Validation of
`
`Pharmaceutical
`
`Processes
`
`Third Edition
`
`BRIHSH LIBRARY
`DOCUMENW RUPPIY CENTRE
`
`2 1 JUL 2009
`
`_ m09].
`-— 2597a
`
`—-
`"‘
`
`Edited by
`James Agalloco
`Age/loco & Associates
`Bel/e Mead. New Jersey. USA
`
`Frederick J. Carleton
`
`Carleton Technologies Incorporated
`Boynlon Beach. Florida. USA
`
`informa
`healthcare
`Mr.» Inn tendon
`
`Regeneron Exhibit 1052.001
`
`

`

`Infomia llealthcare USA. Inc.
`:2 Vanderbilt Avenue
`New York, N‘i 10017
`
`I." 2008 by Informa Healthcare USA, Inc.
`Int'nrma Healthcare is an Informa business
`
`No claim to original L33. Government works
`I'rinted in the United States of America on acid-tree paper
`lll987b§4321
`
`International Standard Book Number-l0: 0—84934055-3 (Hardcover)
`International Standard Book Number-13: 978-084934055-7 (Hardcover)
`
`This book contains information obtained from authentic and highly regarded sources. Reprinted material is quoted with
`permission, and sources are indicated. A wide variety of references are listed. Reasonable efforts have been made to publish
`reliable data and information, but the author and the publisher cannot assume responsibility for the validity of all materials or
`for the consequence of their use.
`
`No part of this book may be reprinted, reproduced, transmitted, or utilized in any form by any electronic, mechanical, or other
`means, now known or hereafter invented. including photocopying, microfilming, and recording, or in any information storage
`or retrieval system. without written permission from the publishers.
`
`For permission to photocopy or use material electronically from this work, please access www.copyright.com (httpzf/www.
`copyrightcom/l or contact the Copyright Clearance Center, Inc. (CCC )i 222 Rosewood Drive. Danvers. MA 01923, 978—750-8400.
`CCC is a not-for—profit organization that provides licenses and registration for a variety of users. For organizations that have been
`granted a photocopy license by the CCC, a separate system of payment has been arranged.
`
`Trademark Notice: Product or corporate names may be trademarks or registered trademarks. and are used only for
`identification and explanation without intent to infringe.
`
`
`Library of Congress Catalogingdn-Publicatinn Data
`
`Validation of pharmaceutical processes / edited by James P. Agalltx‘o, Frederick I. Carleton.
`- 3rd ed.
`P. J (m.
`Includes bibliographical references and index.
`[SEN-13: 97WS-l93-7055—7 (hardcover : alk. paper)
`ISBN-10: 0-8493—7055-8 (hardcover : alk. paper)
`I. Sterilization.
`
`2. Pharmaceutical technology—Quality control.
`3. Pharmaceutical technology—Standards.
`I. Agalloco, James I’., 1948- ll. Carleton, Frederick J., I923
`[DNLM2 1. Technology, Pharmaceutical. 2. Quality Control. 3 Sterilization—methods. Q's" 778 V172 2007"]
`R5] 99.573V345 2007
`615'. l q—chZ
`
`20070 I 821%
`
`Vi5it the Informa Web site at
`www.infonna.com
`
`and the Informa Healthcare Web site at
`www.informahealthcare.com
`
`Regeneron Exhibit 1052.002
`
`

`

`
`
` This material may be protected by Copyright law (Title 17 U.S. Code)
`
`
`16—
`
`Validation of Ethylene Oxide Sterilization Processes
`
`Iohn R. Gillie
`SGM Biotech. inc. Bozeman. Montana. U.S.A.
`
`Gregg Mosley
`Biotest Laboratories. Inc, Minneapolis. Minnesota. USA
`
`INTRODUCTION
`
`The recent history of EU sterilization has been dominated
`by advances in engineering technology These advances
`include not only the computerized controls for the
`operation of sterilizers, but also the physical environ-
`mental controls that permit safe use of 100% E0 gas.
