The author, the consulting microbiologist Tony Cundell, Ph.D., examines what he considers the top five challenges to microbiologists working today in the pharmaceutical industry and suggests how the resolution of these challenges can assist our industry in both advancing public health and supporting the business success of our companies.
The challenges facing pharmaceutical microbiologists in 2006 are: 1) maintaining compliance levels in our microbiology programs, 2) promoting advanced aseptic processing technologies within our organizations, 3) increasing microbial testing productivity, flexibility, and cost effectiveness, 4) implementing rapid microbial methods, and 5) setting microbial requirements that take into account the analytical capabilities of the test methods.
Maintaining Compliance Levels in Microbiology Programs
Repositioning our microbiology programs to be in compliance with the 2004 FDA Aseptic Processing Guide and revisions to the compendial microbial tests, i.e., <61> Microbial Limit Tests and <71> Sterility Tests, is a challenge confronting pharmaceutical microbiologists. Once official, none of these documents are grandfathered. This will mean revising standard operating procedures and retraining staff to be compliant. Managers responsible for the operation and oversight of microbiology programs should arrange for compliance audits to be conducted to demonstrate an absence of regulatory risk in these areas. Also, regulatory citations arising from inspections of sterile product manufacturers should be monitored closely to detect any new compliance trends.
Promoting Advanced Aseptic Processing Technologies Within Your Organizations
With recent technology advances there are opportunities to replace steam-sterilized, stainless-steel tanks with single-use sterile plastic bags for handling sterile bulk solutions, buffers, and cell cultures, introduce advanced aseptic sampling and connection technologies to reduce the potential for product contamination, retrofit aseptic filling suites (conventional clean rooms) with grey side maintenance and restricted access barrier systems (RABS) and plan for the use of isolators in new facilities to separate people from exposed packaging components and product. All these technology changes will improve sterility assurance especially if they significantly reduce or eliminate operator interventions and product exposure [1].
The use of sterile plastic bags will eliminate the cost associated with tank cleaning and sterilization validation, the capital investment in stainless steel tanks, clean- and sterilize-in-place systems, and improve scheduling of aseptic processing operations.
The use of non-intrusive aseptic sampling and connections are becoming widely used in the biotechnology industry and certainly contributes to the prevention of microbial contamination during aseptic processing. As it is generally accepted that operators are the major source of microbial contamination during aseptic filling operations, barriers to separate operators from exposed product and packaging components clearly should be employed in the interest of public health and safety.
Increasing Microbial Testing Productivity, Flexibility, and Cost Effectiveness
Opportunities include improved scheduling, resource planning, testing productivity, laboratory investigations, and training. An informal survey of pharmaceutical companies manufacturing sterile injectable products showed that a typical product release cycle was 21 to 28 days [2]. This suggests that reducing the 14-day incubation time for the finished product sterility test by the application of rapid microbial methods, without improving the efficiency of the other activities, will not result in a significant reduction in the product release cycle time.
Also Bublitsky and Howard [3] reported the benchmarking of 15 laboratories in the biopharmaceutical industry as part of a lean manufacturing study that gives an insight into QC laboratory operations. The parameters benchmarked were staffing patterns, product release cycle time, analyst availability and utilization, productivity, quality, and support systems. The authors found a relative low labor input per test, a wide disparity between laboratories, high first time right rates, low efficiencies in scheduling tests, reviewing the results, and release product, and value in support staff to increase analyst utilization. The overall testing cycle (queue time, testing, and QA review) ranged from 5 to 34 days with the testing time ranging from3 to 15 days. Clearly the microbiological testing, because of the extended incubation times of growth-based detection, enumeration, and identification methods, is an area that would be improved by the application of rapid microbial methods. However, it appears possible to achieve 20-30% improvements in cycle times through better scheduling, resource allocation, and QA release efficiencies.
The temptation to outsource microbial testing, as a cost savings, needs to be well considered from both a tactical and strategic viewpoint, so that microbiological expertise is not stripped out from your company. Possible approaches are to look for opportunities to move in-process microbial testing into the production area, improve laboratory efficiencies for routine testing, and outsource highly technical and/or infrequently conducted microbial tests to achieve efficiencies. Furthermore, it may make good business sense to identify and qualify another microbiology laboratory, as an alternate testing site, in case there is a disaster in your laboratory.
