`
`Guidance for Industry
`
`The FDA published Good Guidance Practices in February 1997.
`This guidance was developed and issued prior to that date.
`
`Additional copies are available from:
`Office of Training and Communications
`Division of Communications Management
`Drug Information Branch, HFD-210
`5600 Fishers Lane
`Rockville, MD 20857
`
`(Tel) 301-827-4573
`(Internet) hup://wlVwJda.govlcderlguidancelindex.hrm
`
`u.s. DEPARTMEI'« OF lb:.u.TH AND HU~lAN SERVICES, FOOD A ND DRUG ADMINISTRATION
`
`Apotex v. Abraxis - IPR2018-00152, Ex. 1009, p.OI of45
`
`
`
`GUIDELINE ON STERILE DRUG PRODUCTS
`
`PRODUCED BY ASEPTIC PROCESSING
`
`June, 1987
`
`(Reprinted June, . 1991)
`
`Prepared by: Center for Drug Evaluation and Research
`
`Center for Biologics Evaluation and Research, and
`
`Office of Regulatory Affairs,
`
`Food and Drug Administration
`
`Maintained by: Division of Manufacturing and Product Quality (HFD-320)
`
`Office of Compliance
`
`center for Drug Evaluation and Research
`
`Food and Drug Administration (FDA)
`
`5600 Fishers Lane
`
`Rockville, Maryland 20857
`
`Apotex v. Abraxis - IPR20 18-00 152, Ex. 1009, p.02 of45
`
`
`
`r.
`
`PURPOSE
`
`This guideline informs interested persons on certain practices and
`
`procedures for the preparation of sterile drug products by aseptic
`
`processing that const1tute acceptable means of complying with certain
`
`sections of the Current Good Manufacturing Pl"actice {CGMP}
`
`regulations for drug products (Title 21 Code of Federal Regulatfons.
`
`Parts 210 and 211). For biological products regulated under 21 CFR
`
`Parts 600 through 680, it should be noted that sections l1D.2(a) and
`
`21'.1(b) provide that where it is impossible to comply with the
`
`applicable regulations in both Parts 600 through 680 and Parts 210
`
`and 211. the regulation specifically applicable to the drug product
`
`in question shall apply. Therefore, the s terility testing of
`
`biological products, and the culture media employed for such testing,
`
`must conform to the requirements under sect10n 610.12.
`
`II. INTROOUCTION
`
`This guideline is issued under 21 eFR 10.90, and as such. it states
`
`principles and practices of general applicability that are not legal
`
`requirements but are acceptable to the Food and Drug Administration
`
`(FDA). A person may rely upon tn1s guideline with the assurance of
`
`its acceptability to FDA. or may follow different procedures. When
`
`. 1 -
`
`Apotex v. Abraxis - IPR201 8-00152, Ex. 1009, p.03 of45
`
`
`
`different procedures are chosen. a person may , but i s not requ ired
`
`to , di scuss the matter i n advance with FDA to prevent the expenditure
`
`of money and effort on activity that may later be determi ned to be
`
`una cceptabl e.
`
`This guidel ine may be amended from time to time as the agen cy
`
`.
`
`recognizes the need through its regulatory efforts and through
`
`coornen t s submi tted by interes ted persons.
`
`There are certain differences between the .production of steril e drug
`
`products by ase ptic process ing and by terminal ster i l ization.
`
`Termfnal sterilizati on usually involves f111fng and closing product
`
`containers under conditions of a high qual ity environment; the
`
`product. conta i ner. and closure are usually of a hi gh microbiological
`
`qua l f ty but are not sterile.
`
`It is important that the environment in
`
`wh ich filling and closing is achieved be of a high quality in order
`
`to minimize the microbial content of the prod uct and to help assure
`
`that the subsequent sterilization process is successful . The product
`
`in its final contai ner is then subjected to a ster n i zation
`
`process--usually using heat or ra diat i on.
