Risk Based Particle Monitoring
In Pharmaceutical Manufacturing

Risk Based Particle Monitoring In Pharmaceutical Manufacturing

Airborne particle counters are an important tool used in the environmental

monitoring of pharmaceutical, bio-pharmaceutical, and healthcare facilities

worldwide. In addition to determining air quality as part of the facility

qualification, particle counters are required notification and data recording

tools used in confirming air cleanliness in critical areas where high-risk

operations are carried out.


Non-viable particle counters and their associated systems are part of a

larger environmental monitoring program that includes but is not limited to:


  • Non-viable particle monitoring of air
  • Viable particle monitoring of air, surfaces and personnel
  • Pressure differential monitoring
  • Temperature and relative humidity monitoring

Computer Based Monitoring Systems

Particle counters may be attached to computer based monitoring and alarm

systems that handles data collection, report generation, and alarm

notification. These systems are referred to generally as a “Facility

Monitoring System” (FMS) or more specifically as a “Non-Viable Particle

Monitoring System” (NVPMS) or “Environmental Monitoring Systems”

(EMS). There may be other names for these systems and they may be part

of a larger Laboratory Information Management System (LIMS). In fact,

with the release of EU GMP Annex 1 in August 2022, there is great interest

in a Contamination Control Strategy (CCS). The CCS is facility wide and

gathers environmental and operational data on all contamination events

and activities. With the explosion of digital technology, the lower costs for

data storage, and the increase in AI technology, data is becoming more

and more assessable. A CCS gathers data from all over the facility and

reports on trends that the data provides across multiple cleanroom areas

and operations. In fact the CCS along with Risk Assessments are major

parts of the EU GMP Annex 1 update.


For purposes of discussion in this article NVPMS is used to describe a

computer-based non-viable particle monitoring system. These systems

can monitor other environmental parameters such as temperature, relative

humidity, and room pressure differentials. Monitoring and data collection

provide proof that critical environments are in control prior to and during

manufacturing operations. Any equipment, system, or facility used in

manufacturing pharmaceuticals should be validated1 and maintained in a

validated state. The NVPMS system is more than several particle counters

connected to a computer with a database. The NVPMS is a project that

requires planning, qualification, and validation. A risk-based approach

should be used to determine the scope and extent of the validation. GAMP

5 and the V model are good starting points. A User Requirement

Specification based on a Risk Assessment is also a good starting point for

the specification and operation of a NVPMS.


The NVPMS project, as with any part of an environmental monitoring

program, should systematically assess risks through the review of

manufacturing processes and activities in relationship to equipment,

facilities, and personnel. The primary risk the NVPMS addresses is the

quality of air in the manufacturing process environment.


Though there are various approaches with varying complexity for assessing

risk, ICH Q9 and EU GMP Annex 20 are recommended to ensure QRM

principles are followed. Independent of the approach used, the following

items should be considered:

Considerations for the non-viable particle monitoring system:

  1. Identify high-risk operations for particle monitoring.
  2. Determine the optimal sample locations for monitoring.
  3. Establish a monitoring frequency with alert and action levels.
  4. Establish a system to verify the particle monitoring system is working

effectively.

5. Establish and maintain the validated state of the non-viable particle

monitoring system.

1. Identify high-risk operations for particle monitoring

ISO 14644-21 states that cleanrooms and clean air devices are to be

monitored in operation, with the monitoring locations based on a formal risk

analysis study and the results obtained during the classification of

cleanrooms and/or clean air devices.


Critical areas are where an exposed product is vulnerable to contamination

and will not be subsequently sterilized in its immediate container.

Typical high risk areas in manufacturing are:

? The filling zone (where containers are filled at the fill head).

? Stopper bowls (where stoppers are loaded and kept prior to filling

within the Grade A zone).

? Stopper insertion (the point stoppers are inserted into filled

containers).

? Loading areas for freeze drying (where partially closed containers are

loaded into freeze dryers (Lyophilizers).

? Isolator transfer devices.

? Transfer areas between Grade A and Grade B areas.

? Panels or access points (where operators are most likely to perform

interventions or load components such as stoppers).

? Wherever there are open ampoules, vials, or containers (turntables or

the exit of a sterilization tunnel).

? Where there are aseptic connections.