`Most of these improvements have been driven by econ-
`omics. Faster. cheaper processes are indeed worthy
`objectives. However. sometimes it seems we have lost
`the focus on what we are really attempting to accomplish
`in the sterilization processes. Any sterilization process
`must deliver a lethality that kills the naturally occurring
`bioburden microbes that contaminate the products and
`materials. if the process does not render the products or
`materials free from living microorganisms, then steriliza-
`tion has not been achieved.
`
`The microbiological dimension of the EC) process
`has been overwhelmed by the recent strides in
`engineering and the physical process controls. The
`increasing complexity of medical products would
`be much more difficult
`to sterilize without
`these
`
`improvements.
`corresponding engineering, process
`However. failure to properly address microbial lethality
`renders all of these engineering advancements. mean-
`ingless it the resulting product is not sterile.
`Prix‘ess validation means establishing by obiective
`evidence that a process consistently produce» a result or
`product meeting its predetermined specifications ( l).
`The EU sterilization process is expected to deliver
`sterile products that possess all other specified quality
`attributes. Validation must document all critical process
`controls. The products to be sterilized must be challenged
`with an appropriate microbial system located in the
`"Worst case" or "least
`lethal" product
`location.
`In
`
`‘il'l'fl'i'hlfmlt‘ mm! in ”“5 chapter. AAMl, .-\:~,~'ocialion for the -\dvance—
`munl of Medical Instrumentation; Lil, biological
`indicator BIER.
`biological
`indicator evaluator reshtometi-r, ("(13, carbon dioxide;
`DEC. dynamic environmental conditioning, DLFT, device under
`test; F0. E10. ethylene oxide; FDA, Food and Drug Administration:
`(EC, gas l hromamgraphv, ll‘, inoculated product. IR, mtrared; MW,
`molecular in eight; NlUSi l, National institute of Occupational Safety
`and Health; \llm’. National Institute of Science and Tet hnologv; 1“qu
`permissible mposure limits, RH, relative humidity; R'l‘l“), resistance
`temperature detector. SAC, static atmoiepheric conditioning; SAL,
`sterility assurance level, Si R. spore lo} reduction: TAR. test accuracy
`ratio; Tt'.
`thermocouple. TISR. test uncertaintv ratio. HV‘A. time-
`“ t’tgl‘l ted .n eraize.
`
`addition, this microbial challenge product must be posi-
`tioned in the worst case,
`least lethal
`locationts) in the
`production load.
`If the microbial challenges are not
`located in these least lethal locations, then the resulting
`documented evidence may be biased and result in false
`Conclusions about the adequacy of the sterilization vali—
`dation program.
`The validation of the E0 gas sterilization process is
`one of the more complm programs facing process engin-
`eers and micmbiolngists because some critical process
`parameters are interactive. EO gaseous sterilization has
`been shown to be an extremely effective process that can
`be performed with an infinite number of combinations
`of parameters. Key parameters that affect sterilization
`efficacy are (1') concentration of E0 gas.
`(ii) RH,
`(iii) temperature of the process,
`(it’) accessibility of
`the product and packaging for these parameters, and
`tr) time.
`A validation program must demonstrate that the
`selected combination of these interactive process pare
`ameters result
`in an effective physical and biological
`process. The effectiveness of this process is measured by
`calibrated physical instruments and a calibrated micro—
`bial challenge. These process parameters must then be
`correlated to a calculated SAL for the product. SAL is the
`probability of a single i iable miermirganism occurring on
`a product. The required assurance level may vary
`depending on the product itself or the end use of the
`product. but is typically less than one chance in a million
`of a non-sterile unit or SAL of lit
`Another challenge is the task of assuring that
`the E0 gas used does not create a health hazard
`for
`the employees in the working area or
`leave
`unacceptable residuals in the product delivered to the
`consumer. Adsorbed E0 gas is removed fairly rapidly
`from processed materials, while absorbed E0 gas is
`released much more slowly. This absorption rate is
`highly dependent on the specific process conditions.