`
`In aseptic processing, the
`
`drug product , conta iner, and closure are subjected to ster il i zati on
`
`processes separatel y and then brought together. Because t here is no
`
`further process ing to ste ril ize the product afte r it is i n its final
`
`container , it i s critical to the maintenance of product steri lity
`
`- 1 -
`
`Apotex v. Abraxis -lPR2018-001 52, Ex. 1009 , p.04 of45
`
`
`
`that containers be filled and closed in an environment of extremely
`
`high quality.
`
`In addition, there are usually more variables
`
`attendant to aseptic processing than to terminal processing, a factor
`
`that can malee 1t more di ff1cult to attain a high degree of assurance
`
`that the end product will be sterile. For example, before aseptic
`
`assembly, different parts of the final product may have been
`
`subjected to different sterilization processes -- such as dry heat
`
`for glass containers, steam under pressure for rubber closures. and
`
`f11 tration for a liquid dosage form -- each requiring thorough
`
`validation and control, each with the possi.bility of error.
`
`(For the
`
`terminally sterilized drug product, on the other hand. there is
`
`generally only one sterilization process, thus limiting the
`
`possibilities for error.) Furthermore, any manipulation of the
`
`sterilized dosage form, containers, and closures immediately prior to
`
`aseptic assembly involves the risk. of contamination and thus must be
`
`carefully controlled.
`
`These processing differences have led to several questions on aseptic
`
`processing regarding what FDA believes are acceptable ways of
`
`complying with certain sections of the CGHP regulations for drug
`
`products. The sections most frequently questioned concern buildings
`
`and facilities. components, containers/closures. production time
`
`limitations. validation, laboratory controls, and sterility testing.
`
`Because most of th e questions have concerned process validation in
`
`- 3 -
`
`Apotex v. Abraxis - IPR201 8-00152, Ex. 1009, p.05 of45
`
`
`
`particular, this guideline addresses this area extensively. This
`
`guidel ine ;s intended to respond to these questions and clarify
`
`certa in technical aspects of aseptic processing of sterile drug
`
`products.
`
`It should be noted that this document does not address
`
`several other important aspects of aseptic pr.ocessing--such as
`
`employee hygiene, aseptic gowning. and clean room design. These and
`
`other aspects will be covered 1n future revisions of thfs guideline
`
`as needed. This guideline does not address te~1nally sterilized
`
`drug products. although some portions may be applicable to their
`
`preparation also.
`
`In this guideline stated CGHP requirements of certain sections of 21
`
`CFR Part 211 are followed by discussions of practices and procedures
`
`which FDA considers as acceptable means of meeting the requirements.
`
`It should be noted that not all portions of the regulations which
`
`apply to the preparation of aseptically processed sterile drug ·
`
`products are identified -- only those portions for which pertinent
`
`questions have been raised. The guideline also tncludes a list of
`
`references which may be of value to the reader.
`
`Definitions
`
`Critical areas - Areas where sterilized product or container!
`
`closures are exposed to the environment.
`
`- 4 -
`
`Apotex v. Abraxis - IPR20 18-00 152, Ex. 1009, p.06 of45
`
`
`
`Critical surfaces - Surfaces which come into contact with sterilized
`
`product or containers/closures.
`
`o value - The ti me a t a given tempera ture needed to reduce the number of
`microorgani sms by 90t.
`
`Overk.ill sterilization process - A process which is sufficient to provide
`
`at least a 12 log reduction of microorganisms having a minimum 0 value of
`
`1 minute.
`
`Sterilizing filter - A filter which. when challenged with the
`microorgan1sm Pseudomonas dim1nuta. at a minimum concentration of 107
`organisms per cmZ of filter surface. will produce a stel"ile effluent.
`
`Validation - Establishing documented evidence which provides a high
`
`degree of assurance that a specific process w11l consistently produce a
`
`product meeting its predetermined specifications and quality attributes.
`
`Worst case - A set of conditions encompassing upper and lower processing
`
`limits and circumstances. including those within standard operating
`
`procedures, which pose the greatest chance of process or product failure
`
`when compared to ideal conditions. Such conditions do not necessarily
`
`induce product or process failure.
