Microbial Control During Process Pool Holds of Biological Products – Building Quality into the Production Process

Reyes Candau-Chacon, PhD - Corresponding author; Food and Drug Administration, Center for Drug Evaluation and Research, Office of Pharmaceutical Quality, Office of Pharmaceutical Manufacturing Assessment

Bo Chi, PhD - Food and Drug Administration, Center for Drug Evaluation and Research, Office of Pharmaceutical Quality, Office of Pharmaceutical Manufacturing Assessment

Candace Gomez-Broughton, PhD - Food and Drug Administration, Center for Drug Evaluation and Research, Office of Pharmaceutical Quality, Office of Pharmaceutical Manufacturing Assessment

Patricia Hughes, PhD - Food and Drug Administration, Center for Drug Evaluation and Research, Office of Pharmaceutical Quality, Office of Pharmaceutical Manufacturing Assessment

Abstract

Production processes for biological products are often designed to include in-process hold steps to provide flexibility during operations. During hold conditions, process pools (a.k.a. process intermediates) may be at risk of microbial contamination that can negatively impact product quality and put patient safety at risk. Therefore, manufacturers should conduct studies to demonstrate microbial control during the proposed worst-case hold conditions to ensure that a quality product can be manufactured consistently. Contamination may result either from the production environment or from the proliferation of microorganisms already present in the product. Risk assessments are recommended to identify operations susceptible to microbial contamination to implement appropriate risk mitigation strategies. This article discusses the approaches that may be followed to support hold conditions from a microbiological perspective and provides case studies to illustrate the different approaches.

Keywords

Biotech manufacturing, biological product, hold conditions, process intermediates, process pools, microbial control, bioburden and endotoxin monitoring, validation, risk-based approach.

Introduction

Biological products are regulated by Section 351 of the Public Health Service (PHS) Act, the Federal Food, Drug, and Cosmetic (FD&C) Act, and the Code of Federal Regulations (CFR) 21CFR 210, 211, 600-680. In general, manufacturing of biological products involves multiple highly complex operations: fermentation/cell culture (inoculation, expansion, and production), product recovery and purification by chromatography and, for mammalian cell cultures, adventitious agent clearance and removal by physical/chemical treatments. The purified protein product is typically concentrated and formulated prior to final 0.2 µm filtration and released as a Bulk Drug Substance (BDS). Finally, the BDS is aseptically processed into a finished drug product at a fill/ finish site. During most production steps, process pools are maintained under conditions conducive to microbial growth; therefore, a robust microbial control strategy is necessary to prevent contamination and to maintain a constant supply of a quality product^10,11,18.

At several stages of the manufacturing process, process pools may be held to provide operational flexibility; for example, to meet process design requirements (i.e., several chromatography cycles may need to be combined before the following step) or when equipment availability is limited or delayed due to equipment preparation. The International Conference of Harmonization (ICH) Q7 indicates the need for establishing time limits for the completion of the individual steps and for storing process pools under appropriate conditions to ensure their suitability for use⁶. Biologic License Applications (BLA) typically specify the hold conditions for the process pools and include data from small scale studies supporting their chemical stability. However, data demonstrating microbial control during hold conditions should be obtained from commercial scale studies because small scale studies are not representative of the manufacturing environment and process. Typically, contaminating microorganisms may gain ingress into a process from the facility, equipment, personnel, and raw materials¹⁰.

The Food and Drug Administration (FDA) favors a scientific, risk-based approach to pharmaceutical development as a means to build quality into products based on process knowledge rather than testing products into quality^7,8,16. This article addresses the risk factors and risk mitigation strategies that should be considered when validating hold times from a microbiological quality perspective and describes different approaches that may be followed to demonstrate microbial control during the hold of biological process pools. The focus of this article is biological product pools prior to sterile filtration; hold conditions of the sterile product should be supported by media fill data.

Demonstration of Microbial Control During the Hold and Routine Monitoring

Hold time validation from a microbiology perspective and routine microbiology monitoring are both necessary to demonstrate microbial control of the manufacturing process.

Data providing evidence that a process is capable of consistently delivering a quality drug substance⁹ should be included at the time of the BLA submission¹⁰ to avoid unexpected delays in BLA approval, potential quality problems at launch and during product lifecycle, and to prevent drug shortages after approval. Validation studies, including the Process Performance Qualification (PPQ) batches, are intended to support a process that results in a quality product on a consistent basis^3,12. A BLA should also include supporting microbiological quality data for the proposed hold conditions during the manufacturing process. Biological products are manufactured with raw materials and conditions conducive to microbial proliferation and in some conditions, even very low bioburden at the beginning of the hold may proliferate to unacceptable levels during the hold. While the need of validating chemical stability of process pools is widely accepted²⁰, some applicants avoid conducting studies at commercial scale because the studies can be costly and may disrupt operations by imposing unnecessary or artificial pauses during manufacturing. In addition, some manufacturers think that any increase in bioburden during a hold period can be eliminated by subsequent filtration steps, mitigating any risk to product quality. This practice is not acceptable; although bioburden may be removed by filtration, bioburden byproducts can remain after filtration and may impact product quality by degrading or modifying the product and contributing endo/exotoxins to the product. In addition, microbial byproducts may contribute to product and process variability¹⁰. Therefore, microbiological hold time validation data obtained from commercial scale studies are expected to be provided in the submission to support a process that will consistently result in a quality product.