`material being, processed, as well as the geometry of the
`product, which affects material surface—to—volume ratios.
`Appropriate measures must also be taken to assure that
`E0 gas used in the sterilizing environment is Controlled
`and contained so that environmental insult in affected
`work areas is within acceptable regulated limits.
`During the E0 gas sterilization proceees, the gas
`interacts with the materials processed by reaction,
`absorption or adsorption. The EU gas is also trapped in
`the air spaces within the product or material being
`
`Regeneron Exhibit 1052.003
`
`

`

`242
`
`m
`
`STERILIZATION SANITIZATION AND STERlLIT‘i’ ASSURANCE
`
`sterilized. Unreacted residual gas is rapidly removed
`through evacuation, heated nitrogen or air exchanges.
`Product
`that
`is removed from a sterilizer must be
`
`to the
`insult
`controlled to prevent environmental
`workers, The best procedure is to place the sterilized
`materials in an environment that aids the desorption of
`the gas and is environmentally controlled to minimize
`workplace contamination.
`
`CHARACTERISUCS 0F El]
`
`Chemical Properties
`l, l-epoxyethane, and
`ED is also referred to as EtO,
`dimethylene oxide (2).
`It has a formula of CgHiO. The
`following structure is illustrated:
`
`rug—CH:
`0
`
`is a colorless gas, with a molecular weight of
`it
`44.05. It has a characteristic ether—like odor at toxic levels.
`
`EU has a boiling point of lt).7“C (51.3°Fi at 760 mmHg
`pressure, a melting point of —ilZ.6”C (, 1707?), a
`specific gravity of 0.87“ apparent at 20"C (60’1"), or a
`specific gravity of 0.897 at PC. E0 has a vapor density of
`1.5, with dry air being, equal to 1.0. and a vapor pressure
`at ZUT of [095 mmHg. It is completely miscible in water,
`alcohol. acetone, benzene, ether, carbon tetrachloride.
`HCFCs, and most organic solvents, and is a powerful
`solvent for fats, oils, greases, waxes. some rubber formu-
`lations, and paints. it is highly exothermic and potentially
`explosive when heated or mixed with (i) alkali metal
`hydroxides.
`(ii) highly active catalytic surfaces such
`as anhydrochlorides of iron,
`tin. or aluminum, and
`(iii)
`the oxides of iron and aluminum. The explosive
`limits are 3"?"
`to 97% by volume in air.
`It has a flash
`point of —ti“C (20”F).
`It
`is relatively noncorrosive for
`materials. E0 is relatively stable in neutral aqueous
`solutions and when diluted with liquid or gaseous
`carbon dioxide or halocarbons such as HCFCs. E0 is
`
`relatively unstable in either acidic or alkaline aqueous
`solutions and may rapidly form ethylene glycol.
`
`Biological Activity
`EO. reacts irreversibly with numerous chemical moieties
`on cellular molecules by an alkylation reaction where the
`[CHgOH-{Hg‘l alkyl group is covalently bonded with
`the available moiety via an addition reaction. Reactions
`with ~NH3, —SH, —COOH. and (Ii—[30H groups are
`common and illustrated in Figure l (3).
`Reaction rates vary and depend on the Specific pic.
`for each moiety and the existent pH. For a more compre-
`hensive review of possible reactions we refer the reader to
`Russell H). First-order lethality kinetics require that only
`one molecule per cell is the critical target (5—7). Reactions
`other than the critical reaction leading to microbial
`inactivation must be considered collateral damage
`reactions. Not all microbial
`inactivation obeys firsts
`order kinetics. However, even where multiple sites or
`molecules may be required for inactivation, the concept
`regarding critical reactions and collateral reactions is the
`same. Where inactivation is the result of cumulative
`
`then some
`damage, which is not first-order kinetics.
`damaging reactions must be considered more important
`to the events leading to microbial inactivation tcritical)
`
`
`E0 Alkylatlon Action
`
`
`
`NH-CchHQOH
`S-CHZCHQOH
`I
`
`.
