Aseptic Pharmaceutical Manufacturing II

Applications for the 1990s

Michael J. Groves, Ram Murty

• Publisher: CRC Press
• Year: 2020
• eBook ISBN: 9781000161670
• Topic: Negocios y empresa
• Subtopic: Manufactura

Asceptic Pharmaceutical Manufacturing II explores the sophisticated technology, developments, and applications that allow aseptic processing to approach the sterility levels achieved with terminal sterilization. Written by experts in sterile manufacturing, this book covers aseptic technology, developments, and applications and makes a valuable contribution to understanding the issues involved in aseptic manufacture. Topics include the processing of biopharmaceuticals, lyophilization, personnel training, radiopharmaceuticals, hydrogen peroxide vapor sterilization, regulatory requirements, validation, and quality systems.

1

Introduction

Michael J. Groves
Ram Murty

The exact origins of parenteral therapy are lost in antiquity, but we do know that devices recognizable today as syringes, made from pewter or brass, were found recently in the wreck of the Mary Rose, a ship of the line that sank under the eyes of an embarrassed King Henry VIII during a battle with a French fleet in 1545. Intravenous administration of opium dissolved in wine was attempted by Sir Christopher Wren—better known as the architect of St. Paul’s Cathedral in London, but, coincidentally, a mathematician and professor of surgery—in 1656. Over the next 200 years or so there were sporadic investigations of this route of administration, culminating in the intravenous administration by Latto in Edinburgh of salt solutions to patients during an outbreak of cholera. Alexander Wood, also at Edinburgh, is credited with the invention of a graduated glass syringe fitted with a hollow needle “like the sting of a wasp” in 1853. Solutions of morphine are believed to have been injected by surgeons during and after the American Civil War; the first compendial monograph for morphine injection first appeared in the Addendum to the British Pharmacopoeia (BP) of 1874. Although these early monographs suggested that injections be prepared carefully, there was no sterilization procedure and development of the technology did not proceed rapidly until the discovery of the anti-syphilitic “Salvarsan” at the turn of the present century. Had “Salvarsan” been active orally, it is doubtful if progress would have been as rapid; by the 1920s BP monographs for injections also included sterilization procedures. In the 1930s there was considerable discussion of the effectiveness of these sterilization procedures and the associated methods of testing for sterility. Over the next 20 years technology developed for the production and preparation of blood products and penicillin during World War II. This was coupled with the realization that postproduction sterility testing was unlikely to detect anything except the most badly contaminated product, a point emphasized by reports of deaths due to microbiologically contaminated products during the mid-1960s. Public inquiries on both sides of the Atlantic resulted in the promulgation of the first Good Manufacturing Practices (GMPs) by the Food and Drug Administration (FDA) in 1975. Over the past 20 years the subject has been extensively studied and developed. Today, the main issues are the requirements by the FDA that all sterile products be terminally heat sterilized unless the product is unable to withstand this treatment. Not really an issue with salt solutions, it is rather less so with sugar solutions and almost certainly not feasible with the vast majority of modern biotechnology-derived protein or peptide drugs. The need at this immediate point in time is to thoroughly discuss the problems and issues that arise from this requirement. This present book is designed to provide some insight into a number of aspects of this current situation.

HISTORICAL DEVELOPMENT OF ASEPTIC PROCESSING

Although terminal sterilization of injectable solutions was an established technical process by the 1920s, this was the period in which, arguably, the first injectable protein, insulin, became available. With very little to guide them, manufacturers very quickly discovered that it was necessary to prepare a sterile solution of the drug by filtration through porous ceramic filters, followed later by Seitz or compacted asbestos fiber filters. The solution required handling in a controlled environment with aseptic precautions and packaging in presterilized container components, such as glass vials fitted with rubber closures. Glove boxes and the use of ultraviolet germicidal lamps were explored in an effort to obtain a suitably “clean” environment; however, as testing methods developed, the limitations due to the techniques were revealed. The use of special clothing to protect the product from the workers themselves was slowly developed. Concomitantly, advances in heat sterilization technology were also being made. The Tyndalization process in which spore-forming organisms, resistant to heat below 95°C, were allowed to develop to the more sensitive vegetative forms by alternately exposing the product to heat (for killing) and incubation (for allowing the growth of survivors) over two or more cycles was shown to be less than efficient or effective. Tyndalization was, therefore, abandoned, although the process has been explored again more recently Organisms were found that were resistant to heat exposure to temperatures below 120°C; this temperature became the norm for the autoclaving process. Then it was realized that the operation of a steam autoclave was not without its own problems and methods were devised for recording temperatures inside the autoclave and, eventually, inside samples of the product. The effect of air mixed with steam on the temperature of what should have been saturated steam had been known from work during the previous century on the construction and operation of railway engine boilers. The realization that a poorly operated autoclave, from which not all the air had been removed, could allow the product to be incorrectly considered to be “sterile” was a painful and, in some cases fatal, process. This must be considered unfortunate since a generation earlier, inadequately heat-treated dressings had resulted from the application of an insufficient heat treatment due to a failure to remove all the air prior to steaming. Application of the same temperature-measuring techniques to air ovens produced similar findings about the inadequacy of operation procedures and, again, resulted in improved and reproducible production techniques. The resultant effect of all of these observations was to move away from mere “cookery book” procedures to more intelligent applications of knowledge—essentially the first realization that validation of procedures was a necessary and appropriate way to go.

