Vous êtes sur la page 1sur 10

Title: Sterilization Process Validation

Manual Number: 039

Prepared by: Date: Supersedes:
Checked by: Date: Date Issued:
Approved by: Date: Review Date:

Manual: 039

Sterilization Process Validation

1 Purpose

To provide guidelines for the validation of sterilization processes used in the manufacturing
activities for drug products or active Pharmaceutical ingredients (API) and also to outline
recommendations on how to achieve compliance.

2 Scope and Applicability

This Guideline is applicable to all manufacturing Operations, sites, functions and

departments undertaking work, or providing support services, required to meet Good
Manufacturing Practice (GMP) or, in the absence of a GMP standard, International
Organization for Standardization (ISO) standards.

This Guideline is applicable for sterilization processes used to produce sterile drug products,
components, equipment and other ancillary items required to be sterile for use in the drug
manufacturing process.

3 Definitions

3.1 D value

The time in minutes at a specific temperature required to reduce a surviving microbial

population by 90%, i.e. a one-logarithm reduction.

3.2 F0

The time required at any given temperature between 100°C -140°C that is equivalent to the
sterilization effect of steam at 121.1°C (250°F). Assumes a Z value of 10°C.

3.3 Z value

The number of degrees Celsius required to change the D value by a factor of ten.

3.4 Sterile

State being free from viable microorganisms. In practice no such absolute regarding the
absence of microorganisms can be proven.

3.5 Sterilization

Validated process used to render a product free of all forms of viable microorganisms. In a
sterilization process, the nature of microbial death is described by an exponential function.
Therefore, the presence of microorganisms on any individual item/container can be
expressed in terms of probability. While the probability may be reduced to a very low
number, it can never be reduced to zero.

3.6 Overkill cycle

A sterilization cycle that provides a 12-log reduction of a resistant Biological Indicator (BI)
with a known D value of not less than 1 minute. A typical cycle would provide a minimum
of 15 minutes at or above 121.1°C. This method is appropriate for heat stable products and

4 Responsibilities
All sterile manufacturing sites or its contractors are responsible for ensuring that
sterilization processes used to produce items are properly validated.

5 Guideline

Validation of processes used to sterilize drug products and equipment are the most critical
validation activities undertaken. Common elements in the validation of any sterilization
process include:

• Sterilization Cycle Development

• Biological and Physical Measurement Controls
• Empty Chamber Studies
• Loaded Chamber Studies
• Routine Use/Ongoing Monitoring
• Validation Maintenance/Change Control/Revalidation
The sterilization method chosen depends on the application. The following methods are
typically available:

Method Typical application

Steam For the sterilization of fluids in ampoules, vials etc, or the
sterilization sterilization of processing equipment, reactors, preparation
tanks, solution delivery piping, etc. In general, sterilization
through the application of saturated steam under pressure is the
preferred method of sterilization. The principles apply to SIP
processes as well.
Sterilization by Used for those products that cannot be sterilized due to the
filtration heat sensitivity of the product or where heat labile packaging
is chosen since it provides a distinct patient benefit. Not the
preferred sterilization method.
Dry heat Used to sterilize/depyrogenate containers (ampoules, vials,
sterilization and etc.), pharmaceutical raw materials and processing equipment.
Depyrogenation The use of dry heat has little application for the sterilization of
pharmaceutical drug products.
Radiation To sterilize packaging equipment, consumables, garments etc.
sterilization that are difficult to sterilize using steam or other methods.
Radiation sterilization is mostly used for medical devices.

5.1 Steam Sterilization

Steam Sterilization is the most common type of sterilization employed in the

pharmaceutical manufacturing environment. The principles of steam sterilization
are applicable to processes conducted within autoclaves as well as sterilization-
in-place (SIP) processes.

For the sterilization of fluids in e.g. vials and ampoules, a fluids load autoclave
cycle is used. The steam (or superheated water) is used as a heat transfer medium
to heat the contents of the vials/ampoules. The moisture required for sterilization
is derived from the contents in the vial/ampoule.

degradation of the product does not occur.

