By: Naveenganesh Muralidharan, Founder & Principal Consultant, Bench2Batch CMC Lifecycle Partners™
Introduction
How long is too long to hold a process intermediate before microbial risk becomes unacceptable? This question continues to challenge biopharmaceutical manufacturers and remains a frequent focus during regulatory inspections. In most cases, the issue is not the duration of the hold time itself, but the absence of a clear, risk-based justification linking hold-time duration to microbial acceptance criteria and downstream control capability.¹
Current regulatory expectations and industry practice support a lifecycle-based approach in which hold times, bioburden limits, and endotoxin limits are intentionally aligned with stage-specific microbial risk rather than applied uniformly across the manufacturing process.² Early processing stages may tolerate higher microbial loads due to robust downstream clearance, whereas later stages require progressively tighter control as opportunities for mitigation become limited. When acceptance criteria are grounded in process understanding, risk assessment, and validation data, they can support operational flexibility while maintaining regulatory compliance.³
Hold Times as a Control Point for Microbial Risk
Hold times represent defined periods during which process materials remain static between manufacturing operations. During these intervals, microbial proliferation and endotoxin accumulation may occur in the absence of active clearance mechanisms such as sterile filtration, chromatography, or viral reduction steps.⁴ The microbial risk associated with a hold time is influenced by multiple variables, including nutrient availability, temperature, pH, container closure integrity, and exposure to the manufacturing environment.⁵
Regulatory agencies do not prescribe fixed limits for hold times or microbial acceptance criteria. Instead, manufacturers are expected to establish scientifically justified controls based on process knowledge, historical performance, and validation studies.³ Applying identical hold-time limits across all manufacturing stages, without regard to lifecycle-specific risk, can result in unnecessary operational constraints in early stages or insufficient microbial control in later stages. A stage-appropriate strategy is therefore essential.⁶
Early Upstream Operations: Cell Expansion and Seed Bioreactors
Early upstream operations, including cell expansion and seed bioreactor stages, are typically performed under sterile conditions and are not routinely associated with extended static hold periods. As a result, defined hold-time limits are generally not required for these operations. Microbial control at this stage is instead maintained through closed processing, controlled cleanroom environments, and routine microbiological monitoring.⁷
For both cell expansion and seed bioreactor operations, bioburden alert limits are commonly established at ≥1 CFU per 10 mL, with corresponding action limits at ≥10 CFU per 10 mL. Samples are typically collected aseptically from Class D, C, or B cleanroom environments and analyzed in nonclassified microbiological laboratories. Under these conditions, the appearance of low-level colonies on bioburden plates may occasionally result from sampling or handling artifacts rather than true process contamination, increasing the risk of false-positive results and unnecessary batch rejection.²
In early stages of the manufacturing process—including shake flasks, culture bags, and seed bioreactors—true microbial contamination is generally detected through rapid and pronounced process signals such as abrupt pH shifts, dissolved oxygen deviations, or sudden declines in viable cell density. These indicators typically manifest well before completion of the standard seven-day incubation period associated with bioburden testing. Consequently, establishing an action limit of ≥10 CFU per 10 mL in addition to an alert limit of ≥1 CFU per 10 mL is considered an appropriate risk-based strategy that balances process protection with the practical risk of false-positive microbiological results.2,3
Endotoxin testing is not routinely required at these early upstream stages due to the sterile nature of the operations, the absence of extended hold periods, and the availability of multiple downstream clearance steps.³
Production Bioreactor and Harvest Operations
At the end of a production bioreactor run, microbial monitoring focuses on early detection rather than clearance, as active microbial reduction steps have not yet been applied. Bioburden alert and action limits are typically set at ≥1 CFU per 10 mL and ≥10 CFU per 10 mL, respectively, with endotoxin alert and action limits at ≥5 EU/mL and ≥10 EU/mL.³,⁸
Harvest operations introduce the first defined hold periods of potential microbial concern. In large-scale manufacturing, clarified harvest material following centrifugation and/or depth filtration are filtered through 0.2-micron filtration. At this stage, higher microbial limits are acceptable due to the nature of open processing in the clarification steps and presence of downstream microbial reduction steps.
