Microbiology Roundtable

Microbiology Roundtable

In general, what are some current issues or problems pharm/biopharma companies are dealing with in regards to microbiology testing, analysis and identification?

Christine Farrance, PhD, Director of Research and Development, Charles River Laboratories: Microbial identification can be performed with different technologies, which have different capabilities for differentiating closely-related species. The limitations and strengths must be understood when choosing the most appropriate method. Even considering MALDI-TOF and DNA sequencing, closely-related species may be difficult or impossible to distinguish. The most appropriate methods can vary based on the species. Resolution of certain taxonomic groups, such as Streptococcus canis, dysgalactiae, and pyogenes, is challenging for MALDI. Many times, these species can be differentiated using rDNA sequencing. However, it is also well understood that the phylogenetic analysis of the ribosomal gene regions is not always sufficient for species resolution. This is true of the Burkholderia cepacia complex. Sequencing information from other genetic markers such as gyrB, recA, or TUB2 can be used to increase discrimination. In many cases, species resolution cannot be achieved due to intricate taxonomy, and we have to accept an identification at the complex level as this is the most accurate answer. The validation of a new identification platform involves determining its taxonomic resolution using reference strains. If the species resolution of platforms is known, then the appropriate choice of system can be made, and the expected identification will correlate to the system output.

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!

Tony Cundell, PhD, Microbiology Consultants, LLC: In the most part, pharmaceutical companies still rely on traditional growth-based microbiological tests for in-process monitoring and product testing. These tests are inadequate in terms of controlling our ingredients, intermediates and manufacturing processes, conducting timely investigations of microbial failures, and releasing drug products to a fast moving market. Modern microbiological testing technologies are available, but our industry is risk adverse, so we avoid the challenge of validating and introducing these technologies as alternative to compendial methods and linking the updated methods to our regulatory fi lings. To make matters worse, we unfairly blame the regulators for the slow introduction of these technologies.

As a microbiologist working for big pharma companies, on numerous occasions, I was told why would we add a potential complication that may delay the approval of a new block-buster drug with potential sales of $20 million a month in its first year after approval. Microbiology is not the tail that wags the dog!

Cheryl Essex, Head of Biologics Microbiological Control, Biologics Corporate Quality, Sanofi:

  1. Extrinsic contamination of samples continues to distract for true issues
  2. Data integrity of manual analysis – when is 4 eyes verification appropriate
  3. Difficulties with rapid technology – 
    1. no clear winners
    2. very few clearly validatable or calibratable
    3. false positive and false negatives with automated plate counters
    4. development and approval taking many years
    5. microbiology staff not experts in underlining technology of new methods
  4. Fungal identification options – costly with MicroSeq, MALDI not reliable yet
  5. Misinterpretation of USP microbial limits test – some firms summing TAMC and TYMC rather than reported separately

Suzanne Williams, Manager, Bio/Pharmaceutical Microbiology, Eurofins: There appears to be a gap in how regulatory agency auditors and industry professionals are interpreting compliance requirements in microbiology. In some cases, microbiological tests may be expected to demonstrate properties beyond those they were initially designed to assess. Smaller companies or virtual companies may not have immediate, readily available or on-site access to microbiologists that can assist them when questions or issues arise, and differing regulatory approaches and findings may contribute to confusion. This can create challenges when dealing with cleaning and disinfection programs, environmental monitoring, product hold-time studies, overcoming product interference during method suitability testing and assessing whether a microorganism isolated is of a concern for their product, process or manufacturing area.

Robert Porzio, Product Manager, Endotoxin Detection, Lonza: With regards to endotoxin testing, the industry has made huge strides since the introduction of the gel-clot method; however, this testing is still inherently prone to human error. Results may vary widely from analyst to analyst, there is high potential for transcription errors due to the number of manual processes involved, and methods ultimately lack the potential for proper second person review. Not only can these factors result in failure modes, but they also introduce high levels of risk into a company’s data integrity policy and governance. Without an understanding of how to review this level of risk, and ideally eliminate it, many companies may find themselves quite vulnerable.

Paula Peacos, Senior Consultant, ValSource, LLC:

  • Ensuring data integrity and balancing preventative measures against the actual ROI. Are measures implemented truly ensuring data integrity? We are still working with very manual and highly variable methods.
  • Reducing turnaround time while maintaining quality.
  • Rapid changes in microbiological nomenclature: frequently operators, especially new operators have difficulty with this. They may not be aware of the change, and therefore may fail to catch that they have isolated an objectionable or aberrant organism. Identification software has difficulty keeping up.
  • Performing meaningful and robust risk assessments.
  • Sufficiently justifying the details of method qualifications and validations. For example, what organisms are to be tested as part of the challenge battery? Are organisms that pose a particular risk to the product, process or patient ever included, or just the standard type culture battery?
  • Lack of automation.