`
`- 5 -
`
`Apotex v. Abrax is - IP R201 8-0015 2, Ex. 1009, p.07 of45
`
`
`
`III.BUILDINGS AND FACILITIES
`
`Requirements
`
`Section 211.42 (design and construct1on features) requires. in part.
`
`that there be separate or defined areas of operation to prevent
`
`contamination. and that for aseptic processing there be. as
`
`appropriate. an air supply filtered through high efficiency
`
`particulate air (HEPA) filters under positive pressure. and systems
`
`for monitoring the environment an,d 11il1ntain1ng equipment used to
`
`control aseptic conditions.
`
`Section 211.46 (ventilation. air f11tration. air heating and cooling)
`
`requires. in part. that equipment for adequate control over air
`
`pressure, microorganislWs, dust, humid tty, and temperature be provided
`
`where appropriate and that afr filtration systems, including
`
`prefilters and particulate matter air filters. be used when
`
`appropriate on air supplies to production areas.
`
`Gu1danc;e
`
`In aseptic proceSSing there are various areas of operation which
`
`require separation and control. with each area needing dHferent
`
`degree s of air quality depending on the nature of the operation. Two
`
`expos ure ar eas are of parti cular importance to drug product
`
`qual ity-- c ritical ar eas and controlled areas.
`
`- 6 -
`
`Apotex v. Abrax is - IPR20I S-00152, Ex . 1009, p.OS of45
`
`
`
`CRITICAL AREAS
`
`A critical area is one in which the sterilized dosage form.
`
`containers. and closures are exposed to the environment. Activities
`
`that are conducted in this area include manipulations of these
`
`sterilized materials/product prior to and during f111ing/clos1ng
`
`operations. These operations are conducted in what is typically
`
`called the "aseptic core" or "aseptic processing" area.
`
`This area is critical because the product 15 not processed further tn
`
`its immediate container and is vulnerable to contamination.
`
`Therefore, tn order to maintain the quality and, specifically, the
`
`steril 1ty of the product. the environment in the ilTlllediate proximity
`
`of the actual operations should be of the highest qual1ty.
`
`One aspect of environmental quality is the particulate content of the
`
`air. Particulates are significant because they may enter a product
`
`and contaminate it physically or, by acting as a vehicle for
`
`microorganisms, biologically.
`
`It is therefore important to minimize
`
`the particle content of the air and to effectively remove those
`
`particles which are present. Air in the immediate proximity of
`
`exposed sterilized containers/closures and filling/closing operations
`
`is of acceptable particulate qual Hy when it has a per-cubic-foot
`
`- 7 -
`
`Apotex v. Abraxis - IPR201 8-00152, Ex. 1009, p.09 of45
`
`
`
`part icl e count of no more than 100 in a size range of 0 . 5 micron and
`
`larger (Class 100) when measured not more than one foot away from the
`
`work site, and upstream of the air flow, during fflling/clos1ng
`
`operations. The agency recognizes that some powder f1111ng
`
`operations may generate high levels of powder particulates which. by
`
`their nature. do not pose a risk of product contamination.
`
`It "may
`
`not. In these cases, be feasible to measure air quality within the
`
`one foot distance and still differentiate -background noise- levels
`
`of powder particles from air contaminants whfch can fmpeach product
`
`qual1ty.
`
`In theSe instances. It is nonetheless iliportant to sample
`
`the afr in a ~nner which. to the extent possibl e . characterizes the
`
`true level of extrinsic particulate contall'l1nation to which the
`
`product is exposed.
`
`Air in critical areas should be supplied at the point of use as HEPA
`
`filtered laminar flo~ air. haYing a velocity sufficient to sweep
`
`particulate matter away from the fi1"l inglclosing area. Normally. a
`
`velocity of 90 feet per minute, plus or minus ZOI. Is adequate,
`
`(Refs. 1 and Z) although higher velocities may be needed where the
`
`operations generate high levels of particulates or where equipment
`
`configuration disrupts laminar flow.
