The protocol summarizes the best practices to minimize microbial bioburden in a cleanroom environment and includes strategies such as environmental monitoring, process monitoring, and product sterility testing. It is relevant for manufacturing and testing facilities that are required to meet current good tissue practice standards and current good manufacturing practice standards.
A well-validated and holistic program that incorporates robust gowning, cleaning, environmental monitoring, and personnel monitoring measures is critical for minimizing the microbial bioburden in cellular therapy manufacturing suites and the corresponding testing laboratories to ensure that the facilities are operating in a state of control. Ensuring product safety via quality control measures, such as sterility testing, is a regulatory requirement for both minimally manipulated (section 361) and more than minimally manipulated (section 351) human cells, tissues, and cellular and tissue-based products (HCT/Ps). In this video, we provide a stepwise guide for how to develop and incorporate the best aseptic practices for operating in a cleanroom environment, including gowning, cleaning, staging of materials, environmental monitoring, process monitoring, and product sterility testing using direct inoculation, provided by the United States Pharmacopeia (USP<71>) and the National Institutes of Health (NIH) Alternative Sterility Testing Method. This protocol is intended as a reference guide for establishments expected to meet current good tissue practices (cGTP) and current good manufacturing practices (cGMP).
Implementing a strong microbial monitoring program through environmental monitoring (EM), process monitoring, and product sterility testing is a regulatory requirement for current good tissue practices (cGTP) and current good manufacturing practices (cGMP) in cellular therapy laboratories1. Additionally, the United States Food and Drug Administration (FDA) expects that the laboratory performing the quality control (QC) testing of the product should also employ facilities and controls comparable to those used for aseptic filling operations2.
This protocol has four main sections: 1) Aseptic practices, including personnel gowning, cleaning, and staging of materials; 2) EM, including viable air and surface cultures and non-viable particle air monitoring; 3) process monitoring, including settling plates and gloved fingertip sampling; and 4) product sterility testing via the compendial United States Pharmacopeia (USP) <71> method3 or the NIH Alternative Sterility Testing Method4. When used together, these measures can be an effective method for ensuring that a facility remains in a state of control.
The techniques described here are not novel; however, current standards from regulators and professional organizations lack detail, which has led to an absence of microbial monitoring or the implementation of non-standardized practices, particularly in academic centers where on-site manufacturing and product sterility testing are emerging at a rapid rate1,5,6. This protocol can be used as a guide to create a microbial monitoring and control program that meets regulatory requirements when used in conjunction with end-user validation and risk assessments.
1. Aseptic practices
Figure 1: Example of a BSC cleaning pattern. Working from back to front (or top to bottom), clean the BSC using overlapping wipes in the following order: the HEPA diffuser grill (the top of the BSC), the back wall of the BSC, both side walls of the BSC, the sash, and the work surface. Finally, wipe the sash of the BSC using 70% sIPA to remove any residual disinfectant. Please click here to view a larger version of this figure.
2. Environmental monitoring (EM)
Category | Media | Culture conditions | Culture observation | Results | ||||||
Environmental monitoring | TSA (viable air) | 30 °C-35 °C, air, for at least 3 days | End of incubation | The QA group of each facility should establish alert and action limits for each sampling type and location. Action limits for viable samples based on ISO classification can be guided using PIC/S 009-16 (Annexes) 18 and ISO-14644-1 7. Action limits for non-viable air samples are typically set to a percentage of the ISO limit (e.g., 99%). Alert limits for viable samples are typically set to a percentage of the Action limit or ISO limit (e.g., 95%). Refer to PDA TR-13 and USP<1116> for more details regarding establishment of alert and action levels and validating selected culture conditions 8,9. | ||||||
SAB (viable air) | 20 °C-25 °C, air, for at least 7 days | |||||||||
TSALT (viable surface) | 30 °C-35 °C, air, for at least 3 days | Representative images of EM plates are shown in Figure 2, Figure 3, Figure 4, and Figure 5. | ||||||||
SABLT (viable surface) | 20 °C-25 °C, air, for at least 7 days | |||||||||