`
`CH2
`
`COO-CHZCHEOH
`
`Illustration of the alkylation reaction of ethylene oxide
`Figure 1
`with chemically active moieties in the bacterial coil.
`
`than other reactions (collateral). Winaro and Stumbo (ti)
`identified E0 reactions with DNA as the critical reactions
`
`resulting in microbial inactivation. Lawley and Brooke:
`(9) defined specific reactions of ED with the nucleic acid
`tertiary heterocyclic nitrogen sites (=i\i-i in numerous
`experiments resulting in more than 10 publications
`between l957 (l0) and 1963 where they state (9):
`1. Sites in the nucleic acids reactive towards alkylating
`agents are shown to be,
`in order of decreasing
`reactivity: for RNA, N—7 of guanine. N-l of adenine.
`N-l of cytosine and N-3 of adenine for DNA, N-7 of
`guanine. N—3 of adenine and N-l of cytosine.
`Denatured DNA behaves in this respect like RNA.
`The observed differences between DNA and RNA are
`ascribed to the involvement of N-l of adenine and of
`
`[-J
`
`3.
`
`cytosine in hydrogen bond formation in DNA.
`In all cases alkylation results in destabilization of the
`nucleosides or the corresponding moieties in the
`nucleic acids. At neutral pH, with DNA, 7-alkyl-
`guanines and 3—all-tyladenines are slowly liberated
`by hydrolysis, the latter at the greater rate, whereas
`with RNA slow rearrangements occur,
`l-alkyla-
`denim: moieties yielding b-methlaminopurine
`moieties and l-alkylcytosines giving the corre-
`sponding l-alkyluracils.
`More recent studies suggest that disruption of the
`DNA molecule may occur differently depending on
`various repair mechanisms (ll).
`In the case of certain
`repair mechanisms, the reactions with cytosine may be
`the injury which ultimately leads to the inactivation of
`the microbe.
`
`VALIDA'I'IDN UF E0 STEHILIZATIDN PROCESSES
`
`Validation of the E0 process is divided into two phases:
`Engineering Qualification and Process Qualification.
`When these activities are completed successfully and all
`aspects of the process are documented. the process can be
`certified for routine use for manufacturing goods.
`
`Engineering Qualification
`Engineering Qualification deals with the sterilizer and
`associated equipment used in the process. This phase is
`divided into three segments: Installation Qualification,
`Calibration, and Operational Qualification.
`
`Regeneron Exhibit 1052.004
`
`

`

`Installation Qualification
`
`installation Qualification requires an audit of the equip-
`ment as it has been installed in the facility. This audit
`includes checking all utilities and supplies to the equip-
`ment
`to make sure that
`they meet
`the manufacturer's
`recommended specifications. Engineering drawings must
`be evaluated to assure that (i) the equipment is assembled
`according to the manufacturer’s prints, (ii) the equipment
`is installed according to the installation schematics, and
`(iii) all aspects of the equipment are documented with
`appropriate engineering drawings or sketches. These
`drawings are essential for future reference to compare
`the ha rd wa rc validated to any future configurations. This
`segment of the validation program is probably the most
`abused with sterilizers using nonexplosive containers of
`E0. Systems using 1011“» are extremely well documented,
`which is d m en by the safety issue. Once the equipment is
`hooked up and it "runs," little more is ever documented.
`With the pressure to get things working, little attention is
`paid to the documentation for future reference.
`inade-
`quately treated items typically include documentation of
`utilities, spare parts lists and preventive maintenance
`procedures. Many validations have been performed
`with all the necesmiry tests on the hardware relating to
`product loads, but with no record as to the exact configu-
`ration of
`the equipment when the validation was
`executed. Since any mechanical device will routinely
`malfunction, or wear out and require replacement, it is
`absolutely essential
`that a well—prepared Installation
`Qualification document be assembled for each piece of
`equipment to be validated. If this is not done, subsequent
`validation data may prove meaningless.