The sterilization process eventually became more controlled with the application of scientific measurements based on the response of selected, heat-resistant, microorganisms. The microbial death process was considered to be a kinetic process, requiring time to go to completion. Calculations were made of D, the decimal reduction time, or the thermal destruction value, Z, and the F₀ or number of minutes at a specified temperature above, say, 100°C, equivalent to holding the product at 121°C (250°F) for 15 minutes. At the same time it became evident that the success of a sterilization process, by definition killing all microorganisms in a product inside a sealed container, would critically depend on the bioburden or the number of organisms originally present inside the container before sterilization. This led to an exploration of the effectiveness of filtration procedures for the removal of microorganisms, dead or alive. In the 1970s a membrane filter with an average pore diameter of 0.45 μm was considered to be suitable for the purpose of removing bacteria; later this limit was reduced to a pore diameter of 0.2 μm. Membrane manufacturers also had problems in characterizing their products. For example, ideally the nominal diameter of 0.2 μm should be that of the largest pore, not the average pore size. Many of the arguments and discussions at that time were only resolved by refined characterization and measurement techniques. Nevertheless, although particles larger than 0.2 μm diameter and, as it happens, significant quantities of particles smaller than this diameter, will be removed on passage through a membrane of this nominal limiting pore size, particles corresponding to viruses and viable cell fragments could, theoretically, pass through into the filtrate. This consideration alone represents the biggest limitation to the production of any product by an aseptic manufacturing process. Nevertheless, filter manufacturers now routinely challenge their products with suitable suspensions of small microorganisms in order to support their claims of effective sterilization. These claims, perhaps wisely, have not been extended to suggesting that viruses are removed from the product. The unfortunate realization that viruses, such as HIV or hepatitis, were left in blood products and could subsequently infect patients has stimulated research in this area; thus far an effective filtration technology has not been developed.

As noted earlier, contamination of a product with viable microorganisms, either as a result of faulty processing or, in the case of aseptic filling, by the intrinsic nature of the process, is unlikely to be detected by postproduction sampling and sterility testing. The concept of sterility assurance levels (SALs) has arisen over the past decade and numbers provided. For example, although there is really no method to accurately and precisely measure SALs, a terminal heat sterilization process is considered to provide a SAL of 10⁻⁶, indicating that there is a 1 in 1,000,000 chance of finding a contaminated container in a batch of product. Batch sizes of 1,000,000 units are not common but, with modern production methods, need not necessarily be impossible. A SAL of 10⁻⁶ would, therefore, suggest that there would be one contaminated container in that batch. This is, of course, unacceptable to the patient who receives that one container. What the concept should mean is that there is a 1:1,000,000 chance of finding a contaminated container each time the batch is sampled, irrespective of the number of containers waiting to be sampled. By the same convention aseptic processes are considered, without, one might add, a lot of evidence, to provide a SAL of 10⁻³. Thus, in a 3000 lot assembled aseptically, 3 of the containers would be contaminated. This would not be true if, in fact, the chance of contamination at each sampling point was considered to be 1:1000. The FDA considered this to be unacceptable and this is the main reason why terminal heat sterilization of the product was suggested at one point as being mandatory.

The issue is not insignificant because approximately two-thirds of sterile pharmaceutical products are filled aseptically, with up to 87 percent of small-volume parenterals (SVPs) (<100 ml) being filled in the same way in the United States [data from a Parenteral Drug Association (PDA) survey by Agalloco and Akers in 1993]. Indeed, it was suggested that, as more biotechnology-derived therapeutic drugs enter the marketplace, more products will need to be processed aseptically. The FDA requirement is, therefore, inopportune; it will undoubtedly stimulate deeper research into, and understanding of, the topic.