Case Study 1 (Prior Approval Supplement (PAS) to add a new drug substance manufacturing facility): Process holds were validated at small scale to demonstrate chemical stability; the studies did not consider the impact of the facility, equipment, personnel, and practices on microbial control to maintain product quality. The PPQ batches were not held for the proposed maximum holds. In an information request, FDA requested data to support microbial control during the maximum holds; the applicant indicated that routine monitoring was sufficient to support microbial control of the process. FDA communicated that without hold time validation data from a microbiology perspective, the proposed hold times should be limited to 24 hours to mitigate the risk of bioburden excursions and potential for shortages. The applicant decided to conduct studies to support microbial control at the worst-case hold conditions.

Routine monitoring for microbial quality should be conducted even when data supporting the proposed hold conditions have been generated^5,10,17. Routine monitoring provides information about the microbiological status of a particular batch, and in the event of an excursion, it provides a basis for initiating an investigation and promptly implementing corrective and preventive measures. Routine monitoring should be conducted in a manner that minimizes the risk of contamination during sampling of the batch and the sample itself. Multiple sampling devices with aseptic connections are currently available and will help to mitigate the risk of contamination during sampling.

Case Study 2 (new BLA): The manufacturing process did not include routine in-process microbiology samples and there was no validation data to support microbial control during process holds. After an FDA information request, the applicant proposed to monitor bioburden and endotoxin at the end of the maximum hold for three validation batches but not during routine manufacturing. In this case, the applicant intended to rely on the one-time validation study and not monitor the in-process pools during routine operations. FDA did not consider the strategy acceptable due to the risk of undetected contamination during product manufacture. A deficiency on the microbial control strategy was included in the Complete Response letter of this BLA.

Risk Factors to Consider When Demonstrating Microbial Control During Hold

Bioburden mapping studies and risk assessments should be conducted to evaluate which process pools are at risk of contamination during the hold and to identify which process holds should be validated for microbiological quality. For example, process pools that do not support microbial growth or low bioburden pools held for short periods of time (less than 24 hours) do not need supporting validation data because the predicted growth rates of microorganisms under those conditions may not be sufficient to negatively impact product quality. In addition, risk mitigation factors, such as use of closed processes or bioburden-reduction filters, may lower the risk of contamination, and should be taken into consideration.

The following factors should be considered when assessing the risk of microbial ingress in the process pools:

• Facility: area classification, segregation, cleanliness
• Equipment: sanitation of the hold vessel (e.g., sterile, sanitized, or clean vessel), sterile single-use equipment (e.g., sterile disposable bag with in-line filter), vessel integrity (integrity tested in the vessel after sterilization or by the supplier in case of single-use systems), number and use of ports, number and type of interventions, status of the gaskets, types of sampling devices, equipment storage, etc.
• Sanitization and storage conditions of chromatography resins and UFDF membranes
• Raw materials and components
• Procedures: flow of material, product, equipment, and personnel, manufacturing practices, gowning
• Process design: open or closed processes, use of bioburdenreduction filters
• Bioburden carried over from the previous process step

In addition, the following factors may be considered when assessing the risk of microbial proliferation in the in-process pools during the hold^{1, 2, 13, 14, 15, 17}:

• Hold time
• Hold temperature
• Composition of the process pool matrix (nutrients)
• pH of the process pool matrix
• Volume of the process pool in relation to the volume of the vessel and agitation during the hold (these parameters may impact the oxygenation.)

Additional studies, such as growth promotion of the process pools, may be conducted to evaluate some of the risk factors. All the contributing factors in the risk assessment and risk mitigation strategy should be considered for each process pool to establish those pools at risk of contamination and to provide a science- and risk-based decision about the need to validate the process conditions to ensure microbiology quality during the hold at each step. Worst-case conditions may be used to justify other conditions. For example, a validation run conducted at 15 to 25°C is generally acceptable to demonstrate microbial control at lower temperatures because proliferation of most organisms will be slower at the lower temperature. 