`
`Calibration
`
`.
`
`The second segment of the Engineering Qualification is
`the calibration of all process sensing, controlling, indicat—
`ing,
`and
`recording
`devices
`on
`the
`sterilizer
`or independent systems associated with it. Recording
`instruments that appear on the control panel are typically
`calibrated, but many of the control instruments are often
`located out of sight and should not be ignored since they
`may have a tremendous impact on the cycle function. For
`example with the DEC phase of an E0 sterilizing, process.
`it is extremely important to calibrate the stall point of the
`vacuum pump before the actual pressure or temperature
`set points are calibrated. This measurement is critical to
`balance the steam input into the chamber in relation to the
`capacity of the vacuum pump to remove the steam from
`the chamber. All critical process control instruments that
`are recorded and displayed by the control system must be
`calibrated. This is even more complicated when micro
`processor Control units are employed, because not only
`are there specific operating set points for those systems,
`there are also high- and low—limit alarm and other default
`systems that must be documented and calibrated. The
`calibration program will also vary depending on the type
`of computerized system,
`The calibration program should be performed with
`instruments referenced as secondary standards. The
`secondary or transfer standard is a standard that can be
`transported to and from the actual sterilization equip-
`ment because most instruments associated with the
`sterilixer must be calibrated at
`the sterilizer's location.
`
`t6
`
`VALIDATION a; ETHVLENE OXIDE sremumuow PROCESSES
`
`243
`
`Secondary standards must be traceable to a recognized
`standard such as those maintained by the hilt-ST.
`A measurement or calibration compares a BET to a
`standard or reference. This standard should outperform
`the DUT by a specific ratio, called the ”TUR“ also known
`as the TAR. As a rule of thumb, the TUR should be greater
`or equal to 4:] (12).
`Primary standards should have an even greater
`sensitivity.
`It
`is recommended that these primary stan—
`dards be submitted to the NIST for calibration and
`
`recertification on a periodic basis. Primary standards
`are usually recertified annually. It is extremely important
`that detailed procedures be established including limits
`and acceptable correction variances allowed and cali—
`bration frequency for all
`the instruments on the
`sterilizer. Adequate records must be maintained. A
`tracking system is essential to assist metrology, assuring
`that required calibrations occur at their designated
`frequencies. A history file should be maintained for
`each instrument. and the records reviewed to assure
`
`established calibration frequencies are appropriate.
`
`Operational Qualification
`The third segment of Engineering Qualification is Oper—
`ational Qualification that deals with the operating
`parameters of the sterilizer: their function, adjustment,
`and control. These tests are performed with an empty
`chamber. The various parameters for the cycle are eval-
`uated to determine if they perform as specified by the
`manufacturer. Temperature controllers are set and eval-
`uated to determine performance. The temperature
`distribution within the sterilizer is documented. The
`
`unit is sequenced through its operating steps to assure
`that the sequencing is appropriate. Every operating
`parameter must be documented to determine its compli-
`ance with the manufacturer's operating specification. The
`Operational Qualification protocol will serve as the basis
`for developing the Standard Operating Procedure for
`routine operation of the sterilizer. The Operational Quali—
`fication testing specifies in detail how the equipment
`operates.
`
`Process (Perturmance) Qualificatinn
`The final phase of validation deals with Process Qualifica-
`tion. Even though the unit functions appropriately with
`an empty chamber, it must now be demonstrated that it
`sterilizes product This phase may require repetition with
`different products and loads.
`
`Load Configuration
`There are several key aspects of Process Qualification.
`First. the specific product and all its packaging must be
`defined. The next step is to define the way master cartons
`are arranged into pallets. Pallet arrangement within the
`sterilizer is also part of the load configuration definition.
`Many manufacturers have numerous products that must
`be mixed together in order to achieve effective sterilizer
`throughput.
`Categoriring product for the sterilizing load is an
`extremely important element.