Process monitoring | TSA (settling plate) | 30 °C-35 °C, air, for at least 3 days | For information only. Provides useful information in the event of an OOS investigation in response to a failed product sterility test. | |||||||
SAB (settling plate) | 20 °C-25 °C, air, for at least 7 days | See Figure 6 for an example of a positive settling plate. | ||||||||
Gloved fingertip sampling | TSALT | 30 °C-35 °C, air, for at least 48 hours days followed by 20 °C-25 °C for at least 5 days 19. | The acceptability criteria for GFS is <1 CFU/plate (i.e., no growth) as per PIC/S 009-16 (Annexes) 18. Acceptability criteria may be modified at the discretion of facility QA. | |||||||
Product sterility testing | TSB (USP<71>) | 20 °C-25 °C, air, for at least 14 days | Periodically throughout the incubation period (days 3, 5, 7, and 14) | No growth. | ||||||
FTM (USP<71>) | 30 °C-35 °C, air, for at least 14 days | |||||||||
iFA+ (NIH method) | 30 °C-35 °C, air, for at least 14 days | Monitoring automatically by the BacT/ALERT Dual-T instrument. Visual check of each bottle at the end of incubation for mold balls is strongly recommended. | See Figure 8 for an example of visible mold balls that failed to be automatically detected by the BacT/ALERT. | |||||||
iFN+ (NIH method) | ||||||||||
SAB (NIH method) | 20 °C-25 °C for at least 14 days | Periodically throughout the incubation period (days 3, 5, 7, and 14) |
Table 1: Summary of the recommended culture conditions and expected results. The culture conditions described here are recommendations based on a validated program used at the NIH. Each end-user is required to validate their own microbiology testing program. The microbial control and testing strategies may differ between institutes depending on variables including the facility design, facility flora, and product risk classification.
Figure 2: Growth on the TSALT plate. The TSALT surface sampling plate showing three CFUs of two distinct colony morphologies. Please click here to view a larger version of this figure.
Figure 3: Contamination of the TSALT plate during collection. The TSALT surface culture showing a single colony on the edge of the plate, indicative of poor aseptic handling during the sampling process. Please click here to view a larger version of this figure.
Figure 4: Culture obtained using a contaminated air sampling head. Example of a TSA active air sampling culture showing >100 colony forming units (CFU) of mixed morphologies. The pattern of growth indicates contamination of the sampling head. Please click here to view a larger version of this figure.
Figure 5: No growth on a TSA active viable air plate. TSA active viable air plate illustrating no growth following incubation. Indents from the active air sampler head can be seen in the image. Please click here to view a larger version of this figure.
3. Process monitoring
Figure 6: Growth on a TSA air settling plate. A TSA air settling plate illustrating a single colony of a contaminant cultured during passive air process monitoring in the BSC. Please click here to view a larger version of this figure.
Figure 7: Gloved fingertip sampling. The correct method for obtaining gloved fingertip samples using the largest surface area (or pad) of each finger/thumb is shown on the left. The incorrect process where only the fingertip is sampled is shown on the right. Please click here to view a larger version of this figure.
4. Sterility testing by direct product inoculation
Figure 8: Growth of mold that failed to be detected by the BacT/ALERT. Example of mold balls, visible to the naked eye, that failed to be automatically detected by the BacT/ALERT system. Based on these findings, we recommend terminal visual inspection of all BacT/ALERT bottles and the addition of the SAB plate for fungal culture using the NIH Alternative Sterility Testing Method. Please click here to view a larger version of this figure.
The expected results are described in Table 1. The EM data should be reviewed and followed up with an appropriate investigation and response to action, alert, or ISO limit excursions. If an excursion occurs for non-viable particles, one should proceed as per ISO 14644-Annex A, sec A.5.57. If the excursion can be attributed to an immediately identifiable abnormal occurrence, the original sampling results should be documented, a note should be added to disregard the original results, and a detailed description of the abnormal occurrence should be provided. Another sample may be taken at the site in question. If the excursion cannot be attributed to an immediately identifiable abnormal occurrence, document the sampling result. Take another set of non-viable samples at the site in question immediately after identifying the out of specification (OOS) result. This can be used as part of the investigation. As per ISO 14644-Annex A, section A.5.6, investigate the OOS finding and remediate the identified root cause.