Product Specific Method

The Product Specific Method (PSM) applies to sterilization cycles where an

overkill approach is not possible because the product cannot withstand the
temperatures and/or exposure times required for an overkill cycle. The PSM is
more applicable to the validation of heat sensitive fluids, for example, where the
sterilization cycle is developed to adequately destroy the microbial load and yet
not result in product degradation.

The PSM relies on determining the population and heat resistance of the
environmental, product, or items-to-be-sterilized bioburden along with ongoing
monitoring and control over the bioburden during routine operation. A PSM
cycle would deliver less lethality than an overkill cycle, but would still deliver a
Sterility Assurance Level of at least 10-6 (for the most heat resistant bioburden in
the product).

In the PSM, once the population of the overall bioburden and the heat resistance
(generally only for spore-forming environmental or product/item isolates) has
been determined, these values (plus additional safety margins based on
professional judgment, the extent of the bioburden data, and the degree of product
bioburden testing that will be conducted on an ongoing basis) are used to
determine the lethality necessary to achieve a minimum SAL of 10-6.The safety
margin selected inversely correlates to the frequency and magnitude of ongoing
tests conducted for bioburden population and resistance. For example, if the
observed worst-case product spore bioburden resistance is 0.3 minutes and, for
lethality determination, a D-value of 0.4 minutes is selected, then extensive
ongoing bioburden resistance testing would be necessary. However, if a D-value
of 1.0 minute were selected for lethality determination, then minimal ongoing
bioburden resistance testing would be necessary.Ongoing monitoring of
bioburden population must be conducted.

Once the bioburden data has been used to determine the lethality required for the
PSM derived cycle, the attributes (population, resistance) of the BI challenge
system can be determined. The semi-logarithmic survivor curve equation is used
for determination of both the required cycle lethality and the attributes of the BI

When the resulting log reduction for the BI challenge system is compared to the
log reduction of the typical bioburden organism (whose D value is substantially
less) the log reduction delivered for the actual bioburden is significantly higher
than the demonstrated log reduction using the BI challenge system.

The F-value for steam sterilization (F0) is a measurement tool used to demonstrate
accumulated lethality and must be used in both validation and routine monitoring of cycles
to demonstrate that acceptance criteria for F0 has been met.

For terminally sterilized drug products validated using the PSM approach,
reductions in cycle parameter set points during the validation are less common.

Terminally sterilized drug product cycles generally include an upper limit for the

minute of the exposure stage.

The difference between the control probe, recording chart probe and independent
sensor (usually a thermocouple) during the exposure stage should not exceed + 1.0°C. Loaded Chamber Temperature Distribution

This study must be included to demonstrate that the equipment loading

patterns do not significantly change the chamber temperature distribution
within the chamber. Typically thermocouples are distributed throughout the
chamber (not in contact with load items) as for the Empty Chamber Temperature
Distribution study and cycles may be run using both the maximum and minimum

Loaded Chamber Temperature Distribution studies should meet an acceptance

criteria for the temperature to be within + 1°C of the mean loaded chamber
temperature after one minute of the exposure stage.

The difference between the control probe, recording chart probe and independent
sensor during the exposure stage should not exceed + 1.0°C. Heat Penetration Studies

Loaded chamber heat penetration studies must be performed to demonstrate

that the pre-required time at temperature criteria are met for the loads being
validated. The heat penetration locations to be monitored are assessed
using both temperature probes and BIs. The thermocouples and the BIs should
be placed at the same locations wherever possible. Special emphasis is placed on
those locations identified during cycle development studies as being difficult for
steam to penetrate/difficult to heat (cold spot determination). Such locations
typically include the interior of hoses, filter housings, large objects, filling
apparatus and items with multiple layers of protective wrapping.

The placement of temperature probes and BIs within the load must not enhance the
penetration of steam into the load item. Where the load includes multiple items of
the same configuration in the load (e.g. bags of stoppers), BIs should be placed in a
second item adjacent to that containing the thermocouple.

This is to prevent the presence of the thermocouple from enhancing the penetration of
steam to the BI.Where items in the load are unique, the BIs must be placed in the load
item near the probe and precautions taken to prevent enhanced steam penetration.