Looking at the latest in technology, can you tell us about some new technologies and/or processes that are available now that you are excited about? Do any of these new technologies have the potential to have a noticeable impact on pharma/biopharma company’s microbiology testing and analyzing operations?

Bindhu Verghese, PhD, Senior Staff Scientist, R&D, Charles River Laboratories: The pharmaceutical industry has embraced several microbial identification platforms to support quality systems and EM programs. To date, the application of next generation sequencing (NGS) in microbiology has been limited to whole genome sequencing and microbiome studies. Recently, NGS has been changing the landscape of medical diagnostics, especially cancer and prenatal testing, and is receiving more attention in clinical microbiology. The CDC and FDA have embraced the utility of NGS for strain typing, especially for tracking down contaminant strains during food outbreak investigations. NGS has great potential as part of pharmaceutical quality systems and EM programs, especially for culture-independent microbial identification. The major advantages of NGS technologies is that they are highly sensitive. They are capable of identifying a spectrum of organisms within the same reaction, even from a mixed sample, with no prior knowledge of the microbial population. Even though, when compared to most identification platforms, the time to result for NGS looks much longer, the time may in fact be similar as there is no need for culturing samples on different media based on the organisms of interest. However, the big question is: are we ready to see the un-culturable organisms that are in our facilities.

Cundell: The technologies that excite me the most include MALDI TOF mass spectrometry for microbial identification, LASERinduced fluorescent viable particle monitoring of cleanroom air and processing and ingredient water, rapid panel-based molecular screening methods for testing bio-fermenters for adventitious microorganisms, and whole genomic sequencing (WGS) for strain typing to support microbial investigations.

Why my enthusiasm? The low unit cost and rapidity of MALDI TOF mass spectrometry makes it suitable to identify colonies directly from the primary isolation plates during environmental and water monitoring. Viable particle monitoring of air and water may be viewed as a process analytical technology, if we can get past reacting to isolated microbial excursions that will occur during continuous monitoring. Huge manufacturing disruptions and biological product rejections have occurred due to virus contamination of cell cultures and can be prevented using molecular screening methods. With the enormous reductions in the cost of WGS, this technology will aid investigations into the source of microbial contamination.


    1. Fluorescence based real time viable air samplers, e.g. BioTrak®, BioVigilant®
      1. strong potential to accelerate investigations and aseptic behavior training
      2. capable of identifying and confirming correction of environmental issues in real time
      3. when used as a supplement to culture-based monitoring, can possibly reduce (but not replace) traditional EM in cleanrooms

Williams: Rapid Microbiology Testing – rapid sterility techniques have the ability to impact turn-around times for sterility testing and therefore facilitate faster product release/distribution (i.e., vaccines in a pandemic situation). Cell therapy products would also benefit from a more timely sterility test result.

Porzio: Many top pharmaceutical companies are now adopting Industry 4.0 practices, driven by digital transformation technologies. At Lonza, we are excited to see a move towards automating the endotoxin test and believe it will bring a range of benefits to the industry. To aid this, we have been looking for ways to enhance our endotoxin detection software, WinKQCL™ Software, to drive the automation of traditionally manual processes. By automating assays and template creation, such a technology could add metadata that did not exist previously. Essentially, this could result in improved data integrity by automating the sample lifecycle throughout the entire endotoxin testing process.

Mike Brewer, Director, Pharmaceutical Analytics, Thermo Fisher Scientific: One thing I am really excited about isn’t a new technology but rather some new developments around an older technology. There is a growing list of pharma/biopharma manufacturers who are using our proprietary MycoSEQ Mycoplasma Detection kit as a replacement for the 28-day culture based mycoplasma test. Following validation, regulatory filing and review, our customers have received regulatory acceptance to use MycoSEQ for lot release testing applications across multiple therapeutic modalities. Product quality and safety are the utmost priority for manufacturers of cell-culture based therapies. They can now meet that priority with faster lot disposition while eliminating the cost and delay of the conventional culture-based method.

Rapid testing is critical to the industry goal of real-time or near-real time lot release. Assays utilizing technological advances in automated sample preparation, multiplex PCR, and high-throughput next generation sequencing have demonstrated the ability to provide sensitive, rapid results for detection of large panels of virus considered to be threats to mammalian cell cultures. With acceptable validation this technology offers the potential to move the required viral test to a rapid molecular method. The shift in the regulatory paradigm of acceptance of these types of assays as alternatives to conventional culture-based methods moves the industry goal of accelerating lot deposition cycles closer to reality.