For clarified bulk material prior to 0.2-micron filtration, bioburden alert limits of ≥200 CFU/mL and action limits of ≥1,000 CFU/mL are commonly justified, with endotoxin alert and action limits remaining at ≥5 EU/mL and ≥10 EU/mL. Typical hold times at this stage range from six to eight hours before sterile filtration, depending on manufacturing scale and process design.¹,²
Following 0.2-micron filtration, microbial risk is significantly reduced. Post-filtration clarified bulk is therefore subject to tighter bioburden limits, typically with alert limits of ≥1 CFU per 10 mL and action limits of ≥10 CFU per 10 mL. Endotoxin limits generally remain unchanged. Hold times following sterile filtration are commonly justified for a minimum of 12 to 24 hours, with maximum durations determined by chemical stability and process-specific risk assessments.²,⁴
Media, Feed, and Buffer Preparation
Microbial control strategies must also address hold times associated with the preparation and storage of cell culture media, feeds, and downstream buffers. Media and feed solutions prepared prior to sterile filtration are inherently growth-promoting due to their nutrient composition and therefore present an elevated microbial risk.²
|
Table 1. Typical Hold Times and Microbial Acceptance Criteria by Manufacturing Stage
|
|---|
|
Sample Type
|
Unit Operation
|
Bioburden Acceptance Criteria
|
Endotoxin Acceptance Criteria
|
Typical Hold Time
|
|---|
|
Non-routine
|
Cell expansion
|
Alert: ≥1 CFU/10 mL
Action: ≥10 CFU/10 mL
|
Not routinely required
|
N/A
|
|
Non-routine
|
Seed bioreactor
|
Alert: ≥1 CFU/10 mL
Action: ≥10 CFU/10 mL
|
Not routinely required
|
N/A
|
|
Routine in-process control
|
Production bioreactor (end of run)
|
Alert: ≥1 CFU/10 mL
Action: ≥10 CFU/10 mL
|
Alert: ≥5 EU/mL
Action: ≥10 EU/mL
|
N/A
|
|
Routine in-process control
|
Harvest (clarified bulk), post-centrifugation and/or depth filtration, pre-0.2 µm filtration
|
Alert: ≥200 CFU/mL
Action: ≥1,000 CFU/mL
(Justified by risk assessment considering filter type, surface area, and microbial reduction capability)
|
Alert: ≥5 EU/mL
Action: ≥10 EU/mL
|
Typically 6–8 hours prior to 0.2 µm filtration
|
|
Routine in-process control
|
Harvest (clarified bulk), post-0.2 µm filtration
|
Alert: ≥1 CFU/10 mL
Action: ≥10 CFU/10 mL
|
Alert: ≥5 EU/mL
Action: ≥10 EU/mL
|
Minimum 12–24 hours; maximum based on chemical stability
|
|
Process validation
|
Media, feed, or buffer preparation (pre-filtration)
|
Alert: ≥200 CFU/mL
Action: ≥1,000 CFU/mL
(Justified by risk assessment considering filter type, surface area, and microbial reduction capability)
|
Alert: ≥5 EU/mL
Action: ≥10 EU/mL
|
Up to 12 hours
|
|
Process validation
|
Storage of filtered media or buffers
|
Alert: ≥1 CFU/10 mL
Action: ≥10 CFU/10 mL
|
Alert: ≥5 EU/mL
Action: ≥10 EU/mL
|
Based on chemical stability and facility scheduling
|
|
Process validation
|
Drug substance purification intermediates (Capture chromatography, viral inactivation, viral filtration, polishing chromatography including CEX, AEX, or HIC)
|
Alert: >/= 10 to 30 CFU per 10 mL
Action: ≥100 CFU/10 mL
(Justified by risk assessment considering filter type, surface area, and microbial reduction capability)
|
Alert: ≥5 EU/mL
Action: ≥10 EU/mL
|
Up to 24 hours
|
|
Routine in-process control
|
Late downstream drug substance intermediate (UF/DF)
|
Acceptance criterion: <10 CFU/100 mL
|
Acceptance criterion: 0.25–1 EU/mL
|
6–12 hours
|
|
Routine in-process control
|
Final formulation/excipient addition (pre-sterile filtration)
|
Acceptance criterion: <10 CFU/100 mL
|
Acceptance criterion: 0.25–1 EU/mL
|
6–12 hours
|
For pre-filtration media, feed, and buffer preparation, bioburden alert limits of ≥200 CFU/mL and action limits of ≥1,000 CFU/mL are commonly applied, with corresponding endotoxin alert and action limits of ≥5 EU/mL and ≥10 EU/mL. Prefiltration hold times are typically limited to up to 12 hours to minimize microbial proliferation while accommodating operational requirements.²,⁵ These limits are further justified through risk assessments that consider the microbial retention capability of the selected filters, the effective filtration surface area, and the potential impact of bioburden, endotoxin, or exotoxin buildup on subsequent sterile filtration performance.