`
`- B -
`
`Apotex v. Abraxis - LPR20 18-00 152, Ex. 1009, p.l 0 of 45
`
`
`
`Air should also be of a high microbial quali t y. An incidence of no
`
`more than one colony fanning unit per 10 cubic feet is considered as
`
`attainable and desirable (Ref. 3).
`
`Air is not the only gas in the proximity of fi11-ingJclosing
`
`operations which should be of a high particulate and microbial
`
`quality. Other gases, such as nitrogen or carbon dioxide, which
`
`contact the product. contatner/closure, or product contact surfaces,
`
`e.g •• purging or overlaying. should be sterile filtered.
`
`In
`
`addition, compressed air should be free from demonstrable 011 vapors.
`
`Cr1tical areas should have a positive pressure differential relative
`
`to adjacent less clean areas; a pressure differential of 0.05 inch of
`
`water ;s acceptable.
`
`CONTROLLED AREAS
`
`The controlled area. the second type of area in which it is important
`
`to control the environment, is the area where unsterilized product,
`
`in-process materials, and container/closures are prepared. This
`
`includes areas where components are compounded, and where components,
`
`in-process materials, drug products and drug product contact surfaces
`
`of equipment. containers, and closures. after final rinse of such
`
`surfaces, are exposed to the plant environment. This environment
`
`- 9 -
`
`Apotex v. Abraxis - IPR201 8-00152, Ex . 1009, p.11 of45
`
`
`
`should be of a high mi c robial and particulate quality 1n order to
`
`minimize the level of particulate contaminants in the final produ ct
`
`and to control the microbiol ogica l content (bfoburden) of articles
`
`and components which are subsequently sterilized.
`
`Air in controlled areas is generally of acceptable partfculate
`
`quality 1f it has a per-cubic-foot particle count of not more than
`
`100,000 in a size range of 0.5 micron and larger (Class 100.000) when
`
`measured in the vicinity of the exposed articles during periods of
`
`activity. W1th regard to microb1al . quality. an incidence of no more
`
`than 25 colony formi ng units per 10 cubic feet is acceptable
`
`(Ref. 3).
`
`In order to maintain air quality in controlled areas, it i s important
`
`to achieve a sufficient air flow and a positive pressure differential
`
`relative to adjacent uncontrolled areas.
`
`In this regard. an air flow
`
`sufficient to achieve at least 20 air changes per hour and. in
`
`general, a pressure di fferential of at least 0.05 inch of water (with
`
`all doors closed). are acceptable. When doors are open. outward
`
`airflow should be sufficient to minimize ingress of contamination.
`
`Gases other than ambient air may also be used in controlled areas.
`
`Such gases should. if vented to the area. be of the same qual ity as
`
`ambient afro Compressed air shou ld be free from demonstrable oil
`
`vapor.
`
`- 10 -
`
`Apotex v. Abraxis -LPR20I S-00152, Ex. 1009, p.12 of45
`
`
`
`In addition to these production areas, there may be certain pieces of
`
`equipment which should be supplied \</ith high quality filtered air.
`
`This is especially important where the air in the equipment will
`
`contact sterilized material or material which should have a low
`
`microbial or particulate content. For example\ bacterial retentive
`
`filters should be used for lyophilizer vacuum breaks and hot air
`
`sterilizer vents to ensure that air coming fn contact with a
`
`sterilized product is sterile. Likewise, air admitted to
`
`unpressur1zed vessels containing sterilized liquid should also be
`
`filtered. Air in tanks used to hold mater.ial which must be of a high
`
`microbial quality should be filtered too, and the filters should be
`
`dry to prevent wetting by condensation with subsequent blockage or
`
`microbial grow-through (two ways of achieving this are providing heat
`
`to the filter and use of hydrophobic filters.)
`
`It 1s important that
`
`these equipment air filters be periodically integrity tested.
`
`An acceptable system for maintaining air quality includes testing
`
`HEPA filters for integrity.