`It
`is important
`that
`the
`particular product mix is configured with a rationale
`that packaging is similar and products should be of
`consistent mass and materials and actual product
`
`Regeneron Exhibit 1052.005
`
`

`

`244
`
`ill
`
`summation SANITIZATlON aria STERiLirmssutiatict
`
`is possible that a manufacturer may
`it
`configuration,
`hat e in its catalog, hundreds ol different products.
`It
`these all have the same characteristics and packaging,
`it
`is possible that the} could be sterilized within one or
`two dilterent sterili/ation cycles.
`It
`is also possible that
`a manufacturer ma}, produce unit a few products, each
`hemp so different from the other products manutattured
`that each product sterilization prm‘ess hill hate to he
`validated in different cvcies.
`
`Unce product categories have been identified, it is
`also possible to \‘ar\' the load configurations. Loads must
`be e\ti‘eiiielt specific in the i\ at thei are defined. Small
`tolerances are permissible without changing the oiei'all
`impact on the biological effectiveness of the sterilization.
`However. changes in the qualified load must he evalie
`ated and properly documented to determine the
`potential
`impact on the biological ett‘ectii‘eness ot
`the process.
`Conditions which influence the lethality delitered
`by the process are mass. density. packaging. product
`design and materials. External preconditioning of
`product
`loads is common and allows easy process
`measurements. Preconditioning time may vary with
`different
`loads. Determining moisture. EO gas and
`temperature penetration into the palletized load is
`much more difficult in the sterilizer.
`if time to achieve
`
`acceptable levels of the- . parameters is similar to the
`originally defined load.
`then loads can be considered
`equivalent. Measured lethality should be similar with
`similar loads.
`It should he noted that configuration
`changes may influence the location of ”worst case—
`least
`lethal" position. Confirmation will have to he
`performed and appropriate adiustmei‘its may have to
`be made to assure that a proper monitoring location
`is documented.
`
`Once the product and load have been defined, then
`the worst case—least lethal locations in the pmduct, within
`each pallet and within the vessel must be determined,
`These locations will have to be monitored physically and
`biologically to provide data on all critical process
`parameters
`
`Pal/e! Configurations
`Pallet construction mat depend in part on how much
`shipping will take place between the time of construction
`until sterilization, When processing was pertormed
`in-house, it was easy toconstruct pallets with “chimneys”
`configured between columns of master cartons l'hese
`chimneys assured that more surface area of the master
`cartons was directly accessible for thermal transfer and
`gas ewhange. This ti'pe of configuration provides the
`greatest homogeneity of steriliration conditions across
`the product load.
`('ontract EU sterilization is extremely popular
`today and provides users with “state~ot—thc~art‘
`systems at reasonable expense. the problem comes not
`from the sterilizer. but from the logistics involied in
`transporting the product
`tiff~site to the contractor,
`Pallets are constructed at
`the product manufacturing
`site with transportation in mind, not sterili/ation
`Pallets are denseli packed because they survii e the
`rigors of overland shipping much better than pallets
`configured with \‘oid spaces
`(chimneys)
`tor gas
`
`
`
`Flgura 2 An example 01 a banded pallet ol product prowding
`mammurn surface exposure to sterilization vapors. Note: Corner
`protectors on pallet protecting the master cartons.
`
`permeation. Stretch wrap is commonly used to hold
`the palletired hows together. Stretch W rap is eweptional
`tor maintaining pallet
`integriti during shipping,
`but
`it ma). create a tremendous, barrier to sterilizing,
`\apor penetration. Stretch wrap manutactiirers are now
`uttering a "net" type of o rapping material which signi-
`ticantly increases the surface ot
`the master cartons.
`directly exposed to the sterilizing vapors. The best tech-
`nique from a sterilization perspective is to use strapping:
`to band the pallets together. ”its requires the use of
`corner protectors so 1h not to crush the outside Corners
`of the master cartons (Fig. 2], An example of uniformly
`constructed pallets loaded into a sterilizer vessel appears
`in Figure 3. More pallet configurations can be sterilized
`successfully of course, but prix‘i‘ss times may be longer
`
`
`
`Figure 3 An example ot a uniform load configuration Two
`identical pallets side by SlCle All pallets are exactly the same in
`construction and product
`
`Regeneron Exhibit 1052.006
`
`

`

`lethality and £0 residuals
`and variations of microbial
`across the load may be greater.