Following sterile filtration, microbial risk is substantially reduced, and bioburden acceptance criteria are correspondingly tightened. Bioburden alert and action limits are typically established at ≥1 CFU per 10 mL and ≥10 CFU per 10 mL, respectively, while endotoxin limits generally remain unchanged. Hold times for filtered media and buffers are determined based on chemical stability data, integrated with facility scheduling and operational needs.³,⁴
Intermediate Drug Substance Purification Operations
Intermediate drug substance purification operations, including Capture chromatography, viral inactivation, viral filtration, and polishing chromatography steps such as cation exchange, anion exchange, or hydrophobic interaction chromatography, represent a transitional phase in the manufacturing lifecycle. Although nutrient content is reduced, materials may still support microbial growth if held for extended periods.⁶
At this stage, bioburden alert limits are commonly set at >/= (10 to 30) CFU per 10 mL, with action limits at ≥100 CFU per 10 mL. Endotoxin alert and action limits typically remain at ≥5 EU/mL and ≥10EU/mL. ASTM D5465-93 (1998) and FDA Pharmaceutical Microbiology Manual (ORA.007) (2020) guidance indicate that reliable plate count estimation generally ranges from approximately 20–30 CFU on the lower end to 200–300 CFU on the upper end, depending on the enumeration technique; therefore, setting the alert range at 10 to 30 CFU per 10 mL ensures results remain within statistically reliable counting limits while maintaining conservative process control.10,11 Hold times for purification intermediates are generally limited to up to 24 hours, balancing microbial control with manufacturing logistics and downstream clearance capability.⁴,⁶
Late-Stage Processing: UF/DF and Pre-Sterile Filtration Holds
Late downstream stages, including ultrafiltration and diafiltration (UF/DF) and final formulation or excipient addition prior to sterile filtration, represent the highest microbial risk from a product-impact perspective. At this point in the process, materials are highly refined, and opportunities for downstream microbial clearance are limited.⁷,⁸
Accordingly, bioburden acceptance criteria are typically defined as <10 CFU per 100 mL, with endotoxin acceptance criteria tightened to 0.25–1 EU/mL. Hold times at these stages are commonly restricted to six to twelve hours to minimize the potential for microbial proliferation and endotoxin accumulation.⁷,⁸
Lifecycle Integration of Hold Times and Acceptance Criteria
Across all stages of biomanufacturing, a consistent principle emerges: microbial acceptance criteria and hold-time durations must be aligned with lifecycle-specific risk. Early stages tolerate higher microbial loads due to robust downstream clearance, while later stages demand progressively tighter control as mitigation options decrease. Applying this framework allows manufacturers to establish defensible, science-based controls without imposing unnecessary conservatism.³,⁹
When supported by documented risk assessments, validation studies, and process understanding, lifecycle-based microbial control strategies are routinely accepted during regulatory inspections and reduce the likelihood of observations related to hold-time justification.⁹ Table 1 summarizes typical hold times and microbial acceptance criteria by manufacturing stage
Conclusion
Hold times are not merely operational pauses; they are critical decision points at which microbial risk must be actively managed. A lifecycle-based approach that aligns hold-time duration with stage-appropriate bioburden and endotoxin acceptance criteria provides a defensible, regulator-accepted framework for microbial control. The question for manufacturers is no longer whether such alignment is necessary, but whether their current strategies accurately reflect the science and risk inherent in each stage of their process.
References
- Bergheim, S. Choosing the Best Fit Cell Culture Harvesting Technology for Your Process. Biopharma Asia (2018).
- Muralidharan, N.; et al. Validating Prefiltration Dirty-Hold Times for Upstream Media and Feed Solutions. BioProcess International 2023, 21 (9), 26–35.
- Sandle, T. Assessing Process Hold Times for Microbial Risks: Bioburden and Endotoxin. J. GXP Compliance 2015, 19 (3), 1–9.
- Dream, R.; Loxley, B. Process Hold Times: What Is It and Why Is It Important? PDA Letter (2024).
- Krause, S. Alert, Action, and Specification Limits for Bioburden and Endotoxin. PDA Annual Meeting (2015).
- Hughes, P. F. Aseptic Processing of Biological Products: Current Regulatory Issues. Well-Characterized Biological Products Annual Meeting (2016).
- Yang, H.; Li, N.; Chang, S. A Risk-Based Approach to Setting Sterile Filtration Bioburden Limits. PDA J. Pharm. Sci. Technol. 2013, 67 (6), 601–609.
- European Medicines Agency. Guideline on the Sterilisation of the Medicinal Product, Active Substance, Excipient and Primary Container. EMA/CHMP/CVMP/QWP/850374/2015 (2019).
- Parenteral Drug Association. Process Validation: A Lifecycle Approach; Technical Report No. 60; PDA: Bethesda, MD, 2013.
- ASTM International, ASTM D5465-93: Standard Test Method for Microbial Enumeration of Water Used for Pharmaceutical and Bio-Pharmaceutical Processing (1998).
- U.S. Food and Drug Administration (FDA), Pharmaceutical Microbiology Manual ORA.007 (2020).
About the Author
Naveenganesh Muralidharan is the Founder and Principal Consultant of Bench2Batch CMC Lifecycle Partners™, a consultancy specializing in end-to-end biopharmaceutical process development, validation, and CPV strategy. With nearly two decades of industry experience spanning upstream, downstream, and manufacturing sciences, he has authored numerous peer-reviewed publications and led major technical programs at leading biotech and CDMO organizations. He focuses on integrating science, statistics, and regulatory strategy to enable robust bioprocess lifecycles.
Subscribe to our e-Newsletters
Stay up to date with the latest news, articles, and events. Plus, get special
offers from American Pharmaceutical Review delivered to your inbox!
Sign up now!