Heat penetration cycles performed, as part of an initial validation exercise

must be repeated several times, e.g. three times,to demonstrate consistency.

These studies should be performed using established load patterns, though a

minimum and maximum load may be used to represent each particular load
pattern for qualification purposes. The purpose of the heat penetration study is to
document that the load items (including the cold spot) receives the minimum
required pre-determined time at temperature/F0.

a load cold spot due to the relatively uniform temperature rise across the load.
Using defined maximum and minimum load patterns may be useful in elucidating
the presence of a product load cold spot.

Once the container and product load cold spots (if any) have been determined, consistency
of the cycle must be demonstrated by repeat runs with temperature probes and BIs located
at the container and load cold spots. The purpose of consistency study is to document that
any load or container cold spot receives the minimum required pre-determined F0and/or
time at temperature.

5.1.3 Demonstrating Biological Lethality of the Cycle

The use of BIs to demonstrate biological lethality must be included in both porous
(equipment or hard goods) cycles and fluids load cycles, whether validating using the
overkill cycle method or the PSM.

BIs used in steam sterilization are bacterial spores, typically Geobacillus

stearothermophilus, having a defined heat resistance that is significantly greater than the
native bioburden of the items/product to be sterilized. For PSM cycles, other BI challenge
systems such asClostridium sporogenes, Bacillus coagulans or Bacillus atropheus may be
used. BIs provide a direct measure of biological lethality and allow the targeted placement
of highly heat resistant microorganisms in situ. BIs respond to both heat and moisture and
therefore provide proof that the required sterilizing conditions have been achieved
(temperature probes alone cannot do this).

The destruction of large numbers of these resistant organisms supports the efficacy of the
sterilization process for the total destruction of any native bioburden organisms present.
BIs are available commercially as pre-prepared spore strips or coupons, self- contained
units, or as spore suspensions. The heat resistance of BIs is characterized by the specific D-
value and Z-value for the preparation determined under defined conditions.

Spore strips or coupons should be placed within the load at locations where steam
penetration is deemed to be most difficult (sometimes referred to as the ‘cold
spot’). The manufacturer of the strips, coupons or self-contained unit provides
information on the number of organisms present on the inoculated carrier
(population) and an expected resistance level (D-value) under defined conditions.

Spore strips or coupons should only be stored, used and incubated as directed by the
manufacturer. They should never be immersed in liquid or altered or abused in any way with
since this may dramatically change their resistance from the labeled value. The end user
should verify the population of commercially obtained spore strips or coupons.

Self-contained Bis include both the organism and the culture media in a single unit. The end
user positions the BI in the load, runs the cycle, recovers the unit and then incubates
according to the manufacturer’s instructions. Where these units are used in accordance with
the manufacturer’s instructions, the population and D-value provided may be used directly,
though it is good practice to determine the population as part of the quality control of these

Spore suspensions can be used to directly inoculate items to be sterilized. The suspension is

In addition, it is common to use a second sterilizing grade filter immediately
before final filling.

The manufacturing process must be designed to minimize the pre filtration

bioburden. Products that are subject to microbial proliferation should be held for
as short a time as possible. If long holding times are anticipated consider the
options of sterile filtering the solution into a sterile holding tank or holding the
product in either a chilled or a heated state to reduce the probability of an
increased bioburden. Sterile holding tanks and any contained liquids must be
held under positive pressure or appropriately sealed to prevent microbial

Filtration validation should be product-specific. The use of a “matrix approach”

or “family approach” allowing grouping of products or the selection of a worst-
case product to represent a number of products is discouraged due to the
extensive justification needed. Even with such justification, the acceptance by
regulatory authorities would be on a case-by-case basis and in the absence of any
official guidance for industry.

All sterilizing grade filters must be integrity tested before use (preferably
after sterilization) and then again following use. The integrity test parameters
must be clearly defined and, along with the test limits, be correlated to the
microbial validation of the filter.

5.3.1 Validation of Filtration Process

The validation of sterile filtration processes must demonstrate microbial

retention as well as filter compatibility and filter extractable relative to the
solution being sterilized. The microbial retention studies required are highly
specialized and the industry in general, contract this work out to the laboratories
of the filter manufacturers.

