`
`VOLUME 2: FACILITY DESIGN, STERILIZATION AND PROCESSING
`
`62. McCullough KZ, Novitsky TJ. Detoxification of endotoxin by acid and base. In Depyrogenation.
`Technical Report No. 7. Philadelphia: Parenteral Drug Association, Inc, 1985:78 83.
`63. Tripodi D, Nowotny A. Relation of structure to function in bacterial O antigens: V. Nature of
`active sites in endotoxic lipopolysaccharides of Serratia marcescens. Ann NY Acad Sci 1966; 133:
`604 621.
`64. Neter E, Westphal 0, Luderitz 0, et al. Studies of enterobacterial lipopolysaccharides. Effects of heat
`and chemicals on erythrocyte modifying, antigenic, toxic and pyrogenic properties. J Immunol 1956;
`76:377 385.
`65. Niwa M, Milner KC, Ribi E, et al. Alteration of physical, chemical and biological properties of
`endotoxin by treatment with mild alkali. J Bacteriol 1969; 97:1069 1077.
`66. Gould MJ, Novitsky TJ. Depyrogenation by hydrogen peroxide. Technical Report No. 7:
`Depyrogenation. Philadelphia, PA: Parenteral Drug Association, 1985:84 92.
`67. Nebel C, Nebel T. Ozone, the process water sterilant. Pharm Manuf 1984; 1:16 23.
`68. Lee MG, Hunt PB, Vallor J. The rate of endotoxin destruction during water treatment using a
`combination of ozone and ultraviolet radiation. J Parenter Sci Technol 1991; 45(4):183 186.
`69. Tsuji K, Harrison SJ. Limulus amebocyte lysate a means to monitor inactivation of lipopolysaccharide.
`In: Cohen E, ed. Biomedical Applications of the Horseshoe Crab (Limulidae), Progress in Clinical and
`Biological Research, Vol 29. New York: Alan R. Liss, Inc., 1979:367 378.
`70. Hudson CT, Nase R. Inactivation of endotoxin by ethylene oxide and cobalt60 irradiation. Technical
`Report No. 7: Depyrogenation. Philadelphia, PA: Parenteral Drug Association, 1985:113 116.
`71. Reich RR, Anderson HD, Syracuse KA. The possible effect of EtO on the LAL reaction. Med Device
`Diagn Ind 1984; 6:36 39.
`72. Banks HM. A study of hyperpyrexia following intravenous therapy. Am J Clin Pathol 1934; 4:260.
`73. Bamba T, Matsui R, Watabe K, et al. Enhancing effect of non ionic surfactant on the inactivation of
`lipopolysaccharide by steam heat treatment II. J Parenter Sci Technol 1997; 51(4):156 160.
`74. Tsuji K, Harrison SJ. Dry heat destruction of lipopolysaccharide: dry heat destruction kinetics. Appl
`Environ Microbiol 1978; 36(5):710 714.
`75. Hecker W, Witthauer D, Staerk A. Validation of heat inactivation of bacterial endotoxins. J Parenter
`Sci Technol 1994; 48:197 204.
`76. Avis KE, Jewell RC, Ludwig JD. Studies on the thermal destruction of Escherichia coli endotoxin.
`J Parenter Sci Technol 1987; 41(2):49 56.
`77. Endotoxin Indicator for Depyrogenation. USP 33: R478. United States Pharmacopeial Convention,
`Rockville, MD, 2010.
`78. LAL Users Group. Preparation and use of endotoxin indicators for depyrogenation process studies.
`J Parenter Sci Technol 1989; 43(3):109 112.
`
`Regeneron Exhibit 1016.201
`
`
`
`The compendial sterility tests
`
`Scott V. W. Sutton
`
`BACKGROUND
`The compendia! sterility test is frequently presented as a flawed test for its stated purpose. This
`statement, of course, begs the question as to what exactly is the purpose of the sterility test as
`described in the compendia? The test first appeared in 1932 (1) and included the basic features of
`two media, prescribed dilution scheme (for bacteriostasis/ fungistasis or method
`the modern test
`suitability) and a defined incubation time. The original test differed from the contemporary
`method in that it had the media incubated for five days rather than 14 and allowed two retests (all
`three had to fail to fail the test). However, the basic stmcture of the test is present.
