Response to the FDA's Concept Paper: Sterile Drug Products Produced by Aseptic Processing
As Reviewed by the Controlled Environment Testing Association (CETA)
This document is intended to review part IV: "Buildings and Facilities" and offer suggestions for improvement and accuracy. The opinions expressed in this document reflect those of the members of CETA, a professional organization dedicated to ensuring a high standard of quality assurance within the controlled environment testing industry.
- Lines: 47, 172, 192, 284, 299, 300, 304, 311, 327, 368, 1123, 1783
Document Says: both "Cleanroom" and "Clean Room" interchangeably throughout
Comments: "Cleanroom" has been the industry-accepted word for about 20 years. All IEST documents, the recently sunsetted Federal Standard 209 and ISO Standards use "Cleanroom." The definition on Line 1730 is correct.
Suggested Changes: Change all references of "Clean Room(s)" to "Cleanroom(s)." - Lines: 47-48, 75-76, 140
Document Says: "...clean room classifications under dynamic conditions"
Comments: IEST-CC-RP-006.3, the recently sunsetted Federal Standard 209 and ISO 14644-1 all define cleanroom "occupancy states" as "as-built," "at-rest" and "operational." ISO 14644-1 Section 3.3 requires the occupancy state to which the classification applies to be included in the designation of the cleanliness class.
Suggested Changes: Change "dynamic" to "operational" and "static" to "at rest", where appropriate. In addition, include the definitions from ISO 14644-1 as follows:
Operational: "Condition where the installation is functioning in the specified manner, with the specified number of personnel present and working in the manner agreed upon."
At Rest: "Condition where the installation is complete with equipment installed and operating in a manner agreed upon by the customer and supplier, but with no personnel present." - Lines: 80-88, 93, 110, 157-162, 175-177, 266-270, 288, 1101-1110, 1395, 1400-1405, 1531-1536, 1613-1618, 1735-1736
Document Says: TABLE 1--"Clean Area Classification: 100, 1000, 10,000, 100,000"
Comments: Table 1 references ISO 14644-1, but uses classifications (Class 100, 1000, etc.) that do not exist in ISO 14644-1. The way the author uses the classifications throughout the paper is not consistent with either ISO or EU (Grade A, B, C & D) classifications. Suggest the inclusion of "US Customary" classifications in parenthesis with the list of ISO Classifications.
Suggested Changes: Change first column in Table 1(Clean Area Classification) to ISO Class 5 (US Customary Class 100), ISO Class 6 (US Customary Class 1000), ISO Class 7 (US Customary Class 10,000) and ISO Class 8 (US Customary Class 100,000). Change further references throughout document to include the reference to particle size at a specified class.For example: ISO Class 6 (US Customary Class 1000) at .5 µm.
- Lines: 82-88, 109, 135 (footnote #3), 170, 1387
Document Says: "> 0.5 um particles/ft." and "> 0.5 um particles/m3"
Comments: All measurement units should be consistent throughout the paper. In both of these examples, the use of units are mixed as particle size is given in microns (Metric) and counts are given in particles-per-cubic-foot (English). Also, there are several references throughout the paper to inches of water.
Suggested Changes: Use the Metric system of measurement consistently for global harmonization. - Lines: 118-120
Document Says: "Nonviable particulate monitoring with a remote counting system is generally less invasive..."
Comments: Nonviable particulate monitoring with a "remote counting system" may be generally less invasive than portable particle counting units, but long sampling tubing runs (over 10 feet) from the sampling point to the detecting sensor may lead to fallout of larger-sized particles, entrapment and fragmentation of particles. Currently, very little remote sampling is done. Most sampling is being done at locations with information being sent electronically to the building monitoring system (BMS).
Suggested Changes: The use of remote manifold sampling systems that have long runs of tubing from the critical location to the detecting sensor should be avoided. A well-designed manifold system relies on short runs of sampling tubing. - Lines: 131, 353, 360
Document Says: "HEPA filtered laminar flow air"
Comments: Laminar flow is defined as a flow having a Reynold's number less than 2,000. This will never occur in a cleanroom environment. The IEST and Federal Standard 209E made the transition from using "laminar flow" to "unidirectional airflow."
Suggested Changes: Change all references to "laminar flow" to "unidirectional airflow." - Lines: 131-135 (and footnote #3)
Document Says: "A velocity from 90 to 100 feet per minute is generally established with a range of plus or minus 20% around the setpoint."