THIS PRESENT BOOK

It should be noted that Aseptic Pharmaceutical Processing, edited by W. P. Olson and M. J. Groves and published in 1987, contained a number of useful chapters that could, with benefit, be read before turning to this second edition, which is effectively an updated and expanded review of the subject.

The older glove boxes, noted earlier, have made a reappearance under the rubric of “barrier technology.” Discussed at length in the Olson and Groves book, the subject has been developed here by J. P. Lysfjord and his colleagues in the chapter on isolators and filling lines. Lyophilization and barrier systems are also discussed by J. W. Snowman. Both of these chapters are important because, even if a sterile solution could be placed into a sterilized container, for a few moments in time, the product is exposed to the environment during filling and subsequent drying. The use of barrier technology today is intimately connected with the success or otherwise of a sterilizing gas used to sterilize all working surfaces inside the barrier. The most promising sterilizing gas in use currently is hydrogen peroxide vapor, a subject discussed by L. M. Edwards and R. W. Childers.

The ultimate method for minimizing this environmental exposure is the filling procedure described by Leo in Olson and Groves known as form-fill-and-seal. In this machine a container of plastic is actually blown and formed in a clean environment immediately prior to filling it with a sterile solution. The solution also serves to cool the extruded plastic and the final container is sealed immediately after filling. The actual environmental exposure is, therefore, reduced to milliseconds. More to the point, the whole apparatus is automated and enclosed in a controlled environment that also has the effect of minimizing contaminant exposure. The chance of contaminated product coming from a form-fill-and-seal machine is, therefore, very small; indeed, the problem has been the perennial difficulty of how to validate any claims made for the process. Some progress toward providing a solution to this exceptionally difficult issue has been made by filling many thousands of containers with sterile bacterial growth medium and determining how many of these sealed containers demonstrated growth on subsequent incubation. Issues associated with this process are described by C. S. Sinclair and A. Tallentire in their chapter. These authors have deliberately contaminated the controlled environment surrounding the machine in order to simulate a worst-case situation. Even in this case their data demonstrate a high degree of protection of the filled product and it is likely that, under favorable conditions of operation, a form-fill-and-seal machine could produce a product aseptically, while at the same time providing a SAL of approximately 10⁻⁶. This subject is obviously ongoing, but future developments suggest that automation of part or all of the production process in a controlled environment is what is needed to produce an aseptically filled product in accordance with modern regulatory requirements.

Validation is required, however, for each and every aspect of the process. From a scientific perspective, this offers the biggest challenge because it is often very difficult to prove, with appropriate scientific and logical conviction, that a specific claim is valid. This topic is discussd by V. Kumar and R. Murty; sight should not be lost of the fact that adequate validation of any process is difficult, requiring a sound knowledge of the basic sciences and an intimate knowledge of the purpose and procedure of the process under review. Unfortunately, this subject has often been obscured and blurred in recent years so that the actual purpose is sometimes lost in misplaced rhetoric. This is another regulatory issue that will only become acceptable from the broader perspective of time.

It will be evident that the FDA requirement to convert aseptic processes to terminal sterilization, discussed at a FDA conference in October 1993, is not a local issue; R. Dabbah discusses the international perspective of this topic. Dabbah, a staff member of the United States Pharmacopeial Convention, is well positioned to interpret the international scene; it should be noted that the current USP 23 contains a general chapter on sterile drug products for home use, <1206>, and a chapter on sterilization and sterility assurance (of compendial articles), <1211>. It is, therefore, an invaluable resource used by the FDA with the backing of law. This latter chapter, incidentally, defines the SAL as a “microbial survivor probability” so that a 10⁻⁶ probability would suggest that there is an assurance of less than one chance in one million that viable microorganisms are present in the sterilized article. “Survival” and “removal” have quite separate meanings and the use of expressions such as SAL or microbial survivor probability ignore the likelihood that a single surviving organism would proliferate after exposure to a heat sterilization process or, in the case of a filtration process, would be left in the product to proliferate. In the case of heat sterilization, SALs of 10⁻¹² are often sought in what is, obviously, an “overkill.” However, many products do not survive these excessive exposures to heat. Filtration processes are now being judged by the number of logarithmic reductions of exposure that can be accomplished. For example, if a filter fails to allow any organisms to pass through when exposed to a test solution containing 10⁷ organisms/cm², then the log reduction value (LRV) is 7. This emphasizes the point made earlier that the cleaner the solution (i.e., the lower the bioburden) before the final sterilization process, the greater is the likelihood that the process will be successful. In the case of a terminal filtration process through, say, a 0.2 μm nominal pore size membrane, it is extremely unlikely that only one filter-collecting surface would be used; assemblies of filters are more likely to be used in current practice. Because of the thinness of membrane filters, the supporting surfaces required careful and accurate machining. Since stainless steels must be used, the high cost of the supporting filtration equipment has tended to inhibit the use of multiple or redundant filtration units. However, realization that the cost of modem biotechnology-produced products is also high has tended to change the perspective on what has become an important capital item in the production plant.