Approaches to Establishing Maximum Hold Times

Several approaches may be used to assess microbial control during process pool hold conditions. The most common ones are discussed below; however, other approaches may be used, as long as they simulate the process risks and they include a scientific justification¹⁹. In all cases, microbiology (bioburden and endotoxin) samples should be collected at the end of hold.

Validation through concurrent manufacture:

The process pool is held in the vessel for the proposed maximum time and temperature. In this case, hold studies are conducted concurrently during manufacturing using commercial scale equipment in the manufacturing area. The validation may be conducted during the production of the PPQ batches or during a different period of time; in this case, under a validation protocol approved by Quality Assurance that includes defined acceptance criteria. This is the most straightforward strategy used in many BLAs; however, if the process does not have adequate microbial control, this approach may result in unacceptable bioburden levels that could lead to bioburden excursions and batch rejection. Although this approach may be somewhat risky to the manufacturer, it provides assurances of microbial control with the supportive at scale validation information.

Validation using material left behind in the vessel:

In this approach, most of a process pool is processed forward to the next manufacturing step leaving a representative portion of the batch in the hold vessel to obtain supporting microbiology hold data. This approach considers the risk factors derived from the facility (vessel integrity, storage area, manufacturing area environment, personnel, and manufacturing practices), and the risks from the initial microbial load (vessel cleanness and microbial load carried over from the previous step). A limitation of this approach is that relatively large volumes may need to be held to cover vessel sampling ports; the left-behind portion of the batch may not be processed forward after the hold and may be discarded.

Case Study 3 (PAS to add a new drug substance manufacturing facility using a “slipstream” approach): In this case, the process pools were not filtered in the hold vessels after the chromatography steps. The manufacturer transferred approximately 1% of the process pools (2 to 5 L) to identical vessels and held them for the proposed maximum hold time; bioburden and endotoxin samples were collected at the end of the hold. The study was only conducted with one batch. FDA did not consider the strategy acceptable because one run was not sufficient to demonstrate consistency of microbial control; in addition, the left-over volume was too low and insufficient to cover all vessel ports. After interaction with the Agency, the applicant lowered the maximum hold times to 24 hours.

Vessel integrity using surrogate solutions:

For process pools filtered into the hold vessels, there is low risk of bioburden carried over from previous steps; therefore, demonstration of hold vessel integrity with filtered surrogate growth promoting solutions throughout the hold may be sufficient to address contamination risk factors coming from the vessel and the facility. The critical element in this approach is the growth promoting properties of the surrogate.

Case Study 4 (new BLA using a vessel integrity study): In this BLA, the process pools were routinely filtered into the hold vessels with a proposed maximum hold of 35 days at room temperature. The hold time validation strategy consisted of filtering a surrogate solution (Tris/sucrose/mannitol buffer) into the hold vessels; the product hold vessel was monitored for bioburden throughout the hold. The applicant conducted a preliminary “buffer recovery test” with an acceptance criterion of “demonstration that different microbes survive in the test buffer”. The buffer recovery study using five compendial organisms and three environmental isolates resulted in no recovery of three organisms (S. aureus, A. iwoffii, and S. epidermidis) and a downward trend for A. brasiliensis, indicating that the buffer inhibited growth for 50% of the organisms tested. FDA communicated to the applicant that the study did not demonstrate integrity of the vessels because the surrogate solution did not promote microbial growth. The applicant agreed to repeat the study using a growth promoting solution and to lower the maximum hold times to those validated concurrently during PPQ until the results were available.

Multiproduct facilities that use similar hold vessels for several products often use one growth promotion solution to validate the integrity of representative vessels across products. The integrity of several vessels can be demonstrated using a family approach. In this case, it is necessary to have appropriate parameters to justify grouping the vessel into a family; some of those parameters include material of construction, size and geometry, complexity, number of ports, and number of interventions in the vessel.

Case Study 5 (new BLA using a vessel family approach): This BLA included demonstration of vessel integrity with a surrogate solution and a vessel family approach. When FDA asked the rationale for family grouping of the vessels, the only parameters cited by the applicant were “material of construction, validated CIP and SIP processes, and filtration of the pool into the vessel”. In addition, the applicant intended to use the studies conducted in one building to validate the hold in a different building, disregarding the risk factors from the facility. The applicant agreed to conduct an additional study in support of the hold conditions of the process pools.

The vessel integrity approach is not applicable if the pools are not filtered into the vessel due to potential bioburden present in the pools at the beginning of the hold. On the other hand, process pools filtered and held in qualified, single-use, sterile bags with an in-line 0.2-µm filter may not need validation because microbial contamination risks are low due to the sterility and integrity of the bags (certified by the supplier) and in-line bioburden reduction filtration of the process pools.