A failed product sterility test will result in turbidity of the USP<71> culture or growth in the BacT/ALERT bottles and/or SAB plate (Figure 8). The organism should be identified to the species level, and appropriate susceptibility testing should be performed, if clinically warranted. An investigation should be completed by the testing laboratory to determine if the OOS result is valid or if there was any potential laboratory error that may have contributed to the OOS. Testing of the retention sample is strongly recommended. The decision to withhold or release the product for infusion should be made by the patient's clinical care team, and by the manufacturing, microbiology, infectious diseases, quality assurance, and regulatory affairs team. For HCT/Ps that may have already been infused based on preliminary in-process testing, the patient should be carefully monitored, and an event report should be submitted to the appropriate regulatory agency.
There are several critical areas in this protocol, including the maintenance of aseptic technique and unidirectional airflow within cleanrooms and the BSCs. Best practices include moving slowly and deliberately to minimize turbulence. Aseptic manipulations should be performed from the side of the product, not from above. Closed system processing and the use of terminally sterilized raw materials are recommended. Speaking in critical areas and leaning against walls or equipment should be avoided. Similarly, unnecessary touching of non-sterile items and picking up fallen items should be avoided until the sterile processing has been completed. The materials in the BSC should be arranged to prevent blockage of the airflow and to maintain unidirectional movement from clean to dirty. Operator movement in and out of the BSC should be minimized. The documentation of all activities including lot numbers, expiration dates, calibration dates, start/stop times, and testing personnel for all the materials, equipment, and processes within a sampling or testing session is also critical.
The EM procedure described here relies on manual incubation and colony counting. The design of an EM program is at the discretion of the end user. The sampling locations and test frequency should be guided by PDA Technical Report 138 and justified by a risk assessment that incorporates the laboratory workflow, product proximity, site criticality, duration, and the number of personnel for each workflow step8. A typical EM program should incorporate the total air particulates (0.5 μm non-viable particles), viable air sampling (1,000 L), and viable surface sampling collected under dynamic conditions at a frequency based on product risk and historical trends. Weekly sampling is generally recommended for new facilities until sufficient data trends have been established. General guidance for EM and culture conditions is provided in Table 1 and in USP<1116>9; however, others have evaluated different culture conditions, such as medium that was limited to TSA/TSALT only, and different incubation temperatures, including dual temperatures10,11. Automated EM instruments that combine incubation with colony count plate reading are commonly used in the pharmaceutical market as rapid methods8,12. The chosen EM program must be validated and is typically dependent on the facility flora, incubator capacity, test volume, and personnel capacity. Furthermore, an established EM program should have an active life cycle, with relevant adjustments made to the collection sites, test frequency, and/or alert/action limits based on an annual review and a data-driven risk assessment8. Major changes to EM trends may trigger the re-evaluation of the cleaning program. Disinfectants should be validated via a disinfectant efficacy study using representative material surfaces against facility flora for the prescribed contact time, as described in USP<1072>13. Typical disinfectants may include Vesphene III (contact time: 10 min), LpH III (contact time: 10 min), and a sporicidal agent such as Peridox RTU (contact time: 5 min). For best practices, disinfectants should be rotated on a monthly basis, and the outlets, mechanical connections, and work surfaces that are under equipment in the BSC should be cleaned routinely.
The minimum product volume that must be tested for sterility is defined in USP<71>3 and is based on the total final product volume. Unfortunately, USP<71> was designed originally for sterile drug products and is recognized to be unsuitable for short shelf-life products due to the slow turnaround time and high product test volume14. USP<1071> acknowledges that alternative rapid microbial methods may be beneficial and that lower product testing volumes may be acceptable when a risk-based approach has been applied14. Given the limitations of USP<71> for the sterility testing of cellular therapy products, alternative methods such as automated BacT/ALERT respiration methods have been used in industry1,4,6. However, any use of an alternative testing method requires rigorous end-user validation to demonstrate non-inferiority to the compendial USP<71> method for that product5,15,16. Method suitability testing for each new product is also required to ensure that the product, at the chosen inoculation volume and test conditions, is not inhibitory to the detection of low-level contaminants6. Therefore, it is possible that the test volume and inoculation volume may vary between products depending on the outcome of the validation studies. Membrane filtration, a secondary method outlined in USP<71>, may be used to sample a larger sample volume in a single test. Validation and method suitability testing should include organisms beyond the six compendial QC isolates. Focus should be given to frequently recovered facility flora and previous product contaminants. Particular attention should be given to fungi, as respiration methods alone have demonstrated suboptimal performance4,17, which is why the NIH Alternative Sterility Testing Method includes a paired fungal culture on SAB and a terminal visual inspection of the BacT/ALERT bottles.