No alt text provided for this image
Figure 1 Isokinetic Sample Probe Placement Example


It is in these locations

within 1 foot (30.5 CM) of

the operation where

particle counting (when

suitable) should occur prior

to and during operations.


The NVPMS monitors air quality and addresses the risks associated with critical

locations by providing an early warning system to detect and prevent

contamination.


Per the FDA guidance3, remote airborne particle counters are best suited

for carrying out routine particle monitoring of critical locations. This is

because remote particle counters are typically attached to the equipment

and have a small footprint compared to portable particle counters. They are

installed in such a way as to not interfere with manufacturing operations nor

disturb the airflow. The selection of the precise monitoring location as well

as the attachment of the particle counter to the location is very important to

ensuring meaningful data is obtained.

Suitability of NVPMS components

The suitability of the remote particle counter, isokinetic sample probe, and

the associated vacuum system for the manufacturing operations must be

considered not only by cost but also durability, continued operations, and

availability of vendor technical support to mitigate unexpected downtime

events.


The remote particle counter needs to be compatible with the environment

being monitored. Points to consider for avoiding additional risks are:


a. The ability to withstand the operational environment: High heat

and excessive humidity may be present during the operations carried

out and may impact particle counter performance and service life.

Particle counter failure or marginal operation during production creates

additional risk.


b. Chemical compatibility: The frequency and nature of cleaning and

decontamination should be considered when selecting a remote

particle counter, probes, and fixtures. As the isokinetic particle sample

probe will be located within 1 foot (30.5 CM) of filling and closing

operations, it should be assumed the particle counter as well as the

associated sample probe, tubing, and vacuum system will be exposed

to the product being filled. Chemical compatibility as well as

occupational safety and health of maintenance staff when servicing all

these components should be considered, as product residue may be

present during service and calibration activities.


Cleaning solutions may be applied directly to the surface of the particle

counter or directly above it. These solutions may drip onto or

accumulate on the upper surface of the particle counter. The particle

counter should be constructed or installed in such a manner as to

prevent the ingress of these solutions. Where there is a risk of this

occurring, the particle counter should be of a sealed design or should

be installed inside an enclosure to protect the instrument from cleaning

and decontamination materials and activities.


Vapor decontamination cycles may also be frequently used, as some

operations may require decontamination of the particle counter and its

associated sample probe. Vapor decontamination compatibility may

also be an important consideration. Where most particle counter

designs are for the infrequent and accidental exposure to vapor

decontamination, frequent and intended exposure is in itself another

associated risk that may need to be addressed. Failure modes in

remote particle counters associated with vapor decontamination may

include loss of calibration that could go undetected until recalibration.

Particle counters intended for frequent and deliberate vapor

decontamination should have significant test data to support the

concentrations and methods used.

2. Determine the optimal sample probe locations for monitoring.

Once a general location has been selected by a risk assessment utilizing

subject matter experts (SME) and even vendor SME’s, the placement of

the isokinetic probe and particle counter are extremely important.


The isokinetic probe should be placed in a position that will best allow the

particle counter to detect contamination above the critical zone, specifically

from the risks previously identified in the risk assessment.

Probe placement example:

An example of this would be an operator access panel identified as being a

location where an operator would perform an operation or intervention.

The isokinetic probe is placed so that when the operation or intervention

occurs the particle counter detects increased particles generated by the

operation or intervention. The location of the sample probe allows for

operator access without hindering the operators work.

No alt text provided for this image
Figure 2: Isokinetic Probe Placement: Probe located so that elevated particle counts resulting from operators accessing critical areas will be detected. (Interventions or equipment set-up)
No alt text provided for this image
Figure 3: Sample grid at various heights

Testing with a portable particle counter during a simulation for this purpose is recommended. If the location for the critical zone is not obvious, a grid zone approach should be taken to identify the worst location in terms of particle counts. This is performed by using particle sampling grid patterns to test at locations considered and at various heights during actual or simulated interventions. The isokinetic probe and particle counter must be attached to the equipment and equipment design should also be considered to enable a secure location.

No alt text provided for this image
Figure 4: Sample probe height at fill head

The isokinetic probe and any fixtures used to attach the probe should not interfere with uni-directional airflow, equipment or personal and be easy to decontaminate. Airflow visualization is also a useful tool for probe placement .