Studies to support microbial control during the maximum hold are typically conducted under a Quality Assurance approved protocol that includes hold conditions (such as vessels, storage areas, temperatures, maximum times, and surrogate solutions in case of vessel integrity studies) and acceptance criteria. Bioburden and endotoxins are sampled at the beginning and end of the hold and analyzed using a qualified method. The number of runs depends on the process variability, amount of experimental data, and process knowledge⁹ ; typically three runs are conducted to ensure consistency of microbial control. Acceptance criteria may be based on the bioburden limit of the pools or, in case of demonstration of vessel integrity, no growth at the end of the hold.

Demonstration of Maximum Hold Times of Processing Buffers

The need to demonstrate microbial control of growth-promoting process buffers during hold (even if they are monitored or filtered prior to use) follows the same rationale and the same principles used for the process pools. Microbial control of buffers is typically assessed during inspection and lack of control often results in FDA-483 observations. For example, in a drug substance manufacturing inspection, the following observation was made: “Purification buffers do not have adequate microbial control. Specifically, buffers are prepared by mixing filtered stock solutions with filtered water (PW or WFI); stock solutions and buffers can be stored for up to six months without data to demonstrate microbial control at scale.” In this case, some of the buffers were contaminated during the hold and resulted in multiple contaminations of the process pools.

Conclusion

Demonstration of microbial control during hold conditions is important to support a quality product and to ensure a consistent manufacturing process. Contamination risk factors and mitigation strategies can be used to assess the need for validating the maximum hold times. Different validation strategies may be used as long as they are conducted using the commercial process, are scientifically justified and are clearly described in the application.

References

1. Andersen K. and Meyenburg, K. (1980) Are growth rates of Escherichia coli in batch cultures limited by respiration? J. Bacteriol. 144: 114-123
2. Davey, K.R. (1994) Modelling the combined effect of temperature and pH on the rate coefficient for bacterial growth. Int. J. Food Microbiol.23: 295-303
3. FDA Guidance for Industry. Process Validation: General principles and practices. January 2011.
4. ICH Harmonised Tripartite Guideline Q5c. Quality of biotechnological products: stability testing of biotechnological/biological products. November 1995
5. ICH Harmonised Tripartite Guideline Q6b. Specifications: test procedures and acceptance criteria for biotechnological/biological products. March 1999
6. ICH Harmonised Tripartite Guideline Q7. Good manufacturing practice guide for active pharmaceutical ingredients. November 2000
7. ICH Harmonised Tripartite Guideline Q8. Pharmaceutical development. August 2009
8. ICH Harmonised Tripartite Guideline Q10 Pharmaceutical quality system. June 2008
9. ICH Harmonised Tripartite Guideline Q11. Development and manufacture of drug substances (chemical entities and biotechnological/biological entities). May 2012
10. Lolas, A.G., Chi, B. Hughes, P.F., and Suvarna K. (2010), “CMC Microbiology Review of Biologics License Applications and Pre-License/Pre-Approval Inspections: Therapeutic Biological Proteins,” Amer. Pharm. Rev. 13: 56-60
11. Lolas A.G. (2014) The Role of Microbiology in the Design and Development of Pharmaceutical Manufacturing Processes. Pharm. Bioprocessing 2: 125-128
12. Martin-Moe S. et al. (2011) A New Roadmap for Biopharmaceutical Drug Product Development: Integrating Development, Validation, and Quality by Design. J. Pharm. Sci., 100: 3031-3043
13. Membré, J.M. et al. (2005) Temperature effect on bacterial growth rate: quantitative microbiology approach including cardinal values and variability estimates to perform growth simulations on/in food. Int. J. Food Microbiol.100: 179-186
14. Monod, J. (1949) The growth of bacterial cultures. Ann. Rev. Microbiol. 3: 371-394
15. Ratkowsky,D.A. et al. (1982) Relationship between temperature and growth rate of bacterial cultures. J. Bacteriol. 149: 1-5
16. Rathore A.S. and Winkle H. (2009) Quality by design for biopharmaceuticals. Nature, 27: 26-34
17. Sandle T. (2015) Assessing Process Hold Times for Microbial Risks: Bioburden and Endotoxin, Journal of GXP Compliance: 19: 1-9
18. Suvarna, K., Lolas, A., Hughes, P., and Friedman R. L. (2011) Studies of Microbial Contamination in Biologic Product Manufacturing. American Pharmaceutical Review 1: 50-56
19. Suvarna, K. (2012) Hold time studies in therapeutic biological product manufacturing. PDA’s 7th Annual Global Conference on Pharmaceutical Microbiology
20. Thillaivinayagalingam, P. and Newcombe, A.R. (2011) Validation of intermediate hold times. BioProcess International, April: 52-57

Subscribe to our e-Newsletters
Stay up to date with the latest news, articles, and events. Plus, get special
offers from American Pharmaceutical Review – all delivered right to your inbox!
Sign up now!