A major limitation for any microbial control program is the retrospective nature in which results are available. Gold standard testing is culture-based, which is slow and can lead to constant reactive measures for corrective action. Faster testing methods require rigorous validation, but these still cannot provide real-time assurance of microbial monitoring and control. Furthermore, microbiological cultures capture only a small subset of time during a production or testing session. Therefore, it is critical that active microbial monitoring occurs on a frequent risk-justified basis to ensure that the trend data is well-established. This will enable the use of appropriately set alert limits to predict potential deviations from microbial control before they occur.
The establishment of a robust cGTP/cGMP program takes time and effort and, unfortunately, cannot be simply transferred from one facility to another. The United States Food and Drug Administration expects that all programs undergo end-user validation testing despite the potential availability of third-party published results demonstrating their efficacy. Therefore, microbial control and testing strategies may differ between institutes depending on variables including the facility design, facility flora, and product risk classification. The strategies outlined in this protocol help provide a framework through which to establish a robust microbial monitoring and control program for cGTP/cGMP laboratories.
The authors have nothing to disclose.
This work was supported by the Intramural Research Program of the National Institutes of Health Clinical Center. The content is solely the responsibility of the authors and does not represent the official views of the National Institutes of Health.
20-25°C Incubator | Lab preference | ||
30-35°C Incubator | Lab preference | ||
Alcohol-based hand sanitizer | Lab preference | ||
BacT/ALERT Dual-T instrument | BioMerieux Industry | ||
Beard cover | Lab preference | ||
Biosafety cabinet (BSC) | Lab preference | ||
Cleanroom shoes | Lab preference | ||
Fluidthioglycollate medium (FTM) | Hardy Diagnostics | U84 | USP |
Handheld cleaning mop | Contec | 2665LF | |
Hypodermic needle | Lab preference | ||
iFA+ BacT/ALERT bottle | Biomerieux | 412990 | |
iFN+ BacT/ALERT bottle | Biomerieux | 412991 | |
Isokinetic head | Lab preference | ||
Laser particle counter | TSI Incorporated | 9500-01 | |
LpH III | Steris | 1S16CX | |
Mirror | Lab preference | ||
Non-sterile bouffant | Lab preference | ||
Non-sterile gloves | Lab preference | ||
Non-sterile shoe covers | Lab preference | ||
Non-sterile sleeve covers | Lab preference | ||
Parafilm | Lab preference | ||
Peridox RTU | Contec | CR85335IR | |
Plastic bag | Lab preference | ||
Sabouraud Dextrose Agar with Lecithinase and Tween (SABLT) | Hardy Diagnostics | P595 | USP, irradiated |
Sabouraurd Dextrose Agar (SAB) | Hardy Diagnostics | W565 | USP, irradiated |
Safety glasses | Lab preference | ||
Scrubs (top and bottom) | Lab preference | ||
Spor-Klenx RTU | Steris | 6525M2 | |
Sterile 70% isopropyl alcohol (IPA) | Decon CiDehol | 8316 | |
Sterile alcohol wipe | Lab preference | ||
Sterile boot covers | Kimberly Clark | Cat# varies based on size | |
Sterile coveralls | Kimberly Clark | Cat# varies based on size | |
Sterile face mask | Lab preference | ||
Sterile gloves | Lab preference | ||
Sterile hood | Kimberly Clark | Cat# varies based on size | |
Sterile low-lint wipes | Texwipe | TX3210 | |
Sterile mop cleaning pads | Contec | MEQT0002SZ | |
Sterile sleeve covers | Kimberly Clark | 36077 | |
Sterile spreading rod | Fisher Scientific | 14665231 | |
Sterile syringe | Lab preference | ||
Tacky mats | Lab preference | ||
Tryptic Soy Agar (TSA) | Hardy Diagnostics | W570R | USP, irradiated |
Tryptic Soy Agar with Lecithinase and Tween (TSALT) | Hardy Diagnostics | P520R | USP, irradiated |
Tryptic Soy Broth (TSB) | Hardy Diagnostics | U46 | USP |
Tubing | Lab preference | ||
Vesphene III | Steris | 1S15CX | |
Viable air sampler | Hardy Diagnostics | BAS22K | |
Vortex | Lab preference |