`This test has generated controversy as to its role in product quality testing for decades. Part
`of the problem is in understanding the role of the compendia! tests. Those chapters in USP
`numbered less than 1000 (for example, the Sterility Test is USP chapter <71>) are referee tests
`in
`other words they are in place solely to demonstrate conformance to qualities specified in the
`product monograph as described in the current National Formulary (the other part of the book). A
`rigid interpretation would have it that if the product is not described by NP monograph, the test
`does not directly apply. In fact, the preface to the internationally harmonized sterility tests reads:
`
`The following procedures are applicable for determining whether a Pharmacopeial article
`purporting to be sterile complies with the requirements set forth in the individual monograph
`with respect to the test for sterility.
`
`In a similar vein, sterile finished dosage forms have the following requirement in USP
`(from <1> Injections):
`
`"Sterility tests: Preparations for injection meet the requirements under Sterility Tests <71 >"
`
`the test states that it is applicable for meeting the require(cid:173)
`This has a nice symmetry
`ments set forth in the monograph, the requirement being that the material meets the
`requirements of the test.
`So, one would have to conclude that the test is not flawed for its intended purpose, that
`purpose being to show that the material tested meets the requirements of the test. How did we
`come to think that this test was designed to show the sterility of the product?
`We need something to demonstrate product sterility. 21 CFR 211 states the requirement:
`
`"211.167 Special testing requirements.
`(a) For each batch of drug product purporting to be sterile and/ or pyrogen free, there shall be
`appropriate laboratory testing to determine conformance to such requirements. The test
`procedures shall be in writing and shall be followed."
`
`The difficulty, of course, is that there is really no way, given current technology, to
`demonstrate sterility of a batch. This imposes significant validation issues as the most direct
`and persuasive documentation of product sterility.
`However, there is an expectation in the GMP that a sterile finished product will have a
`release test. How are we to determine a suitable, "validated" release test for a characteristic
`that cannot be measured? A way to satisfy this requirement is provided in:
`
`"211.194 Laboratory records.
`(a) Laboratory records shall include complete data derived from all tests necessary to assure
`compliance with established specifications and standards, including examinations and assays,
`as follows: ...
`(2) A statement of each method used in the testing of the sample. The statement shall indicate
`the location of data that establish that the methods used in the testing of the sample meet
`proper standards of accuracy and reliability as applied to the product tested. (If the method
`employed is in the current revision of the United States Pharmacopeia, National Formulary,
`AOAC INTERNATIONAL, Book of Methods,(1) or in other recognized standard references, or
`
`Regeneron Exhibit 1016.202
`
`
`
`188
`
`VOLUME 2: FACILITY DESIGN, STERILIZATION AND PROCESSING
`
`is detailed in an approved new drug application and the referenced method is not modified, a
`statement indicating the method and reference will suffice). The suitability of all testing
`methods used shall be verified under actual conditions of use."
`
`So if we can cite a "validated" test we do not need to develop one ourselves. Thus, the
`internationally harmonized Sterility Test is pressed into service as a product quality test, even
`though that is not its design nor its purpose.
`
`THE STERILITY TESTS
`There are two different GMPs describing sterility in the United States. The first is 21 CFR 211
`and the second is the "Biologics" 21 CFR 610 and 612. By common consensus, the 21 CFR 211
`cGMP looks to the compendial sterility tests, while 21 CFR 610 describes a separate test in 21
`CFR 610.12. The Biologics test is similar in fundamental aspects to the compendial sterility
`tests. There is a finite (and small) sample size and two recovery media are used, each with
`specified incubation conditions. So both types (compendial and Biologics) share some common
`limitations (see the following text).
`The compendial sterility tests describe two separate types of tests, the membrane
`filtration and the direct transfer methods. In the first, solution from a specified number of
`containers (volume and number determined by batch size and unit fill volume) is filtered
`through a filter of nominal pore size 0.45 um. Recovery of viable cells from the filter(s) is
`performed by submerging the filter in one of two recovery media followed by incubation as
`specified temperatures for 14 days. The second test is a direct immersion of the product or
`suspensions into a suitable volume of the two media to allow growth. The media are designed
`to support growth in aerobic, or growth in an environment of limited oxygen availability. Both
`types of tests require demonstration that the specific method used is suitable for that product.