`
`gas supply lines allowing makeup charges to maintain the
`target gas concentration.
`
`‘16
`
`VALIDATION OF ETHYLEME OXIDE STERIL|ZATl0N PROCESSES
`
`245
`
`CRITICAL STEHILIZATION PROCESS PARAMETERS
`
`There are several major considerations to be aware of in
`order to structure a validation program that will assure
`that the sterilization process does what it
`is intended
`to do.
`
`These considerations include: (1') controlled process
`parameters and their interaction; (ii) an integration of the
`physical process conditions; (iii) the selection of appro-
`priate process conditions; (in) the product design; to) how
`the product is pretreated prior to exposure; (hi) how the
`product is handled following sterilization: (I-iil how the
`process is monitored. including physical, chemical, and
`biological methods; and (viii) the effect of residual E0 and
`its reaction products on the material being sterilized.
`There are [our critical interactive parameters that
`must be controlled for E0 sterilization process: ii) E0 gas
`concentration; (ii) moisture; (iii)
`temperature: and (mi
`time. All these parameters interact to affect the lethality
`delivered by the process.
`
`E0 Gas Concentration
`
`General Use Range of E0 Gas Concentration
`50 gas concentrations below 300 mg/L and above
`1200 mg/ L are not commonly used in the industry F0
`gas concentrations less than 300 mg/ L are not effective in
`practical process times. Concentrations above 901)
`to
`1200 mg/L do not shorten the process times sufficiently
`to warrant the additional cost of gas. Sterilization effec-
`tiveness is dependent on the molecular collision of the F0
`molecule and the biological entity that is being sterilized.
`Therefore. more E0 molecules lead to more rapid micro-
`bial lethality. A sterilizing process using (>01) mg/l. of E0
`delivers approximately twice the lethality as a process
`using Still mg/ L in the same time. However, considering
`the cost of E0. processes are generally designed toward
`the lower concentrations of E0. Concentrations of «100 to
`
`btltlmg/I. appear to be the more popular conditions
`today for operations to balance the cost of ED, equipment
`and throughput time.
`
`ED Gas Concentration Controllers
`
`The I30 gas concentration is controlled in one of two
`ways. The most common method of control is the indirect
`method through the use of a pressure control system. The
`EU gas concentration desired is calculated as to the
`corresponding increase in pressure. The desired pressure
`settings are then maintained by conventional pressure
`controllers. The direct control method uses analytical
`instruments that achaally detect the E0 gas concentration
`in the environment inside the sterilizer.
`The analytical systems are either gas chrom~
`atographic,
`[R
`or microwave
`detectors. These
`instruments are installed directly to the sterilizer. Periodic
`gas samples are withdrawn from the sterilizer or gas
`circulation lines and passed through the detector. Some
`[it or microwave detectors maybe mounted on the exterior
`chamber wall using an access port or in the gas circulation
`system. Electronic signals are sent to control valves in the
`
`Indirect Methods
`
`There are two approaches for the indirect method
`of measuring E0 gas concentration in the sterilizer—
`they are weight and pressure. The indirect methods are
`dependent on using gas cylinders containing certified
`mixtures of EO. When the chamber is pressurized, it is
`assumed that the mixture contains the given percentage of
`E0 relative to the change in pressure. Therefore, this
`change in pressure can be equated to an aSSumed gas
`Concentration. This system is very easy to monitor using
`pressure transducers and recorders.
`The second indirect method measures the weight of
`the gas cylinder contents dispensed into the vessel. This
`method assumes that a uniform mixture of the E0 and
`
`diluent gas was dispersed into the vessel, yielding an
`assumed concentration of gas in the sterilizing chamber.