In the past few years have regulatory expectations from the FDA and other global agencies resulted in more scrutiny placed on microbiology processes? Do inconsistent or different global regulations place a regulatory strain on pharmaceutical companies?

Cundell: Global regulatory guidelines like those issued by the International Council for Harmonization (ICH) has been important in standardizing regulatory expectations. The ICH has ten members, including the Chinese Food and Drug Administration who joined in June 2017, and twelve observer nations. Standard settling organizations such as the International Standards Organization (ISO) are useful too, especially for air cleanliness standards in cleanrooms. The leading pharmacopeias, i.e., the USP, Ph. Eur. and J.P. have an important role to play. For example, the USP Microbiology Expert Committee has been active in writing general informational chapter to provide up-to date technical guidance on sterilization processes, container-closure integrity, and bioburden control in non-sterile manufacturing.

Regulatory inspectional finding still routinely includes observations around failure to set microbial specifications and validate test methods, inadequate investigations, lack of data integrity, failure to exclude objectionable microorganisms and a lack of sterility assurance. Different regulatory agencies reach conflicting compliance conclusions about individual manufacturing facilities after their inspections. This is disconcerting. However, the FDA membership in PIC/S and mutual inspection recognition may overcome some of this disagreement, after the dust settles.

Antimicrobial effectiveness continues to be a thorny issue. I recall that an important Asian regulatory agency when reviewing my former company’s regulatory submission requested that preservative concentration in a nasal spray be reduced because in support of the lower shelf specification we showed effectiveness at 80% of the label claim. This was after scores of national boards of health had approved the product for sale in their countries. Go figure!

In the near term, the biggest challenge to our industry may be the revision to the EU cGMPs Annex 1 that has taken restrictive positions in the area of microbiology. Examples include the continued insistence on the use of settle plates when better technologies are available and the requirement that new microbiological testing techniques, although they rely on different signals than the colony-forming unit, must give equivalent results. USPValidation of Alternative Microbiological Test Methods has more flexible and technically defensible approaches.


  1. Yes, regulatory scrutiny has increased for microbial methods/studies and data integrity, but also for the process microbial control. Specifically I have seen these concerns:


    1. Too low inoculum of challenge organisms
    2. No in-house isolates included
    3. Disinfectant Efficacy (DES) – Concern for 1 challenge organism/1 surface not meeting acceptance criteria. Misunderstanding that a disinfectant that does not reduce all bacteria to 3 logs can still be part of an effective disinfection strategy.


  1. Lack of holistic contamination control strategy
  2. Inappropriately high microbial limits
  3. Lack of microbiology expertise in bioburden investigations and risk assessment for the environment and process; leads to inadequate RCA and inappropriate CAPA
    b. Regulatory bodies all have increasing focus on the “why”; seeing that firms have strong, data-driven rationale for their strategies. Some differences are noticeable.
  1. US FDA is generally the most risk averse for microbiological issues and willing to raise concerns on “what if” scenarios, even when contamination history is low.
  2. Other global agencies focus on microbiology issues but are generally less conservative.
  3. Regulatory differences do create complexity for maintaining one set corporate

Porzio: Companies are definitely facing more scrutiny these days. In fact, so many warning letters were sent to major pharma companies for their lack of data integrity, that the FDA issued a guidance document, “Data Integrity and Compliance with Current Good Manufacturing practices” in April 2016. At the time, many instances were noted where data had been manipulated, falsified or omitted. Since then, however, many companies and international regulatory bodies have adopted data integrity approaches. These are frequently encouraged from the top down via internal company policies.


  1. Yes. For example, it is common knowledge that the FDA has hired a number of microbiologists over the last few years, and we are seeing the results of that in our inspections. Site microbiologists are addressing technical questions that they have never been asked before. I see this as a good thing. However, there are still some growing pains that need to be resolved. Sometimes there is a conflict between theoretical and practical science - the premise is scientifically sound, but it cannot be practically executed with the existing technology.
  2. I do think that inconsistent or ambiguous regulations place a strain on pharmaceutical firms. When the regulations are ambiguous, they are subject to interpretation. Sometimes the firm’s interpretation is not in lock step with that of the inspectorate, particularly when it concerns the finer details. It is this room for interpretation that can make it difficult for a firm to ensure compliance.

What advice would you give to a pharmaceutical company struggling with their microbiology processes/applications? Is there a top 10 list of items you would suggest to these companies to help them?