`
`Integrity testing should be performed
`
`initially when the units are first installed in order to detect leaks
`
`around the sealing gaskets, through the frames or through the filter
`
`media. Thereafter, integrity tests ~: hould be performed at suitable
`
`intervals. Usually it 1s sufficient to perform such testing at least
`
`twice a year for critical areas. however, more frequent testing may
`
`be needed when air qua 1 i ty is found to be unacceptably low or as pa rt
`
`of an investigation into a finding of non-s.terility in a drug product.
`
`- 11 -
`
`Apotex v. Abraxis - IPR2018-00152, Ex . 1009, p.13 of45
`
`
`
`One acce pta bl e method of testin9 the i nte grity of HEPA filters 1s use
`
`of a dioctylphthalate (DOP) aerosol cha llenge.
`
`Inasmu ch as a HEPA
`
`f il ter is one capable of retaining 99.97 percent of particulates
`
`greater than 0.3 micron in diameter, it is important to as sure that
`
`whatever substance is used as a challenge will have a sufficient
`
`number of particles of this size range. An acceptable DOP challenge
`
`involves introducing a DOP aerosol upstream of the filter in a
`
`concentration of 80 to 100 micrograms/liter of afr at the filter's
`
`designed airflow rating and then scanning the downstream s1d~of the
`
`fil ter with an appropriate photometer pro.be at a sampling rate of at
`
`l east one cu bi c foot per minute. The probe should scan the entire
`
`fil ter face and frallle at a position about one to two inches from the
`
`face of the filter (Ref. 1). A s ingle probe reading equivalent to
`
`0.01 percent of the upstream challenge is considered as indicative of
`
`a significant leak which should be repaired.
`
`Use of particle counters without introducing particles of known size
`
`upstream of the filter is ineffective for detecting leaks.
`
`The r eade r shou ld note that there i s a differe nce between filter
`
`integrity test ing and efficiency testing.
`
`Integrity testing is
`
`performed to detec t leaks from the filte r media, fi lter frame and
`
`seal . The cha llenge is a polydispersed aerosol usually composed of
`
`particles ranging i n size from one to three mic ron s . The test is
`
`done in place and the filter fa ce i s scaned with a pr obe; the
`
`- 12 -
`
`Apotex v. Abrax is - LPR20 18-00 152, Ex. 1009, p. 14 of 45
`
`
`
`mea sured downstream leakage is taken as a percent of the upstream
`
`challenge. The effic1ency test, on the other hand, is used to
`
`determine the filter's rating . The test uses a monodispersed aerosol
`
`of 0.3 micron size particles, relates to filter media, and usually
`
`requires specialized equipment. Downstream readings represent an
`
`average over the entire filter surface. Therefore~ leaks 1n a filter
`
`may not be detected by an efficiency test .
`
`It is also important to monitor air flow veloc1t1es for each HEPA
`
`filter accordfng to a program of estab11shed intervals because
`
`significant reductions fn velocity can increase the possibility of
`
`contamination and changes fn velocity can affect the laminarity of
`
`the airflow. Airflow patterns should be tested for turbulence that
`
`would interfere with the sweeping action of the afro
`
`IV. COMPONENTS
`
`Requf rements
`
`Section 211.80 (general requirements) requires. in part, the
`
`establishment of written procedures for the storage, handling and
`
`testing. and approval or reject ion of components .
`
`• 13 .
`
`Apotex v. Abraxis - IPR20 18-0015 2, Ex. 1009, p.15 of45
`
`
`
`Section 211.84 (testing and approval or rejection of components, drug
`
`product containers. and closures) requires. in part. that components
`
`liable to microbiological contamination that is objectionable tn vtew
`
`of their intended use be subjected to microbiological tests before
`
`use.
`
`Guidance
`
`One of the most important aspects of cOllIponents used in sterile drug
`
`products made by aseptic processing 1s m1,crobiological quality. A
`
`finished drug product produced by aseptic processing may become
`
`contaminated through use of one or more components which contain
`
`microorganisms. Therefore~ unless an overkill sterilization process
`
`is appl ied to components. it 1s important to routinely characterize
`
`the microbial content of each component liable to contamination and
`
`to establish appropriate acceptance/rejection limits based on this
`
`bioburden. Knowledge of this b10burden 1s especially significant in
`
`attaining a high degree of sterility assurance when the component is
`
`subjected. to a non-overkill sterilization process.
