as to what can be expected based on the momentum of current cleaning programs as well the general health and public welfare. One of the most significant emerging truisms is that cleaning of pharmaceutical equipment and facilities is an integral part of product quality as well as a major parameter in the health equation. One can only speculate as to where we are headed with cleaning, but I would like to give you my assessment of what is logical, practical, scientific, and probable.
First of all, I believe the cleaning programs of the future will start with each company embarking on a fact-finding mission to learn more about their products, how they are used by their customers or patients, how they could be misused, how they are stored, controlled, and further processed. I was amazed a few years ago when auditing suppliers of Active Pharmaceutical Ingredients (APIs) to find that many of the API manufacturers were not aware of the specific manner (dosage form, daily doses, how administered, etc.) in which their customers, the dosage form manufacturers, actually used the APIs and thus in what manner they were used by the patient or customer. My personal opinion is that the ideal process will start with risk assessment. I further believe that the risk assessment must be comprehensive and should be more global in nature, that is: that it should consider the product activities both inside and outside the manufacturing facility. It should be 'cradle to grave' in nature.
Risk assessment should be done in phases. First, at the most general level, each company should ask and answer basic questions:
* Do I need to protect the raw materials, intermediates, and products from other products? How do I do that?
* What is the potential for cross contamination of products? What controls are in place to prevent, minimize, or detect these cross contaminations?
* Do I need to protect the operators from the products? How do I do that?
* How am I protecting the environment from the raw materials, intermediates, and products? And how am I protecting the raw materials, intermediates, and products from the environment?
* How can I prevent intentional criminal or bioterrorism contamination of my products? What controls and methods of detection of contamination or tampering are in place to ensure my product reaches the patient or consumer without contamination or degradation?
Once these basic risk questions and answers are evaluated, then more specific questions can be addressed and the branching can continue much as a decision tree might look. The details of the additional phases of risk assessment are beyond the scope of this article, but obviously should include identification of worst case products, most difficult to clean equipment, and examination of locations where contamination and cross contamination are most likely to occur.
The risk assessment should be quantitative. If risk is too high, what can be done to reduce the risk? What controls can be put in place to reduce the risk? It is already obvious that if minimization of cross contamination is one of the main goals, then there must be greater control of potent and potentially toxic products than for less potent products. However, if protection of all products from tampering, bioterrorism, or intentional contamination is the goal, then a different approach or strategy must be used and looking solely at the cleaning of the most potent product(s) would not suffice.
The point here is that the result of the risk assessment should be the development of a strategy which is directed toward achieving finite goals. In some cases, the goal may be to clean the equipment. In other cases, it may be to contain the product. Recently, there have been numerous studies documenting the exposure of health professionals to cytotoxic and other potent products. (References 3-7) In these cases, significant blood levels of potent products were found in blood and urine specimens of doctors, nurses, pharmacists, and other health professionals involved in the administration of potent drugs to patients. The articles speculate that the entry routes are via the skin, lungs, and oral pathways. In some cases, residues of drugs have been found on the outside of vials and ampoules of injectables. Another theory that has been offered is that during the removal of air from injections immediately prior to injection, micro-aerosols of the product may be "injected" into the environmental air and then breathed in by the health professional.
Much of the control of cleaning will depend on the availability of fast and sensitive analytical and testing technology. It seems to me that we may want to learn from the food and beverage industry in some of these areas. In many of the trade publications for food safety there is currently tremendous interest in the rapid testing of food processing surfaces for microbiological residues. This is important because of recent contaminations of various foods both domestic and imported. There is a need to know, on a continuing and rapid basis, that our food processing equipment surfaces are free of microbial contamination. One has only to pick up a copy of Food Safety Magazine, a trade publication, and review the advertising for available rapid microbial testing methodology. Most of these are based on the ATP bioluminescence principle and involve swabbing the equipment with a special swab which is then exposed to chemicals that amplify and quantitate the bioluminescence.