`As early as 1956 Bryce published an article describing the two critical limitations of this
`test. He put forward that the test was limited in that it can only recognize organisms able to
`grow under the conditions of the test, and that the sample size is so restricted that it provides
`only a gross estimate of the state of "sterility" of the product lot (2). Other concerns about the
`Sterility Test (e.g., choice of sample size, choice of media, time and temperature of incubation)
`were extensively reviewed in an article by Bowman (3).
`There have been several changes in the compendial Sterility Test since that time,
`culminating in the internationally harmonized test (4). However, the two basic problems
`outlined in 1956 by Bryce remain today.
`
`Limitations to the Sterility Tests
`Sample Size
`The sample size is set arbitrarily and does not provide a statistically significant population to
`estimate sterility (5). This is indisputable and unavoidable with a test of this type, which is
`destructive in nature. Let's look at some of the numbers:
`
`Let the likelihood of a contaminated unit
`A
`By the Poisson distribution, the probability of picking a sterile unit from the fill (denoted P) is
`e ", or 2.7182818 J.
`Then, if you are picking 20 samples from an infinite supply (or for this discussion, from a
`pharmaceutical batch),The probability of passing the sterility test is F2°
`Conversely, the probability of failing the sterility test is 1 F2°
`Therefore, given a known frequency of contaminated units in the batch:
`
`Frequency of
`contaminated units
`in the batch
`
`Probability of failing
`sterility test with the
`current sample size
`
`0.001
`0.005
`0.01
`0.05
`0.1
`0.5
`
`0.0198 2%
`0.0952 9.5%
`0.1813 18%
`0.6321 63.2%
`0.8647 86.5%
`1.0000 100%
`
`Regeneron Exhibit 1016.203
`
`
`
`THE COMPENDIAL STERILITY TESTS
`
`189
`
`The only way to modify this limitation would be to degrade the media (resulting in lesser
`recovery and therefore false negatives) or to increase the number of samples. Changes of this
`sort seem unlikely in the compendial sterility tests at this point in time. A discussion of
`different sampling plans that might be used is presented in Bryce (2), and a more full
`discussion of the controversy over the final resolution of the current procedure is provided in
`Bowman (3). After extensive review, all of the proposed sampling plans were found wanting
`for one reason or another.
`One frequently overlooked aspect of discussions of sampling plans is that the statistical
`analyses all assume that the test system would recover even a single microorganism if it were
`present in the sample. In other words, one contaminating cell would result in media turbidity.
`This (unverified and unlikely) assumption leads us to the next topic.
`
`Recovery Conditions
`The harmonized test utilizes Trypticase Soy Casein Digest Broth and Fluid Thyioglycollate
`Medium. These media and their corresponding incubation temperatures were chosen to
`maximize recovery of potential contaminants early in the development of the tests. However,
`some authors have questioned the choice of media (6), while others have suggested the use of
`solid media rather than liquid media would be appropriate (7). The choices in the current
`harmonized procedure reflect those media to which all parties in the harmonization process
`could agree.
`Then there was a concern about incubation duration. USP 23 (8) allowed a 7-day
`incubation period for products tested by membrane filtration; 14 days for those tested by the
`direct transfer method. This requirement changed in USP 24 (9) to include a 14-day incubation
`period for both types of tests with the exception of products sterilized by terminal sterilization
`(this exception was removed by USP 27 (10)). Similarly, the Phann Eur 3rd Edition (1997)
`allowed a 7-day incubation period (unless mandated by local authorities). This allowance was
`amended in 1998 with the 4th edition to 14-day incubation. This extension was the result of
`concerns that the methodology might not be able to detect "slow-growing" microorganisms.
`The incubation period was identified as a concern by Ernst et al. (11) who recommended
`a longer period of incubation time than 7 days might be necessary, perhaps as long as 30 days.