Cundell: A list of ten items is a bit of a challenge. My top five pieces of advice would be as follows:

  • Apply risk assessment to your formulation and manufacturing process design and microbial testing and specification setting.
  • Exclude personnel from your aseptic filling operations by the use of restrictive access barrier systems (RABS) and isolator systems.
  • Introduce modern microbiological testing technologies to take advantage of their improved quality of results, timeliness, and ability to capture data electronically.
  • Improve the quality of your microbial failure investigations and institute real corrective actions, preventative actions (CAPAs) and not just rounding up the usual suspects and recommending retraining.
  • Hire well-trained and experienced pharmaceutical microbiologists, pay them comparable salaries to other pharmaceutical professionals, empower them within your organization and listen to their advice.


    1. Start with building staff awareness, top down (leaders and management) and bottom up (shopfloor and laboratory staff – the true gatekeepers)… you will need this for all subsequent efforts to be truly effective.
    2. Get top leadership buy-in to spend the resources on correcting the microbiological issues. No remediation effort will succeed without this.
      1. For process issues – many firms have been forced to take a “no tolerance” policy to microbial failures to regain control, meaning all bioburden issues are heavily scrutinized
    3. Bring in expertise during remediation efforts; hire a dedicated Microbiology champion(s) for the site
    4. For actual remediation, go back to basics:


      1. build an SOP on how to properly validated Micro methods and then evaluate your past method validations against it
      2. repeat or supplement validations are needed; modify or replace methods as needed
      3. Process
      4. complete the fundamental risk assessment on processes, often missing for legacy products. These RA should identify all potential points of ingress and controls. If you aren’t sure if a control is working, start testing that step.
      5. accept that big changes may be needed
      1. Ban the saying “This is how we’ve always done it”

Williams: Look at internal resources and see if someone has a background in order to assist in addressing the problem.

      • Research guidance documents.
      • Reach out to industry experts and/or user groups to gain knowledge, ask questions or attend training.
      • Assess your current capabilities/capacity and consider outsourcing for any microbiological process or testing where it is a cost-effective alternative, ensuring clear communication of requirements and expectations is established with your contract laboratory or service provider.
      • When possible, break down the process into smaller steps or project to be more manageable or to help evaluate where an issue might be occurring.

Porzio: First, companies need to critically review and assess their microbiology processes in relation to their application(s) of interest. With regards to endotoxin testing, Lonza would recommend performing Failure Mode and Effects Analysis (FMEA) to identify any potential failures that could compromise data integrity. This involves performing a risk analysis on processes to identify flaws and act upon them.

The top ten actions we would recommend include:

          • Review your company’s data integrity policy.
          • Ensure uniformity by adopting the same policy throughout the company. Ultimately, senior management is responsible for establishing a culture of integrity. Training needs to be provided to raise awareness across the organization and employees need to be encouraged to openly report errors.
          • Train employees on the new policy.
          • Establish and perform FMEA to identify and address issues. Identify key people to lead data integrity investigations.
          • Review current standard operating procedures, past and current deviations, and provide recommendations for improvement.
          • Define your workflow, data flow, systems and risk points. Below is an example of a kinetic endotoxin assay workflow:
Microbiology Roundtable
        • Interview all lab personnel involved in endotoxin testing to understand each step of the process.
          Microbiology Roundtable
      • Based on the interviews and your reviews, use FMEA to identify ways of eliminating gaps and minimizing risk.
      • Build an audit trail review/schedule to identify failure modes and risk levels based on your FMEA findings.
      • Reflect on Industry 4.0 and find ways to automate assays, reduce manual transcriptions, and minimize human error.


The entire staff, including bench staff, should read and understand the documented regulations. They need to understand the basis for what they do.