Comments: The statements in lines 131-135 are nearly perfect as written...as long as footnote #3 is taken ONLY as a guidance value. The 90 to 100 fpm +/- 20% should be a "guideline value" (in MCA terminology) or "informative" (in ISO terminology). The magic of 90 fpm has been known to be a fallacy within the cleanroom industry for over 25 years. This value has been used since the 1960's when it was mentioned in the "Non-Mandatory Appendix" of Federal Standard 209A. It was based on a simple calculation that a 5 um particle would stay airborne (settle less than 2 feet) over a distance of 20 feet in a horizontal flow cleanroom. The true tests are airflow pattern testing and viable/nonviable sampling.
Suggested Changes: Keep first sentence in Footnote #3 as is. Second sentence to read: Higher or lower velocities may be more appropriate depending on the specific process. - Lines: 137-144
Document Says: "Air Visualization" studies demonstrating unidirectional airflow and sweeping action over and away from the product under dynamic conditions should be conducted."
Comments: a.) Airflow smoke pattern testing is probably the most important component of good cleanroom velocity testing. Although, there have been cases where firms have used a method of generating "smoke" that creates a water vapor (CO2 or similar). Anytime the "smoke" is heavier than air, it will "fall out." Video clips exist that show rooms where air passes along the floor, up the wall and across HEPA filters before re-entering the air stream when observing with a Glycol-based smoke. When using a CO2-based smoke on the same system, it appears as though the unit has good unidirectional airflow. CO2-based units can be used to show "vertical unidirectional airflow" even in a room with NO airflow (or worse, in a room with turbulent flow including updrafts). It is important to avoid the use of smoke generating systems where the effluent is heavier than atmosphere (such as CO2-based systems). Consideration should be given to "Bracketing" velocities. This is the practice of testing at the upper and lower limit of the acceptable range of velocities, which would allow operation at velocities that fall anywhere in the bracketed range.
b.) ISO 14644-2 uses the term "Airflow Visualization" for "smoke."
Suggested Changes: Consider the above suggestions for content. Also, change "air pattern" or "smoke" to "airflow visualization." - Line: 168
Document Says: "...of generally at least 0.05 inch of water (with doors closed)."
Comments: The requirement of 0.05 inches of water to adjacent areas of lower classification is unobtainable in many applications especially in controlled areas like ISO Class 7 gown rooms to ISO Class 8 anterooms without additional HEPA filtration or engineering controls. A 0.05-inch pressure from the aseptic area to controlled areas is necessary, desirable and easily obtainable, while a cascading 0.02"-0.05" between controlled areas of lower classification is sufficient to prevent particulate ingress.
Suggested Changes: Consider the suggestion above for content. - Lines: 175-177
Document Says: "...20 air changes per hour."
Comments: This rate is historic and without scientific basis. The air cleanliness will depend on the aerosol generation rate (i.e., powder operation) as well as the ventilation efficiency. The dirtiest locale in the room will determine the air cleanliness class. High ceilings will reduce the air exchange rate without an adverse effect on performance.
Suggested Changes: a.) Change to "An adequate air change rate should be established to maintain the desired air cleanliness. Many factors must be considered when establishing this air change rate," or b.) Room Recovery Test is a true performance test of how quickly the room actually recovers from an upset condition and may be used to satisfy room requirements. - Lines: 218-220
Document Says: "...filters that should be integrity tested are those installed in dry heat depyrogenation tunnels..."
Comments: HEPA filters (media and seals) cannot withstand the temperatures within the "hot zone" of depyrogenation tunnels. The binder burns off the media leaving a glass residue similar to the mantel of a gas lantern. The expansion and contraction of the metal frame destroys the seals between the filter frame and the structure as well as the media to the frame. Any oils (PAO or DOP) used to challenge filters in zones that will be heated will outgas. The hot oil vapor can deposit on colder surfaces (glass vials) due to thermophoresis. The aerosol challenge required to test the filter integrity may become a product contaminant. It is for this reason that, in cases where hot zone HEPA filters are tested routinely and successfully, the preferred method is not the integrity test method referenced in this document.
Suggested Changes: Change statement to, "Any HEPA filters that may be subjected to extreme temperature conditions should not be tested for integrity using an aerosol challenge and a photometer. The air cleanliness for the entire path of the vials should be measured and verified to meet ISO Class 5; at-rest; @ 0.5 um." This requires particle concentrations to be measured in the depyrogenation zone. - Lines: 222
Document Says: "One recognized method of testing...is use of DOP aerosol challenge. However, alternative aerosols may be acceptable."