`More recently this position was repeated with retrospective data provided by German and
`Australian workers who wished to ensure that a harmonized procedure included an
`incubation period of at least 14 days (12,13).
`However, even with the longer incubation period there is no assurance that all
`microorganisms can grow under these conditions, but are metabolically active. In fact a growing
`body of evidence suggests that there are a large number of microorganisms that are unable to
`replicate under standard laboratory conditions (viable but not culturable VBNC) (14 16).
`
`CLARIFICATIONS AND ENHANCEMENTS TO THE HARMONIZED
`STERILITY TEST
`There have been quite a few clarifications offered by different regulatory agencies to the
`compendial sterility tests. This section will not be a review of the genesis of the sterility tests;
`that discussion is outside the scope of this chapter. We will, however, take a look at a few of
`the clarifications offered by different regulatory agencies on the implementation of the
`harmonized test.
`
`US FDA/CBER
`US FDA/CBER (the Center for Biologics Evaluation and Research) has a section of the GMP
`under section 21 CFR 610. In this section, 610.12 describes a separate sterility test to be used
`with those products under CBER purview. There are several differences in the test from the
`internationally harmonized tests that include controls, method suitability requirements, media
`growth promotion procedures, etc. A major difference between the tests is that the CBER test
`allows a retest if the original sterility test fails. This retest must also fail for the product lot to be
`out of specification. While the manufacturer is urged not to attempt this approach by the
`author of this chapter, this is still technically allowed in the Biologics sterility test.
`
`Regeneron Exhibit 1016.204
`
`
`
`190
`
`VOLUME 2: FACILITY DESIGN, STERILIZATION AND PROCESSING
`
`As an aside, the pharmacopeias and 21 CFR 610.12 do not reference or provide sterility
`guidelines for unprocessed bulk samples for protein and virus products, although the FDA
`guidance documents "Points to Consider in the Manufacture and Testing of Monoclonal
`Antibody Products for Human Use" (17) and "Points to Consider in the Characterization of
`Cell lines Used to Produce Biologicals" (18) require this testing. Common practice is to use
`10 mL/media (for a total of 20 mL) for this testing.
`
`USP
`The USP introduced clarification in 2007 with a new chapter <1208> "Sterility Testing
`Validation of Isolator Systems" (19). This informational chapter provides background in
`isolator design and construction, the equipment qualification considerations for the isolator,
`validation of the decontamination cycle (this would include the internal environment, the
`exterior of the product containers entering for testing and the protection of the product from
`the decontamination cycle), and the maintenance of asepsis within the isolator environment.
`The reader is also instructed that the sterility test performed in a properly functioning isolator
`is very unlikely to result in a false-positive result. Finally, instruction is provided on the
`training and safety aspects of the isolator operation.
`
`Pharm Eur
`The European Pharmacopeia have published a nonmandatory chapter "5.1.9 Guidelines for
`Using the Test for Sterility" (20) in which further information on the sterility tests is provided.
`The user is instructed that the test can be performed in a class A laminar air flow cabinet
`located in a class B room, or an isolator. The reader is also reminded that this test cannot
`demonstrate sterility of a batch, and that it is the manufacturer's responsibility to adopt a
`representative sampling plan. Finally, elaboration is provided on "Observation and Interpre(cid:173)
`tation of Results" in that during an investigation,
`
`" ... if a manufacturer wishes to use condition (d) as the sole criterion for invalidating a sterility
`test, it may be necessary to employ sensitive typing techniques to demonstrate that a
`microorganism isolated from the product test is identical to a microorganism isolated from the
`test materials and/or the testing environment. While routine microbiological/biochemical
`identification techniques can demonstrate that 2 isolates are not identical, these methods may
`not be sufficiently sensitive or reliable enough to provide unequivocal evidence that 2 isolates
`are from the same source. More sensitive tests, for example, molecular typing with RNA/DNA
`homology, may be necessary to determine that microorganisms are clonally related and have a
`con1n1on origin."
`
`TGA
`The Australian Therapeutic Goods Administration (TGA) has published a 33-page document
`entitled TGA Guidelines 011 Sterility Testing of Therapeutic Goods (21) to explain how the
`harmonized sterility tests are to be interpreted when submitting a product into Australia while
`noting that the British Pharmacopeia (and therefore Phann Eur) is the official test. This document
`is extensive and expands the details provided on controls recommended in the harmonized
`Sterility Test.