Crucial issues are those of personnel and, especially, personnel training. There is now a realization that the product should have minimal exposure to “people” since people represent the major source of microbiological contamination. However, even in a highly automated production line, people are needed at key points in the process. M. J. Akers deals with the training of personnel, emphasizing a vital need for workers in an aseptic process to understand the purpose of their work and the need to develop the appropriate skills necessary for aseptic product assembly. This subject leads naturally to the laboratory techniques associated with aseptic processing, discussed by R. Mehta and R. Murty. A rather more specialized involvement on a laboratory scale, because of some unique issues, is reviewed in the chapter on radiopharmaceuticals by P. O. Bremer. Here the issues are scale and the fact that the shelf-life of a radioactive diagnostic agent is often too brief to allow testing of the product prior to administration to the patient. Procedures, therefore, have to be in place that guarantee safety and efficacy, allowing validation “after the fact.”

The problems of handling modern biopharmaceutical drugs have been alluded to throughout this discussion. In a chapter by N. M. Lugo, the issues are reviewed and placed into perspective since, after all, insulin, a complex protein drug, has been on the market for nearly 70 years.

This volume contains rather less information on the technical issues that are involved in aseptic production of the sterile product. Nevertheless, there have been some. significant advances in lyophilization over the past five or so years and the subject is updated in the review by E. Trappier. Clearly this has come about through a deeper understanding of the subject brought about by academic studies allied with the involvement of many equipment manufacturers. This has been to the net benefit of the producer and, ultimately, to the product consumer.

This general need to improve “quality” from the consumer’s perspective has not been confined to aseptically produced sterile products in just one industry; it is a worldwide movement seen in many, if not most, industries at this time. For the most part, this is associated with the ISO 9000 initiative; this movement is reviewed by K. Stephens from a GMPs point of view. This chapter is also valuable for the overall perspective it provides of the movement toward ISO 9000 certification. The author has introduced an element of demystification about the process and summarizes the overall benefit it provides for the producer and, ultimately, the consumer. However, the ISO process itself is, by the very nature of international collaboration, a very complex and involved business. Some feeling for this is conveyed in the chapter by M. Korczynski and his colleagues, who have been involved in the certification process for aseptic manufacturers in the United States. Finally, we have included a review of packaging and labels by R. Murty, an area of much development in recent years that will continue to provide a challenge into the future.

CONCLUSIONS

The general theme of this introduction has been the way stepwise technical progress has occurred since the very earliest attempts to inject solutions into the human body. Generally, attention has been focused on one or two limited aspects of the overall subject. ...

Table of contents

• Cover
• Title Page
• Copyright Page
• Table of Contents
• Preface
• Acknowledgements
• 1. Introduction
• 2. Controlled Environments in the Pharmaceutical and Medical Products Industry: A Global View from Regulatory, Compendial, and Industrial Perspectives
• 3. Quality Systems and Total Quality
• 4. Aseptic Processing of Healthcare Products—A Pending ISO Document
• 5. Validation of Aseptic Processes
• 6. Laboratory Techniques in Aseptic Manufacturing
• 7. Aseptic Production of Radiopharmaceuticals
• 8. Good Aseptic Practices: Education and Training of Personnel Involved in Aseptic Processing
• 9. Predictive Sterility Assurance for Aseptic Processing
• 10. Aseptic Processing of Biopharmaceuticals
• 11. Lyophilization
• 12. Lyophilization Under Barrier Technology
• 14. Barrier Isolation Technology: A Systems Approach
• 15. Hydrogen Peroxide Vapor Sterilization: Applications in the Production Environment
• Appendix: USP 23 Chapter <1116> : Microbiological Evaluation of Clean Rooms and Other Controlled Environments
• Glossary
• Index