`
`In aseptic processing, each component may be individually sterilized
`
`or several components may be combined. with the resulting mixture
`
`s teril ized. There are several methods to steril ize components. and
`
`each can be acceptable when properly validated. A wi dely used method
`
`- 14 -
`
`Apotex v. Abraxis - IPR201 8-00152, Ex . 1009, p.1 6 of45
`
`
`
`is filtration of a solution formed by dissolving the component in a
`
`solvent such as USP water for injection; the solution is passed
`
`through a sterilizing membrane or cartridge filter. This method can
`
`be useful where the component is likely to be adversely affected by
`
`heat. A variation of this method involves subjecting the filtered
`
`solution to aseptic crystallization and precipitation of the
`
`component as a sterile powder. However. this method involves more
`
`handling and manipulation than other methods and therefore has a
`
`higher potential for contamination during processing.
`
`If a component is not adversely affected by heat. and it is soluble.
`
`it may be made into a solution and subjected to steam sterilization
`
`either in a separate autoclave or within a steam-jacketed pressurized
`
`preparation vessel.
`
`Dry heat steril ization is a suitable method for components that are
`
`heat stable and may be insoluble. However, this method can pose
`
`problems of inadequate heat penetration and distribution. For
`
`example. the treatment of powders by this method necessitates
`
`suitable heat penetration and distribution studies because of the
`
`powders' insulating effects.
`
`- 15 -
`
`Apotex v. Abraxis - [PR20 18-00 152, Ex. 1009, p.17 of 45
`
`
`
`Ethylene oxide exposure is another method of sterilizing components.
`
`However. its effectiveness as a primary method is questionable
`
`because of a lack of consistent penetration of the sterilant to the
`
`crystal core of a powder. Ethylene oxide may be useful for the
`
`surface sterilization of powders as a precaution against potential
`
`microbial contamination during aseptic handling.
`
`For products intended to be pyrogen free. there should be written
`
`procedures for acceptance or rejection of ~omponents which are
`
`susceptible to pyrogens. Those components found to be cont~minated
`
`with pyrogens should be rejected or processed to remove the pyrogenic
`
`properties provided that the resultant components will meet
`
`appropriate standards. specifications. and characteristics.
`
`CONTAINERS/CLOSURES
`
`Requirements
`
`Section 211.94 (drug product containers and closures) requires. in
`
`part. that drug product containers and closures be clean and. where
`
`indicated by the nature of the drug, sterilized and depyrogenated.
`
`Standards and testing methods and, where indicated, methods of
`
`cleaning, sterilizing and processing to remove pyrogenic properties
`
`must be written and followed.
`
`- 16 -
`
`Apotex v. Abraxis - IPR20IS-00152, Ex. 1009, p.IS of45
`
`
`
`Guidance
`
`In the case of sterile drug products made by aseptic processing.
`
`preparation of containers and closures prior to filling and closing
`
`operations should go beyond mere cleaning to remove surface debris.
`
`It is critical to the integrity of the final product that containers
`
`and closures be rendered sterile and. in the case of injectable
`
`products. pyrogen free. The type of processes used to sterilize and
`
`depyrogenate will depend primarily on the nature of the material
`
`which comprises the container/closure. Any properly validated
`
`process can be acceptable.
`
`In tne case of glass containers. Ilresterllfzation preparation usually
`
`involves a series of wash and rinse cycles. Not only is it important
`
`that these washes effectively remove debris. it is also important
`
`that the f1nal rinse water be of a high quality. Final rinse water
`
`is acceptable if it meets the requirements of USP water for
`
`injection. Depyrogenation may be accomplished by a variety of
`
`methods; for example. by initial washings with chemical solutions
`
`followed by rinses with water for injection. Dry heat may be used to
`
`sterilize and depyrogenate glass containers. Validation of dry heat
`
`ster1ltzation/depyrogenation should include heat penetration and heat
`
`distribution studies as well as use of representative process ·cycles
`
`- 17 -
`
`Apotex v. Abraxis - IPR2018-00152, Ex. 1009, p.19 of45
`
`
`
`and loading configurations to simulate actual production practices.