`The Stasis Test is an additional control recommended here. In this test, spent media from
`a negative Sterility Test (media that has seen the membrane that filtered product and 14 days
`of incubation) is subjected to an additional growth promotion test to demonstrate its
`continuing nutritive properties.
`There is also a great deal of discussion in this document on the interpretation of the test
`results and on how to investigate Sterility Test failures (see below).
`
`PIC/S
`The Pharmaceutical Inspection Convention and Pharmaceutical Inspection Co-operation
`Scheme (jointly referred to as PIC/S) has as its mission, "... to lead the international
`development, implementation and maintenance of harmonized Good Manufacturing Practice
`
`Regeneron Exhibit 1016.205
`
`
`
`THE COMPENDIAL STERILITY TESTS
`
`191
`
`(GMP) standards and quality systems of inspectorates in the field of medicinal products."
`There are currently 37 Participating Authorities in PIC/S (as of October 2009
`see http://
`www.picsscheme.org for current information). The US FDA has applied for membership
`several years ago and awaits disposition of its application (22).
`
`PI 012 2 "Recommendations on Sterility Testing"
`PI 012-2 "Recommendations on Sterility Testing" provides a great deal of additional information
`that the inspectors are instructed to ask about. This includes direction on acceptable training of
`personnel, the sterility test facilities (including clean room design, airlocks, aseptic gowning, and
`clean room fittings), cleaning and sanitization, as well as environmental monitoring of the
`sterility test area. Additional detail is also provided on the test method.
`The Sterility Test controls are also provided some attention in this document. In addition
`to their execution, the inspector is instructed to require a table of negative control failures and
`positive control failures.
`The instruction provided for "validation" (or bacteriostasis/fungistasis) by PIC/S in this
`document is in conflict with the harmonized chapter. Where the harmonized chapter informs
`the user to add the inoculum to the final rinse, the PIC/S document states that the product
`should be inoculated unless it is not practical due to product interference (such interference,
`presumably, would have to be documented). In addition, the PIC/S document asserts that it is
`good pharmaceutical practice to revalidate all products every 12 months. The author is
`unaware of this practice outside this document for the pharmaceutical industry. The Stasis Test
`is also recommended in the PIC/S document. This test is also recommended to be repeated at
`least every 12 months.
`Finally, there is a good deal of discussion on investigations (as in the TGA guidance).
`This will be discussed below.
`
`PI 014 3 "Recommendation: Isolators Used for Aseptic Processing and Sterility Testing"
`This guidance document covers the same basic material as described in the preceding text for
`USP chapter <1208> with some significant expansion on validation considerations, the nature
`of the sporicidal decontaminant, and the logistics of the isolator's operation. While this
`guidance is directed primarily to the use of isolators in manufacturing, it also claims sterility
`testing to be within its scope.
`
`RMM AND THE STERILITY TESTS
`A frequently discussed option for the sterility testing of finished dosage forms is to use a "rapid"
`method (23). Currently marketed rapid microbiological methods (RMM) can be grouped into two
`types
`those that require amplification (growth) to show low-level contamination and those that
`do not. In the first group would be technologies such as ATP bioluminescence, head-space
`analysis, and others. Examples of the second type might be technologies such as PCR and vital
`dye/ chromatography methods. Why is this distinction important?
`The concern with recovery conditions is that we do not know how to grow all micro(cid:173)
`organisms that might contaminate pharmaceutical products. Applying an alternate technology
`that requires growth does not result in an improvement in the sterility test method, since
`organisms that currently do not grow would not grow in the new method either (24). In
`addition, there is the continuing concern about the duration of the incubation period.
`The currently required 14-day incubation period imposes a significant burden on the
`manufacturer who must quarantine product until successful completion of the test. Can this
`be shortened in an alternate test? The time required for microbial growth to turbidity can be
`thought of as the sum of two stages: a lag phase where the microorganism prepares to grow
`and the generation time requirements for a low level of microorganisms to ~ow to a
`concentration where they are visible using human vision, that is, approximately 10 cfu/mL.