Filtration validation is process specific and process parameters must be

provided to the filter manufacturer in order to assure that the laboratory-
based validation conditions reflect both actual and elements of worst-case
use conditions in our facilities. Filters must never be subjected to conditions
beyond those mandated by the filter manufacturer. Process parameters such
as pressures, maximum use time, hydraulic shock, flow rates etc. as well as
product solution characteristics such as pH, temperature, osmolarity, etc.
are some of the parameters that must be factored into the validation
exercise. While the actual work to validate microbial retention may be contracted
out, the responsibility for the validation lies with the sponsor company .A copy of
the validation documentation must be held at the sponsor site.

5.3.2 Microbial Retention Test

Sterilizing grade filters must pass a microbial retention test using

Brevundimonas diminuta as the challenge organism. The requirements for
culturing the organism as well as the conditions of the test are rigorous and highly
specialized. It is particularly important to perform this test by directly inoculating

5.4 Dry Heat Sterilization and Depyrogenation

Dry heat sterilization processes are typically used for equipment parts while dry
heat-based depyrogenation processes are used for glass containers for drug
products. Dry heat sterilization and depyrogenation can be performed in ‘batch’
dry heat ovens or continuous sterilizing tunnels. Rubber closures are also subject
to a validated depyrogenation process using a washing/rinsing process - See
section 5.3.4 below). There is limited use for dry heat in sterilizing drug products,
though raw materials are sometimes sterilized using dry heat. The validation of
these dry heat based cycles follows many of the general principles outlined for
steam sterilization. Concepts such as equilibration time, temperature mapping,
heat penetration, biological lethality and worst-case cycle parameters (shortened
times, lower temperatures, faster belt speeds, etc.) are relevant, though in a
different context.

It is common practice to determine FH values for dry heat processes in a similar

manner to F0 for steam sterilization. However, most applications of dry heat are
to depyrogenate and not just to sterilize. The depyrogenation process is not as
well understood as the sterilization process. Some evidence indicates that
inactivation of endotoxin is a second-order reaction whereas the standard use of
FH, D and Z values assumes a single-order reaction. Therefore for depyrogenation
cycles there is even more reliance on the biological proof (inactivation of
endotoxin) as opposed to only the accumulation of physical data. Even with these
limitations it is valuable to collect the FH data to demonstrate reproducibility.

The temperature range allowed in both empty and loaded chamber studies for dry
heat cycles is significantly greater than for steam sterilization. In particular, for
empty chamber studies, USP 29 states that ± 15ºC is a typical acceptable
temperature range when the oven is operating at not less than 250ºC. For loaded
chamber studies, a ± 5ºC temperature range is generally achievable.

Depyrogenation processes must be validated to achieve a minimum 3-log

reduction of bacterial endotoxin. The endotoxin must be spiked onto the item
in a liquid state and allowed to air dry prior to being subject to the
depyrogenation cycle. Recovery studies and appropriate controls are a
necessary part of these studies.

5.4.1 Dry Heat Ovens

The validation of a dry heat process in an oven, whether it is used for sterilization
or depyrogenation, is similar to the validation of a steam sterilization process.
Biological indicators, typically Bacillus atrophaeus spores (for sterilization
processes) or bacterial endotoxin (for depyrogenation processes) must be
used in conjunction with temperature measurements. A successful endotoxin
challenge in the validation of a depyrogenation process assures that
sterilization has also occurred and there is no need for additional studies
with bacterial spores.

The load size and configuration are very important in dry heat processing. These
must be carefully designed and duplicated between the validation studies
and routine operation.

Blowers/Fans - for proper air balance and the differential pressure
between the tunnel and the room environment
Room pressure differentials Empty Tunnel Temperature Distribution

Temperature probes (thermocouples) are distributed across the width of the

tunnel and a temperature profile is produced across the conveyor and along the
tunnel during operation. A reasonably uniform temperature profile is expected. A
probe needs to be located next to the controlling sensor. Note the come-up time to
temperature and cool-down times, as these should be consistent in an empty
tunnel. Loaded Chamber Heat Penetration

In dry heat tunnel applications heat penetration studies must be performed.