`This system is easy to monitor using acceptably
`sensitive scales.
`These indirect methods are reasonably good est-is
`mates
`for most gas mixtures. Neither method
`compensates for absorption of E0 by the packaging
`materials or the product. Different materials absorb E0
`at different rates than they do diluent gases (I3). Further-
`more, indirect methods do not consider physical leaks in
`the sterilization system. Thus the indirect method. at best,
`provides an approximation of the E0 gas concentration in
`the vessel.
`
`Direct Gas Measurement
`
`Direct analysis of the E0 in a sterilizing chamber can be
`performed by specific analytical instruments. Two of the
`most common analytical methods are the GC and the
`IR spectrophotometer.
`
`Gas Chromatography. CC has been the most widely
`used method for determining the level of EC.)
`in the
`sterilizing environment Some EO processes operate at
`atmospheric or positive pressure, making withdrawal of
`a gas sample easy. Sterilization processes that use 100%
`EU with a nitrogen blanket may operate at slightly
`subatmospheric pressures and sampling is slightly more
`difficult. CC is not used in It
`":2, E0 processes with no
`nitrogen overlay because they operate under a deep
`vacuum. When dealing with explosive mixtures of E0
`or pure E0, only intrinsically safe instrumentation must
`be used.
`
`Sample removal is extremely important in order to
`assure meaningful data. Sample lines must be heated and
`insulated upon exiting the sterilizer. It cold spots occur in
`the sampling lines, the E0 and water vapor may condense,
`yielding false data. These samples may be collected using
`gas collection bottles or with lines attached directly to the
`CC if an automatic injection system such as a gas sampling
`loop is used.
`Multiple sample sites also present a problem.
`Representative sites are generally selected throughout
`the sterilizing chamber. Small capillary tubes serving as
`sample delivery lines are fitted to the gas sample ports.
`Care must be taken to permit these sample tubes to be
`flushed to assure that
`the sample being extracted is,
`
`Regeneron Exhibit 1052.007
`
`

`

`245
`
`m STEHILIZATION. summation AND STERlLlTY ASSURANCE
`
`from the chamber environment and not a
`indeed.
`in the sample delivery line. For this reason,
`residual
`this method is not acceptable for sampling,r within the
`product or product packages. The flushing of the sample
`lines accelerates the gas penetration into these restricted
`locations and yields data that are not representative of
`actual load conditions,
`
`Gas samples can be extracted from the gas recircula—
`tion system. This provides a good estimate of the gas
`concentration in the chamber.
`
`The GC unit must be calibrated prior to sample
`analysis with a certified standard gas. This certified
`standard may be either a diluted gas mixture or lll"'..
`EO. Most laboratories that are established to perform CC
`analysis are qualified to use 100% E0 as the standard for
`calibration. However,certitied mixtures are available from
`
`gas suppliers. The (SC is calibrated at one point with this
`standard gas and expressed as mole percent. These cali-
`bration results are independent of temperature and
`pressure. The mole percent concentration of the steriliza-
`tion chamber is compared to the standard gas and is then
`converted into mg/ L
`
`100%
`x
`
`mol
`273°C
`( 273°C + ”Cl
`
`g
`=xr
`
`Ill-L7 psia 4: l' psigl
`14.7 psia
`
`where Y psig, pressure of the sterilizing chamber;
`"C, temperature in the sterilizing chamber; XF, scaling
`factor.
`Therefore, the E0 concentration from the GC data
`in mole percent multiplied by the scaling factor (XF)
`yields mg/ L.
`
`[R Analysis. Most gases have a characteristic IR spec-
`trum that can be used to identify them. These spectra are
`usually rather complex; however, each usually contains a
`small number of strong analytical bands that are used in
`this analysis.
`These lR analyzers incorporate a fixed wave-length
`filter that corresponds to one of these strong bands. An
`optical path is also chosen that provides the sensitivity
`range required for the particular analysis.
`The analytical wavelengt