`This separation of stages is important, as it seems that the lag phase is the most significant
`portion of time required for turbidity (25). Therefore, any alternate methodology that requires
`growth to amplify the microorganism will likely be required to incorporate a lengthy
`incubation period to ensure the recovery of "slow-growing" microorganisms.
`
`Regeneron Exhibit 1016.206
`
`
`
`192
`
`VOLUME 2: FACILITY DESIGN, STERILIZATION AND PROCESSING
`
`Duguid and du Moulin (26) describe one approach to overcoming this issue. Using an
`amplification stage for an ATP bioluminescence technology, they started in 1999 to validate a
`sterility test for an autologous cell therapy product. This sterility test, which provided for
`product release in 72 hours with confirmatory results at the standard 14 days, was approved
`by FDA/CBER in 2004. In the time since they report almost 6000 sterility test results (samples
`included primary, expansion, and final product from this process) were collected including
`four positives detected. The alternate method detected them, on average, approximately 35
`hours earlier than the confirmatory test (19 vs. 54 hours incubation).
`Interestingly, US FDA/CBER (the Biologics group) has issued a draft guidance document
`on the validation of growth-based rapid methods for use in sterility testing (27). This CBER
`document is remarkable in its complete avoidance of any mention or consideration of the
`previous work done in validation of RMM by FDA/CDER, Pharm Eur, USP, or PDA.
`The limiting aspects of growth-based methods as an alternative for the sterility test
`can be avoided by use of a rapid microbiological method (RMM) technique that does
`not require growth (24). The use of a method that avoids growth requirements offers an
`additional advantage in that the question of VBNC organisms is completely side-stepped.
`As no culturing is required, the recovery phase of the sterility tests can be optimized to
`all microorganisms regardless of growth requirements. This approach is described by
`Gressett et al. (28).
`
`INVESTIGATIONS IN THE STERILITY TEST
`There is a significant amount of literature written on OOS and investigations. Most of this
`concern, of course, stems from the 1993 Barr Decision (29). Barr Laboratories had a history of
`repeated current good manufacturing practice (cGMP) deficiencies, including repeated
`retesting and resampling of product as well as reprocessing of defective product without
`adequate justification in a practice that has come to be known as "testing to compliance." This
`is not good practice
`the out-of-specification (OOS) data is telling the manufachirer important
`information about the product and must be resolved. Unfortunately for the microbiology
`community, this initial situation, as well as most of the subsequent writing on this topic, has
`focused on OOS from an analytical chemistry perspective. The Food and Drug Administration
`(FDA) has provided guidance following the Barr decision, and drafted the "Guidance for
`Industry
`Investigating Out of Specification (OOS) Test Results for Pharmaceutical Produc(cid:173)
`tion" (30). Interestingly, this guidance document only briefly touches upon microbiological
`data, stating that "the USP prefers the use of averages because of the innate variability of the
`biological test system." In addition, this guidance document specifically excludes microbiology
`from its scope in footnote 3.
`A PDA task force that was assembled to look into this issue recommended the use of the
`phrase "Microbial Data Deviation" (MDD) in the investigation of issues in microbiology, at
`least until it is clear that the issue is a true product specification failure, as opposed to a lab
`error or process monitoring concern (reviewed in Ref. 31).
`The harmonized Sterility Tests provide some guidance on MDD investigations:
`
`If evidence of microbial growth is found, the product to be examined does not comply with the
`test for sterility, unless it can be clearly demonstrated that the test was invalid for causes
`unrelated to the product to be examined. The test may be considered invalid only if one or more
`of the following conditions are fulfilled:
`
`a. The data of the microbiological monitoring of the sterility testing facility show a fault.
`b. A review of the testing procedure used during the test in question reveals a fault.
`c. Microbial growth is found in the negative controls.
`d. After determination of the identity of the microorganisms isolated from the test, the growth
`of this species (or these species) may be ascribed unequivocally to faults with respect to
`the material and or the technique used in conducting the sterility test procedure.
`
`If the test is declared to be invalid, it is repeated with the same number of units as in the original
`test. If no evidence of microbial growth is found in the repeat test, the product examined
`complies with the test for sterility."