Thermocouples must be placed in contact with vials distributed across the tunnel. The
temperature of the vial is usually lower than the air temperature.

Test should be performed with all vial sizes. Endotoxin spiked vials should be
located adjacent to the penetration thermocouples. Heat penetration studies,
conducted as part of the initial validation, must be repeated several times (e.g.
three times) to demonstrate consistency. Consider the loading arrangement of the
tunnel i.e. at the start, middle and end of a run.

The temperature profile for a specific load should show good reproducibility.

5.4.3 Demonstrating Biological Lethality

The use of BIs to demonstrate biological lethality must be included in the

validation approach.

For dry heat sterilization the indicator of choice is Bacillus atrophaeusspores.

However, dry heat is also used to depyrogenate. In this case endotoxin (E. coli) is
used to validate the depyrogenation activity. If endotoxin is used then Bacillus
atrophaeusspore strips are not required as endotoxin is more heat resistant than
Bacillus atrophaeus. These biological studies can be performed at the same time
as the loaded chamber temperature mapping. If this is done the carriers should be
located as close as possible to the thermocouples.

5.4.4 Washing Process for Depyrogenation

In some cases, the depyrogenation of items that cannot be dry heat sterilised (e.g.
rubber stoppers) is carried out in a washing/rinsing process. It must be demonstrated
that a 3-log reduction in the endotoxin concentration on spiked stoppers can be achieved.

Recovery studies on the spiked stoppers must be performed. The stoppers must be
subsequently sterilized using a validated steam sterilization cycle.

5.5 Radiation Sterilization

5.6 Validation Maintenance/Change Control/Revalidation

Validation of sterilization processes is not a one-time exercise. Validation

maintenance is a phrase that describes a number of activities that support the
ongoing validated state of a process. In the case of sterilization processes, routine
bioburden monitoring, preventive maintenance, calibration, cycle review and
approval and annual reviews are activities that make up validation maintenance.

Bioburden monitoring must take place for every batch of aseptically

produced drug product. The preventive maintenance program should provide
clear instructions on reporting any unusual observations or equipment
breakdowns/mechanical failures so that an evaluation can be made on impact to
the validated state. Cycle review and approval should be in accordance with
detailed SOPs that have a direct traceability to the qualification and validation
activities, set point parameters and acceptance ranges.

Chamber vacuum leak testing, air detector device performance (if so equipped),
Bowie-Dick type testing for steam penetration, thermometric testing for small
loads, and other types of testing or review that may be required by local
regulatory authority expectation or requirement should be understood and
implemented as appropriate. For example, steam quality testing is a requirement
of the UK authorities. These tests are also mentioned in the PDA Technical
Monograph #1 and FDA’s Guideline on Drug Products Produced By Aseptic

The change control program should ensure that technical experts and the Quality
Assurance Function assess any planned changes to the equipment, process, loads,
procedures or documentation as to whether the qualified or validated state may be
impacted by the change. Any additional work necessary to demonstrate the
ongoing validated state should be reviewed and approved as part of the change
control process. Sterilization processes are critical processes. Be vigilant in
assuring that any changes or repairs are fully assessed for potential impact
via the change control process. Perform a periodic assessment of the
potential cumulative effect of changes that individually may not be
significant enough to prompt revalidation but, taken together, indicate a
need to re qualify or revalidate.

Sterilization processes must be revalidated at least annually in the absence of

any change-driven revalidation. The revalidation can be a subset of the original
validation work. Some acceptable approaches include:

• Single runs rather than 3 consecutive runs are sufficient in the absence of
recurring problems
• The selection of a worst case load pattern for revalidation

• The revalidation of each type of cycle (e.g. cycle for stoppers, cycle for
filling parts, cycle for products, etc.) but not the revalidation of each cycle
loading pattern.
• The revalidation of the worst case loading pattern and one selected
loading pattern, with the rotation of the remaining loading patterns at
subsequent annual revalidations.
• The revalidation of the empty chamber, maximum and minimum loading