Implementing Rapid Microbial Methods
As stated above, microbial testing, because of the extended incubation times of growth-based detection, enumeration, and identification methods, is an area that would be improved by the application of rapid microbial methods. However, since the 2000 publication of the PDA Technical Report Number 33 Evaluation, Validation, and Implementation of New Microbiological Testing Methods, rapid microbial methods have not lived up to their promise for changing routine quality control microbiological testing in the pharmaceutical industry. Recent studies in the areas of lean manufacturing cited in the previous paragraph highlighted that it is possible to achieve significant improvements in product release cycle times through better scheduling, resource allocation, and QA release efficiencies so these improvements can be considered low-hanging fruit compared to the implementation of rapid microbial methods. Recent efforts by the FDA in their GMPs for the 21st century initiative is leading to renewed interest in the application of rapid microbial methods. The advantages in developing, validating, and implementing these methods include reducing the product release cycle time, improving the quality of the microbial testing, automating the microbial testing, electronic capture of test data and information creation, the ability to start investigations earlier in response to out-of-specification (OOS) results, potentially reducing risk of microbial contamination of our products and improving the manufacturing processes, and the professional enrichment within the laboratory. Shorter product release cycle times will result in significant reduction in backorders, and pharmaceutical excipients, packaging components, ware-in-process and finished product inventories that may represent three to six months of the sales of a global pharmaceutical company.
Setting Microbial Requirements That Take into Account the Analytical Capabilities of the Test Methods
Microbial testing of pharmaceutical excipients, intermediates, and finished products are designed to answer the questions whether microorganisms are present in a sterile product (sterility testing), the number of microorganisms in a non-sterile pharmaceutical or overthe- counter drug product (microbial enumeration), and if objectionable microorganisms are in a non-sterile product (screening of specified microorganisms). The low probability of detecting contaminated units in a lot of an aseptically filled sterile drug product using the compendial sterility test with a sample size of 20 units is well known. Less well known is the poor accuracy and precision of microbial enumeration employed in microbial limit testing, water monitoring, and environmental monitoring. Chapters within the proposed harmonized USP Microbial Limit Tests acknowledge the variability of the microbial enumeration by adopting the Eur. Ph approach by allowing for a maximum acceptable limit for each dosage form. For example, a topical cream with a recommended quality criterion for a Total Aerobic Microbial Count of not more than 100 cfu per g would have a maximum acceptable level of 200 cfu per g based on a twofold or 0.5 log variability for the counting methods. In a similar vein, the USP <51> Antimicrobial Effectiveness Test defines no change in microbial counts as within a 0.5 log variability.
In their article on the analysis of microbial monitoring data from pharmaceutical clean rooms, Hussong and Madsen [4] argued that the frequency of isolation of microorganisms is more meaningful than numerical values and reviewed the early literature for assessing microbial enumerations. They rightly point out the variability of microbialcounts, that follow a Poisson distribution, expressed as % Standard Deviation (SD) is inversely related to the mean of the count according to the equation SD = ± (100/ square root of T) where T is the colony count. For mean counts of 1, 3, 5, and 10 cfu per plate the SD were 100, 58, 45, and 32% respectively, so there is not real difference between these counts. Examples of the experimental measurement of the precision of plate count methods may be found in the literature. A classic study by Ziegler and Halvorson [5] determined that the RSD of plate counts with an average number of colonies per plate in the range of 48 to 215 was 32 to 35%. More recently, Marino et al [6] reported RSDs of 15 to 35% for plate and membrane filter counts of water samples containing 25 cfu per ml.
As decisions are being made on the level of environmental control in aseptic filling areas based on air, surface, and personnel microbial monitoring, the analytical capability of these methods bears examination. For example, the current USP surface cleanliness guidelines for equipment and facilities recommend not more than 3 cfu per contract plate or swab when sampling an area of 24 to 30 cm2. A2005 draft revision to USP informational chapter <1116> Microbiological evaluation of clean rooms and other controlled environments recognizes that the discrimination between 1, 2, 3, and 5 cfu per 30 cm2 is beyond the analytical capability of the methods. The draft revision moves from numerical levels to trending the frequency of isolation. Surface monitoring methods are described in the Standard Methods for the Examination of Dairy Products but the accuracy and precision of these methods are not truly known. This was highlighted by a CDC-sponsored study investigating the efficiency in recovery of Bacillus anthracis spores from steel coupons using pre-moistened cotton, macrofoam, polyester, and rayon swabs to assist the bioterrorism response program [7]. Air-dried inocula of 5 x 104 spores were recovered from the steel coupons. Pre-moistened macrofoam and cotton swabs that were vortexed during processing recovered the greatest proportions of spores with a mean recovery of 43.6% (SD 11.1%) and 41.7% (SD 14.6%), respectively. Pre-moistened and vortexed polyester and rayon swabs were less efficient, at 9.9% (SD 3.8%) and 11.5% (SD 7.9%), respectively. The accuracy and precision of surface monitoring in clean rooms where microbiologists attempt to quantify low numbers of stressed vegetative microorganisms as wells as spores can only be worse. Furthermore, when a typical result is no recovery of microorganisms, a proportion of the isolations may be the result of the sampler contaminating the sampling devices and/or the media. An earlier Nordic collaborative study comparing the recovery of a range of microorganisms including vegetative bacteria from stainless-steel surfaces in the range of 1, 10, and 40 cfu/cm2, using a commercially available dip slide, contact plates, and swabs demonstrated recoveries in the range of 16 to 31% [8].