Comments: PAO is by far the most commonly used aerosol today for filter integrity testing. DOP is only used in a few FDA regulated manufacturing facilities. It is still required where the firm does not want to risk changing to PAO for fear of "proving equivalence" to the existing DOP aerosol.
Suggested Changes: Delete the first two sentences on lines 222 and 223. Change sentence 3 to "Poly-alpha-olefin (PAO) aerosol is a recognized aerosol challenge to test the integrity of HEPA filters." - Lines: 229-243
Document Says: "...99.97% of particulates greater than 0.3 micron in diameter..."
Comments: This entire paragraph is very weak and in many cases incorrect. A.) The statements regarding particle size are true only for performing factory filter efficiency tests. HEPA filters should be rated at their Most Penetrating Particle Size (MPPS). Assume a marginal hole (leak) in HEPA filter medium of 200 um in diameter. Particles 10 um in size will pass through the defect as easily as 0.1 um particles. Challenge aerosol size distribution is not a consideration when scanning filters for integrity. In fact, if the challenge aerosol is near the MPPS of the filter, penetration will be detected downstream of the intact filter media, masking the presence of minor media defects. IEST 's recommended practice for HEPA filters (IEST-RP-CC 001.3) very carefully specifies thermally generated particles when using a test aerosol for filter efficiency testing. That aerosol is actually approximately 0.19 micrometer count mean diameter or .3 micrometer light scattering mean geometric diameter. Specifying the use of a Type "C" filter per the IEST-RP-CC 001.3 greatly reduced the potential for leakage across the entire filter media when using a near MPPS particle (such as a thermal aerosol generator). We do not efficiency test filters in the field, we integrity test them. It therefore is important that we use filters that are properly efficiency tested at the factory. B.) Line 232 states, "The DOP challenge should introduce the upstream of the filter in a concentration of 80 to 100 micrograms/liter..." This concentration is unnecessarily too high and was established in the 1950's for early generation aerosol photometers. Modern photometers are over 10 times as sensitive as the original equipment. IEST-RP-CC-034.1 Section 6.2.1 states, "A concentration of approximately 10-90 ug/l of air is an adequate challenge." The NEBB "Procedural Standards for Certified Testing of Cleanrooms" Second edition 1996, Section 6.2.2.2.a states, "adjusted to produce an upstream concentration of 10-20 ug of DOP per liter of air." C.) The document states, "...photometer probe at a sampling rate of at least one cubic foot per minute." The sample rate should be 1.0 cubic foot per minute. The reported leak size is inversely proportional to the photometer sample rate (i.e., a 1% leak measured with a 1.0 cfm photometer would be only 0.5% with a 2 cfm photometer. D.) Scanning of the filter face and frame should be conducted within one inch of the face of the filter. At distances farther away, the leak may be diluted by the surrounding HEPA filtered air and may produce slightly less than the actual or true penetrations through the filter.
Suggested Changes: Consider deleting the entire paragraph or address the above concerns. Also, the document does not address the IES-RP-CC-034.1 Section 6.2.3 "Aerosol photometer total leakage test method." A paragraph should be added to discuss the testing of filters in ducts or air handling units, where filters are challenged with a uniform challenge of 10-20 ug of PAO per liter of air. - Lines: 247-250
Document Says: "...particles ranging in size from one to three microns...rating of the filter."
Comments: The count mean diameter of both thermally generated and Laskin nozzle generated PAO and DOP is less than 0.3 um (in the submicron range-not one to three micron range).
Suggested Changes: Change to "...particles ranging in size from 0.1 um to 1.0 microns." - Line: 294
Document Says: "Airlocks and interlocking doors facilitate better control of air balance throughout the aseptic processing area."
Comments: See below for suggested improvement in diction.
Suggested Changes: Change to "Airlocks and interlocking doors are critical to maintain appropriate pressure differentials throughout the aseptic processing area." - Line: 1785
Document Says: "Laminarity" definition
Comments: In past cleanroom literature, the term was laminar flow. Now, the industry uses "unidirectional airflow."
Suggested Changes: Replace laminarity definition with unidirectional airflow definition: "Air that flows in a single pass in a single direction through an air device or clean zone with generally parallel streamlines." - Line: 1803
Document Says: "ULPA filter" definition
Comments: Use the IEST-RP-CC-034.1 definition.
Suggested Changes: Change to "ULPA (Ultralow-penetration air) filter-An extended-medium, dry type filter in a rigid frame having a minimum particle collection efficiency of 99.999% (i.e., a maximum particle penetration of 0.001%) for particles 0.1 to 0.2 um when tested in accordance with Methods of IEST-RP-CC-007."