I believe that in the future we will be doing the same sort of swabbing of our pharmaceutical equipment immediately prior to use. This will help us in many ways. We often clean equipment and then it may become recontaminated during long idle periods of storage time. If those surfaces have become contaminated either accidentally or even intentionally (bioterrorism) would it not be a great control to be able, not only to test the equipment, but to constantly monitor the condition of these critical equipment surfaces? I believe this is at least related to the goals of the process analytical technology (PAT) FDA initiative to constantly monitor our critical processes. Again, the cleaning process must be considered, in my opinion, just as critical as the manufacturing process.
In addition to the emerging rapid testing for microbial residues, there is currently an emphasis being placed on the rapid analysis for chemical residues. The "need for speed" is the same as for airport security units that must process passengers for thousands of domestic air flights each day. The evaluation must be fast since people are lined up and waiting to board their planes and they must be sensitive in order to detect the trace levels of telltale residues. The technology used again involves swabbing of carry-ons, typically laptop computers, and evaluation of the swab by an analytical device that accomplishes the scan in microseconds. Some rapid and sensitive techniques that are currently under study include ion mobility spectrometry (IMS) and diffuse reflectance spectroscopy (DRS).
SUMMARY
The pharmaceutical industry has come a long way over the brief 50 year time span that I have worked in it, but the next 50 years truly promise to be even more exciting and more demanding. The risks are enormous, the responsibility is huge, and the stakes (the quality of our products and the health of the nation) are the highest they have ever been. Cleaning, containment, control--the 3 C's--will continue to be important in achieving and improving the quality of our products and thus the health of our people.
My best wishes for your happiness, health, and productivity in your chosen profession, my friend. Take good care
REFERENCES
1. FDA, Current Good Manufacturing Practices for Finished Pharmaceuticals, Title 21, Vol 2, Part 211.
2. FDA, "Guide to Inspection of Validation of Cleaning Processes" (Division of Field Investigations, Office of Regional Operations, Office of Regulatory Affairs) (July, 1993).
3. Preventing Occupational Exposure to Antineoplastic and other Hazardous Drugs in Health Care Settings, National Institute for Occupational Safety and Health (Sep 2004).
4. Baker, E.S and T.H. Connor, Monitoring Occupational Exposure to Cancer Chemotherapy Drugs, American Journal of Health System Pharmacy (Nov 1996).
5. Robays, M, the Pharmaceutical Journal, Vol 263, No 7070 (Nov 1999).
6. Kaijer, G.P., W.J.M. Underberg, and J.H. Berjnen, Risks of Handling Cytotoxic Drugs, Pharmacy World and Science Journal (May 1990).
7. U.S. Department of Health and Human Services, Centers for Disease Conrol, National Institute for Occupational Safety and Health, "Guidelines for Protecting the Safety and Health of Health Care Workers," Publication No. 88-119 (1988).
ABOUT THE AUTHOR
Dr. William (Bill) E. Hall has over 45 years industry experience in areas including Research and Development, Quality Assurance, and Quality Control. Bill is a member of the JVT Editorial Advisory Board and has given hundreds of presentations on the subjects of process and cleaning validation, quality assurance and compliance, as well as drug abuse. He serves as an expert witness for the FDA in the area of GMP compliance and has taught FDA inspectors in the evaluation of cleaning programs. Before entering industry, Bill worked in academia for seven years as a professor at the University of North Carolina at Chapel Hill in the School of Pharmacy. Dr. Hall can be reached by email at CleanDoct@aol.com or by phone at 910-458-1087.
Article Acronym Listing
API Active Pharmaceutical Ingredient
DRS Diffuse Reflectance Spectroscopy
ELISA Enzyme Linked Immunosorbant Assay
FDA Food and Drug Administration
GMP Good Manufacturing Practice
HPLC High Performance Liquid Chromatography
IMS Ion Mobility Spectrometry
NIR Near Infrared Spectroscopy
PAT Process Analytical Technology
QA Quality Assurance
QC Quality Control
TOC Total Organic Carbon BY WILLIAM E. HALL, PH.D.
A Personal Tribute to Ken Chapman
6 comments:
Very informative! Thanks much
Post a Comment