With respect to microbial air monitoring, Ljungqvist and Reinmuller [9] concluded in a study comparing different sampling devices that the number of colony forming units detected by one method cannot be directly compared with results from another method. Clearly, microbial monitoring to confirm continued environmental control needs to rely on monitoring to identify adverse trends, not the evaluation of individual numerical results.
References
1. Akers J. and J. Agalloco, The Akers-Agalloco Method Pharm. Technol. November 2, 2005.
2. Moldenhauer, J. Personal Communication. 2005.
3. Bublitsky, T and A. Howard Benchmarking QC Operations. Pharm. Tech. Europe, September, 2004.
4. Hussong, D. and R. E. Madsen Analysis of Environmental Microbiology Data from Clean room Samples Pharm. Technol. Aseptic Processing Supplement p10-15 2004
5. Ziegler N. R and H. O. Halvorson Application of Statistics in Bacteriology. IV Experimental Comparison of the Dilution Method, the Plate Count, and Dilution Count for the Determination of Bacterial Populations. J. Bact. 29 (6) 609- 634 1935
6. Marino, G., C Maier and A M. Cundell. A Comparison of the MicroCount Digital System to Plate Count and Membrane Filtration Methods for the Enumeration of Microorganisms in Water for Pharmaceutical Purposes. PDA Journal of Pharmaceutical Science & Technology. 54(3) 2000.
7. Rose, L, B. Jensen, A. Peterson, S.N. Banerjee, and M. J. Arduino Swab Materials and Bacillus anthracis Spore Recovery from Non-Porous Surfaces Emerg. Infect. Dis. 10 (6) 1023-1029 2004.
8. Sato S., A. Laine, T. Alanko, A.M. Sjoberg and G. Wirtanen. Validation of the microbiological methods Hygicult dipslide, contact plate, and swabbing in surface hygiene control: a Nordic collaborative study. J. AOAC 83 (6) 1357-1365 2000.
9. Ljungqvist, B and B. Reinmuller Airborne viable particles and total number of airborne particles: comparative studies of active air sampling. PDA J Pharm Sci Technol. 2000 Mar- Apr. 54(2) 112-116 2000.
Tony Cundell is an independent Consulting Microbiologist to the pharmaceutical industry working in the areas of rapid microbial methods, QC microbiological laboratory management and microbial contamination risk assessment. Previously, he was Director of QA Microbiological Development and Statistics at Wyeth Pharmaceuticals, a global pharmaceutical company, where he directed a corporate microbiology group responsible for method development and validation, the evaluation, validation and implementation of new microbiological testing technologies, and consulting with the manufacturing sites on microbial issues and a statistical group applying statistical tools to manufacturing and testing operations.
Other professional experience includes running QC testing laboratories, QA programs and market product stability in the pharmaceutical, plasma fractionation, and medical device industries. He is a nationally recognized pharmaceutical microbiologist who has published extensively in the areas of sterilization processes, environmental monitoring, the application of rapid microbial methods to Process Analytical Technology, microbial testing in support of aseptic processing and microbial identification.
Also, he chaired the task force responsible for the publication of PDA Technical Report No. 33 Evaluation, Validation and Implementation of New Microbiological Testing Methods and was a member of the task force responsible for the PDA Technical Report No. 13 Evaluation, Validation and Implementation of New Microbiological Testing Methods.
Tony Cundell is a member of the 2005-2010 USP Microbiology and Sterility Assurance Committee of Experts.