However, ensure you use an air buoyant visualization medium to get a proper visualization of the actual airflow. Each location must be documented as to why this position is monitored with supporting information as to the placement of the isokinetic probe, based on the data from the grid zone testing.

Correct Probe placement is extremely important and a robust practical approach using a risk assessment should be followed. To align with GMP, the location should be orientated to the point of greatest risk as close to the critical zone as possible

EU GMP Annex 1 recommends that the following are accounted for:

  • Room layout
  • Equipment layout
  • Process considerations
  • Airflow patterns
  • Position of air supply and return vents
  • Air change rates
  • Consideration should be given to any unintended bias in the sampling

process

3. Establish a monitoring frequency with alert and action levels.

How often particle counts are taken and reported as well as action and alert

limits must be determined. Particle sample volumes and reporting

frequencies are often a source of confusion as some standards can be

misinterpreted. As particle data is reported in particles per unit of volume,

GMP Annex 1 and ISO 14644-1 indicate particle limits in particles per cubic

meter. Remote particle counter flow rates have flow rates of 1 CFM (28.5

LPM). This leads to some of the confusion as a 1 CFM particle counter will

take greater than 35 minutes to sample 1 m3. However, 35 minutes

between particle data counts is too long for operating personnel to respond.

Therefore, particle data should be recorded every minute with no

interruptions in particle counting for critical locations (Grade A). A rolling

total may also be used that updates every minute showing the accumulated

particles/m3 based upon the last 35 samples. Normalization of particle

count data in this application is not advised. However, the best practice is

to set up trend analysis based on X out of Y events using statistical process

control limits to identify “trends”.


Though GMP Annex 1 does state that for Grade B operations sample

frequencies may be reduced, it is commonly accepted to use the same

monitoring frequency as that of Grade A. Supplemental monitoring with

sequential sampling systems for non-critical locations may also be

deployed with consideration to the limitations of such sampling with respect

to particle losses in long lengths of tubing. For an application that requires

data for particles above 2μm, using a manifold system is not advised.

Manifolds are widely used in Semiconductor cleanrooms where the particle

sizes of interest are around 0.xn--1m-99b, and particle losses at that size range

are more acceptable than at 5μm sizes.


Alert and action limits can only be determined by observation of actual

operations, simulations, or media fills. These limits may be different from

the GMP Annex 1 or ISO Class limits. The PQ is where you should be

adjusting your alert and alarm limits. It is expected that these alarm limits

go under periodic review.


The data for 5μm particles can be problematic as the limits (per GMP

Annex 1) are low. However, occasional detection of low levels of 5μm

particles is acceptable. In Grade A, regular or frequent occurrences of low

levels of 5μm particles should be investigated. A common way to

address this is to set alert limits for 5μm particles and action limits at 3

events in any 10 minute period for Grade A operations. Also use the 0.5μm

data as the main trending data system as there are more particles

generated in the cleanroom at this size range and therefore more statistical

relevance for setting proper control limits.


This approach will enable a basic trend to be setup using meaningful data.

Trending particle count data for monitoring operations is highly

recommended. Looking for “one off events” in an environment where

transient particles are inevitable is not recommended, as this method is

detrimental to the overall operation of a monitoring system, especially in

aseptic processes of manufacturing sterile products. The trending method

is used in ISO 14644-1:2015 for cleanroom classification (certification)

based on a formula. Setting of appropriate alarm limits is highly

recommended, and appropriate alarms should be based on your cleanroom

environment and your aseptic process.

4. Establish a system to verify the particle monitoring system is working effectively.

No alt text provided for this image
Figure 5: NVPMS Alarm notification system and operator interface

It is required to have some type of indicator that the NVPMS is active, recording data, and reporting conditions related to the environments monitored. There should be some way to notify personnel of malfunctions or that alert and action limits have been reached.

This notification system should be visible to operators inside the manufacturing environment and any observers. It also should be such that it draws attention to problems in such a manner as cannot be ignored. This indicator may be an operator panel, light beacon with audible alarm, or some other visual and audible indicator. SMS or email to smart devices is another option for real time notifications. Alarms can be set on personal devices to notify of out of tolerance events in the cleanroom.