`
`Whatever depyrogenation method i s used , the validation data should
`
`demonstrate that the process win reduce the endotoxin content by 3
`
`logs.
`
`One method of assess ing the adequacy of a depyrogenatfon process is
`
`to simulate the process using containers having known quantftfes of
`
`standardized endotoxfns and meas ure the level of reduction. However .
`
`FDA i s aware of one potential problem where challenge endotoxins are
`
`used to assess certain washing process~s. The problem stems from
`
`applying the powdered endotoxin challenge directly to the surface
`
`being tested, rather than first resolubflizing the material and air
`
`drying it onto the surface. The powdered material may be much more
`
`soluble in the wash and rinse water than the re constituted air-dried
`
`material and more so than endotox1ns that may normally be present on
`
`container/closure surfaces. This could result in the perception of
`
`the process under consideration as being much more efficient at
`
`endotoxin removal
`
`than it really is. Therefore, endotoxin challenges
`
`should not be easier to remove from the target surfaces than the
`
`endot oxi ns that may normally be present.
`
`Plastic containers, subjected to uncontrolled ha ndl ing and s t orage
`
`may be a source of pyrogens and should. therefore. be depyrogenated.
`
`Plastic co ntainers may be sterilized with ethylene oxide gas.
`
`Bi ol ogical indicators can be useful to monitor such processes, a long
`
`- 18 -
`
`Apotex v. Abraxis - IPR20 18-0015 2, Ex. 1009, p.20 of45
`
`
`
`with monitoring and control of temperature. pressure, humidity, and
`
`ethylene oxide concentration. The potential for residues. such as
`
`ethylene oxide and its degradation products. to remain on or in the
`
`container should be assessed.
`
`Rubber compound stoppers pose another potential source of mf~robfal
`
`and (of concern for products intended to be pyrogen free) pyrogen
`
`contamination. They are usually cleaned by multiple cycles of
`
`washing and r1nsinfj prior to final steam sterilization. The final
`
`rinse should be with USP water for injection.
`
`It is also 111portant
`
`to minimfze the lapsed time between washing and stertlizing because
`
`moisture on the stoppers can support microbiological growth and the
`
`generation of pyrogens. Because rubber 1s a poor conductor of heat.
`
`proper validation of processes to sterilize rubber stoppers is
`
`part1cularly 1mportant.
`
`VI. TIME LIMITATIONS
`
`Requirements
`
`Section 211.111 (time limitations on production) requires. in part.
`
`the establishment of time limits for completion of each phase of
`
`production. when appropriate. to assure the quality of the drug
`
`product.
`
`- 19 -
`
`Apotex v. Abraxis - IPR201 8-00152, Ex. 1009, p.21 of45
`
`
`
`Guidance
`
`In ase pti c process i ng of s teri 1 e drug products the es tab11 shment of
`
`time limitations is generally appropriate for s everal operations .
`
`The total time for the product fil tratton and filling operations. for
`
`example. should be limited to an established maximum in order ·'to
`
`prevent contamination of the filtrate by microorganisms growing or
`
`passing through the filter over a period of time . Such a limit
`
`should also prevent a significant increase in the number of
`
`microorganisms on the upstream side of the fflter. which increase
`
`could lead to pyrogen formation .
`
`VII. PRODUCTION ANO PROCESS CONTROLS; VALIDATION
`
`Regui rements
`
`Section 211.113 (control of microbiological contamination) requires.
`
`in part, the establishment and adherefl ce to appropriate written
`
`procedu r es designed to prevent microbiological contamination of drug
`
`produc t s purporti ng to be sterile. Such procedures mu st include
`
`vali dati on of any sterilization process .