`
`Regeneron Exhibit 1016.207
`
`
`
`THE COMPENDIAL STERILITY TESTS
`
`193
`
`Conditions "a" and "b" basically refer to a catastrophic failure of control. If it can be
`demonstrated that either the technique or the environment was not in control at the time of the
`test, the test can be declared invalid.
`Condition "c" is interesting in its own right. The assumption when running a control is
`that the effort to run that control is justified by the information provided by the test. However,
`many labs will only consider the results from the negative control if the test fails. In other
`words, although the negative control is supposed to demonstrate the adequacy of the test
`conditions and performance, if the test samples pass, then a failing negative control is ignored.
`If the test samples fail, a failing negative control is used to invalidate the test. The author of this
`chapter urges that a consistent interpretation of controls be used in all testing.
`Condition "d" is one that has received a great deal of attention. Additional detail is
`provided the previously cited Pharm Eur 5.1.6, the PIC/S guidance on sterility test, and the
`TGA document. This topic is also discussed in FDA's Aseptic Manufacturing Guide (32).
`Reduced to its essentials, the user is urged in these documents to use methods sensitive
`enough to demonstrate that the microorganism is not only of the same species, but also of the
`same strain or substrain of that species. It should be noted that even with this detail the best
`that can be done is to show a correlation between the presence of the strain from the two
`sources rather than a causal relationship. In other words, finding the same strain of
`Staphylococcus aureus on the testing technician and in the sterility test does not prove that the
`only possible source of that was the technician (the strain could also be present in the aseptic
`core), but it is accepted as sufficient proof in regulatory guidance that the test was
`compromised and so invalid.
`The pharmaceutical literature provides some examples of Sterility test investigations that
`can be used as guides. Lee (33) described a detailed sterility investigation that included the
`identification of the contaminant, reviews of documents, training records, gowning practices,
`environmental monitoring records, lab procedures, and other critical controls. It should
`be stressed here that most of the work in an investigation occurs reviewing records. The
`practice of complete proactive documentation is critical to the success of any investigation. The
`likelihood of an inconclusive investigation (and therefore surety of failing product) is assured
`if the associated records do not support a definitive finding.
`Schroeder (34) published a thoughtful review of considerations for a sterility failure
`investigation. He argues that for products sterilized by filtration filter failure must also be
`considered in addition to the other commonly cited areas of investigation.
`
`CONCLUSIONS
`The current, harmonized Sterility Test has two fundamental weaknesses, both of which have
`been obvious from its inception. The first is that the sampling plan is insufficient to meet the
`requirements implied by the title of the test. This weakness is not solvable in the current
`regulatory climate (nor has it been for over 70 years). The second weakness of the test
`involves recovery and recognition of microbial contamination in the sample, should it exist.
`There are several different varieties of the Sterility Test, and even when citing the
`harmonized test the user must be sensitive to regional expectations for that test. While there
`is great promise in finding a rapid method for conducting sterility tests, few examples exist
`of this having been successfully accomplished. Finally, there are clear expectations on the
`investigations to conduct into a failed Sterility Test, and the user is urged to be familiar with
`these expectations.
`
`REFERENCES
`1. Anon. Brit Pharm. Tests for sterility. Brit Pharm 1932:632 633.
`2. Bryce DM. J Pharm Pharmacol 1956; 8:561.
`3. Bowman FW. J Pharm Sci 1969; 58(11):1301 1308.
`4. USP. <71> Sterility Tests. USP 32, The United States Pharmacopeial Convention, Rockville, MD,
`2009:80 86.
`5. Knudsen LF. Sample size of parenteral solutions for sterility testing. J Am Pharm Assoc 1949; 38:
`332 337.
`6. Abdou MA F. Comparative study of seven media for sterility testing. J Pharm Sci 1974; 63(1):23 26.
`
`Regeneron Exhibit 1016.208
`
`
`
`194
`
`VOLUME 2: FACILITY DESIGN, STERILIZATION AND PROCESSING
`
`7. Clausen OG. A study of the growth promoting properties of fluid and solid microbial contamination
`tes