This indicator should also provide notification of particle counter operational problems such as low sample flow, sensor issues, laser issues, loss of instrument signal, or calibration date status. This indicator is extremely important as it provides a single point status if the NVPMS is working and the environment is suitable for operations to be carried out. A common application is a three-color light beacon with the following status being displayed:

  • Green: Indicating the NVPMS and all supporting components are

operational and environmental conditions monitored are within

operational specifications. (Environmental conditions may include

particle counts, temperature, relative humidity, and differential

pressure).

  • Yellow: Indicating environmental conditions have reached alert limits.
  • Red: Indicating the NVPMS or supporting components require

attention or that environmental conditions have reached Action

Limits. Red indicators often are attached to an audible indicator that

may be acknowledged or temporarily silenced by operator

interaction.

5. Establish and maintain the validated state of the particle monitoring system.

This process should be undertaken at the Performance Qualification (PQ)

stage. Before a NVPMS goes live it must be fully tested, and the PQ is the

last chance to iron out any issues before the NVPMS goes into service.

The PQ should not be a redo of the vendors Operational Qualification (OQ)

but should be driven by operational performance and process validation.

Examples of items to test before going live are:

  • Alarm and Alert limits. Adjust if necessary
  • Alarm notifications
  • SOPs based on alarm events should be fine-tuned and finalized
  • System recovery and backup
  • Review of all data and setup of a data transfer validation path
  • Operator training for use of the system
  • Operator responses to SOPs; adjust when necessary
  • Access to the system and security


Validation is ongoing and does not stop with the successful installation of

the particle monitoring system. Equipment must have maintenance

schedules and calibration must occur at predetermined times. As

components or equipment may fail and need to be replaced, spare parts

should be available and in stock to mitigate against long lead times.


Spare parts and repair/replacement SOPs should be on hand so that

operations continue uninterrupted. System re-validation should occur at

specific intervals to determine the NVPMS is operating correctly and

remains in the validated state. Calibration to current industry practices5 is

mandatory. Per ISO 14644-1:2015, calibration and re-calibration is to be

done in compliance with ISO 21501-4. Organizations performing this

calibration6 should have the proper equipment, procedures, and

documentation to support all aspects of this calibration4. The latest ISO

21501-4 standard was updated in 2018.

In Summary:

Always follow a Risk Based approach when designing the user

requirements of a NVPMS. Develop a strategic team of SME’s within your

organization. Hire SME’s if you do not have the experts on hand. Engage

with a NVPMS supplier that has SME’S available to assist and guide you

through the process. The URS developed based on QRM approach with

SME engagement will be a far more robust specification. The right URS

and team are a recipe for success with the NVPMS project.


Understand your process and use that knowledge to develop a Risk

Assessment that captures all known risks. The URS details will include

those risk attributes. A vendors Functional Design documentation should

capture all URS requirements or as much as possible.


Sample probe locations are critical in obtaining meaningful process data.

Meaningful data will enable better process decisions. Always use trending

as the process control method following X out of Y events. Alarm and Alert

limits should reflect the environment and your process and be fine-tuned in

the PQ.


The PQ is the last chance to fully test the NVPMS alongside your SOPs.

Auditors will always want to see how operators react to alarms and see that

SOPs are validated and workable to assist in determining alarm conditions.


Alarms should be periodically reviewed against the process to ensure they

are working for you and not against you.

No alt text provided for this image
Figure 6: NVPMS remote particle counter installation on filling machine

Ensure that the particle counters are suitable for the environment where

they are installed. Take into account cleaning/decontamination of the

particle counter enclosures and ease of access/removal for service and

calibration. Keep sample tubing lengths as minimal as possible and without

any bends if possible. Use a remote particle counter where communication,

power cables, and sample tubing are hidden and secure. Smart Brackets?

enable this and have the location ID associated with the Bracket. This

keeps data integrity intact, ensures there are no operator mix-ups in

location IDs, and validates that the data is coming from the right locations.


For more information visit our Knowledge Center at

www.golighthouse.com

Harald Flechl

Don't confuse guidelines with rules and experience with expertise.

1 年

Before deciding the point of sample see the definition of isokinetic! It is always disregarded: isokinetic sampling at the chosen location does not always correspond to the assumed 0.45 m/s (+/-20%) in direction and velocity of the airflow! Smoke studies will show the real flow direction and measuring the velocity at this point is really a challenge.

要查看或添加评论,请登录

社区洞察

其他会员也浏览了