      1. The entire staff, including bench staff, should also understand the process they are monitoring or performing testing for. They need to spend time on the shop floor as this will enable them to make appropriate adjustments to the micro programs (EM, trending, etc.) to maximize value. It will also improve the quality of root cause analyses etc.
      2. Ensure there is robust scientific justification for everything you intend to implement, and everything you do not. This includes things such as choice of challenge organisms etc. If you can’t justify it, you probably shouldn’t be doing it.
      3. Ensure consistent and robust training and identify high quality trainers. Your most knowledgeable person may be a terrible trainer. It is critical that the training be consistent and that correct understanding by the trainees is verified. This determines good or bad performance.
      4. Good trending involves more than just reporting data. It involves data analysis and interpretation of that analysis that lead to better, defendable decision making.
      5. Your bench staff is often the most valuable source of ideas for improvement. Encourage their participation in problem solving.
      6. Benchmarking is a useful tool, but it must be used correctly. Just because everybody else is doing something does not necessarily mean your firm should do it and vice versa. The form in which the question is posed and who is asked obviously determines the outcome of your survey. I have seen firms use benchmarking to almost seek permission to implement a particular procedure or not, and I think this is counterproductive.
      7. Don’t be afraid to challenge the status quo. If you can scientifically justify your position and show a positive effect, in my experience inspectors will generally accept it as long as it does not violate an established regulatory requirement. However, that scientific justification must be sound, logical, robust, address any potential detrimental effects, and make practical sense.
      8. If you know of an issue and you can mitigate or eliminate it, do it even if that issue has not caused you any problems to date. It is a deviation waiting to happen. It is also part of any good continuous improvement plan.
      9. If you are having a great deal of difficulty, seek assistance form someone who has expertise in that area such as a consultant, and learn from them. Know that not all consultants are created equal. It is important to get the right expert.

What do you see as the major industry critical issues over the next five years in regards to microbiology?

Kerry Falgowski, PhD, Systems Validation Specialist, Charles River Laboratories: The FDA, MHRA, and other regulatory bodies have begun to emphasize and strongly enforce the importance of data integrity. As we continue to manage and minimize our data integrity risks, many organizations will aim to convert uncontrolled and complex manual processes into controlled and simple automated processes. The implementation of such sophisticated technologies in the dawning (daunting) era of data integrity is a major concern and will require sufficient resources for several reasons. First, the manual process complexity must be absorbed by software development or other systems. Second, the systems must be designed to be compliant with the ALCOA (Attributable, Legible, Contemporaneous, Original, and Accurate) principles of data integrity. Third, the systems must be fully validated for their intended use. In addition, it is critical that organizations continue to develop and implement a comprehensive Data Integrity Governance Program. Proper data integrity training empowers employees with a sense of data ownership when they have a full understanding of the data life cycle and can identify the critical data within these systems.

Cundell: Two critical issues that have my attention are 1) the cost of emerging gene and cell therapies and 2) continued issues around the exclusion of objectionable microorganisms from non-sterile drug products. With respect to gene and cell therapies, regulators should avoid using the pharmaceutical manufacturing paradigm to this unique industry. As FDA commissioner Scott Gottlieb recently acknowledged, applying pharmaceutical cGMP directly to 503B outsourcing sterile compounding facilities that have significantly different requirements was inappropriate and specific sterile compounding cGMPs will be developed. The same will be true for cell therapies.

Although I worked long and hard with my colleague Anil Sawant in the leadership role on the PDA task force that developed the 2014 PDA Technical Report No. 67 Exclusion of Objectionable Microorganisms from Non-sterile Pharmaceuticals, Medical Devices and Cosmetics recalls continue. For example, in July 2016 the FDA announced a voluntary nationwide recall of all non-expired lots of oral liquid docusate sodium, a stool softener, manufactured by PharmaTech LLC, Davie, Florida in one pint bottles and distributed by Rugby Laboratories for contamination with B. cepacia and linked to a five state outbreak with infant deaths. Later laboratory evidence linked the B. cepacia to the company’s purified water system. More up-front risk analyses are recommended. The USP will be publishing a screening method for the absence of Burkholderia cepacia complex in the September/October Pharmacopeial Forum. This may help.


  1. Implementing modern micro methods, e.g. rapid, automated, more sensitive
  2. Routine toxin testing (exotoxin and mycotoxin)
  3. Toxin clearance studies for the process
  4. Highlight on ineffective equipment cleaning and handling leading to microbial failures
  5. Implementing and optimizing closed systems
  6. Using AI for meta-data analysis, e.g. linking relevant facts from multiple data storage systems, better real-time trending

Williams: Application of data integrity concepts will be an issue in microbiology in regard to manual processes (i.e., counting and recording data). Microbiology data handling doesn’t fit all data integrity interpretations specifically in regard to requiring a second analyst to verify a plate count or a reading of a sterility test on its last day of incubation. The impact could be that we will essentially be requiring two individuals to perform the same tasks for each test.

Porzio: As long as there is the potential for error-prone manual steps involved in microbiology processes, companies will continue facing data integrity issues. It is time to view data integrity as an opportunity to improve product quality. A paperless system like the MODA™ Solution, that enables traceability of a sample’s lifecycle data could provide significant benefits from a quality perspective.


  1. Data integrity and how to effectively (and practically) manage it.
  2. Automation- we need to develop more of it.
  3. Performing robust and meaningful risk assessments. While some firms are very adept at doing this, many are not.
  • <<
  • >>