`
`. 20 -
`
`Apotex v. Abraxis - IPR201 8-001 52, Ex. 1009, p.22 of45
`
`
`
`Gu i da nce
`
`[n order to ass ure the s terility of product s purporting to be sterile
`
`which are prepared by aseptic processing, it i s /'AOst important that
`
`two types of operations in particular be adequately val idated, namely
`
`s t erilization and filling/closing under aseptic conditions. The "
`
`objective of the most effective sterilization processes can be
`
`defeated if the sterilized elements of a product -- the drug, the
`
`container and closure -~ are brought together under conditions that
`
`contamtnate those elements. Conversely. product sterility rna .:,' be
`
`compromised where those conditions add no contamination whatsoever.
`
`but where the product elements are not sterile at the time they are
`
`as sembled.
`
`(luesttons have ari s en as to acceptable ways of val idating the aseptic
`
`assembly of sterile product elements, and the sterilization of those
`
`elements. However, the former operation is considered by some people
`
`to be the most difficult and has generated more questions.
`
`Therefore. the guidance presented places greater emphasis on
`
`val ida t ing the asepti c assembly (i .e .• filling/closing) operations .
`
`- 21 -
`
`Apotex v. Abrax is -LPR20I S-001 52, Ex. 1009, p.23 of45
`
`
`
`Aseptic As sembl y Operations
`
`An acceptable method of validating the asepti c assembly process
`
`involves the use of a microbiological growth nutrient medium to
`
`s imulate sterile product f111ing operatfons. · This has been termed
`
`"sterile media fillsM. The nutrient medium i s manipulated and
`
`exposed to the operators. equipment. surfaces. and environmental
`
`conditions to closely simulate the same exposure which the product
`
`1tsel f will undergo. The sealed drug produ ct containers filled with
`
`the meith are then incubated to detect I1fcrob1olog1cal growth and the
`
`results are assessed to determine the probability that any given unit
`
`of drug product may become contaminated during actual ftlling/cl osi ng
`
`operations. Media filling in conjunction with comprehens ive
`
`environmental monitoring can be pa rti cul arl y valuable in validating
`
`the ase pti c processi ng of st eril e solu tio ns. suspensions, and
`
`powders. Filling liquid media, as part of validating the process ing
`
`of powders. may necessitate use of equipment and/or processing steps
`
`that would otherwise not be attendant to routine powder operations.
`
`However. such additional efforts are valuable and impo rtant in
`
`characteri zing exposure of powders to contamination.
`
`- 22 -
`
`Apotex v. Abraxis -lPR2018-00152, Ex. 1009 , p.24 of45
`
`
`
`Severa 1 ques t ions about medi a fi 11 s have been ra i sed concern i ng
`
`contaminating equipment with media, frequency and number of runs,
`
`size of runs, the medium itself, environment~l conditions. and test
`
`results.
`
`1. Contamination with media - SOllie drug manufacturers have
`
`expressed concern over the possible contamination of the
`
`facility and equipment with the nutrient media during media fill
`
`runs. However. 1f the medium 1s handled properly and is
`
`promptly followed by the cleantng, san1t1z1ng. and, where
`
`necessary. sterilization of equipment, then media fill
`
`operations should not compromise the quality of product
`
`subsequently processed using the same facility and equipment.
`
`2.
`
`Frequency and number of runs - When a process is initially
`
`validated each separate media fill should be repeated enough
`
`times to assure that the resul ts are consistent and meaningful.
`
`This is important because a single run may be faulty, and widely
`
`divergent results of mult1ple runs may signal a process that is
`
`not in control.
`
`FOA believes that. in many cases, at least
`
`three separate runs are needed: this minimum number has been
`
`recognized as a general validation principle in the industry
`
`(Refs. 6 and 7). The frequency of additional media fills needed
`
`after initial validation has been completed will vary depending
`
`- 23 -
`
`Apotex v. Abraxis - IPR201 8-00 152, Ex. 1009,