Microbiological Aspects of GMP Stability Programs
Scott Sutton, Ph.D.
In reviewing Emails from the PMFList, it seems there may be some confusion over the role of microbiological assays in the stability program for pharmaceutical and medical device products. It can only be assumed that a similar uncertainty exists for the cosmetic industry, although not evident in PMFList traffic. This short article will attempt to lay out one view of the need and extent of microbiological testing for stability programs.
* Before you stabilitarians get all cranky at me I do know that there are many more than three designs. I am intentionally leaving out freeze-thaw, light studies and all the others as microbiology rarely plays a role in these studies.
There are three separate types (at least) of stability programs to evaluate*. The first is the product development stability program, used to establish expiry dating of the product in preparation for regulatory submission. The second type of stability program is the post-market stability normally conducted as a regulatory commitment to demonstrate continued QC of the product, or as a means to extend the expiry dating of the product. Finally, microbiological evaluations play a role in raw material stability studies to demonstrate the adequacy of storage conditions for bulk excipient and API awaiting manufacture.
The chemical requirements for these stability programs are well-known. Less recognized, however, is the need to demonstrate the stability of the microbiologically-related product characteristics with time.
Stability studies are conducted under different conditions of temperature and humidity. Leaving aside humidity considerations, the incubation of product at different temperature is done for two reasons. The first is to establish stability under room temperature for different parts of the globe (1). Microbiological evaluation should be part of this examination as the formulation, particularly the more complex formulations, can behave in an unpredictable manner in terms of microbiological response. It must always be remembered that the reason microbiological analysis is being performed is that chemistry cannot predict the microbiological response. The bacteria are living organisms that respond to stimuli, sometimes very complex stimuli well beyond the question of whether a specific chemical is present at a specific level.
The second reason for incubating samples at different temperatures on a stability program is to predict the stability of the product on an “accelerated” schedule. The idea here is that incubation of the formulation at elevated temperatures will mimic the degradation of compounds seen at lower temperatures over longer periods. It has been my experience that this may work in many situations for microbiology, but when it fails to work it does so in a dramatic fashion. The wildly inaccurate results that are collected create a great deal of confusion as the “real-time” data catch up with the accelerated conditions, or falsely describe stability problems that do not materialize on the real-time studies. The microbiological tests on stability are critical in describing the product characteristics over time, but care must be exercised in the design of the program. At least for microbiology data it must always be remembered that the real-time studies are the only ones providing dependable information.
The second consideration in the design of stability programs is the testing frequency. It is not uncommon for initial intervals in the chemistry analysis to be on a monthly basis. This is wildly impractical for microbiological assays which might take 6 weeks from beginning to finalized report. In general, microbiological assays should be performed no more frequently than at the initial, 6, 12 and 24 month time points. This provides sufficient assurance of the microbiological quality of the product, and allows trending of the data (as appropriate). Demonstration of the microbiological quality is strongly suggested at the initial and terminal points of stability, but intermediate demonstration is prudent. What are you left with after 24 months if you fail a specification and the only other data point is the initial test?
Sterile products must meet the requirements for the compendial Sterility Tests. Ideally, they would be demonstrated sterile, but this is impractical given the currently available technology (2, 3). It is prudent to conduct sterility testing on stability at least on an annual basis.
FDA is also interested in the container closure integrity of packaging systems on stability (4). Several years ago a draft guidance document was even issued on the use of container closure testing rather than sterility testing as a stability assay (5). This guidance document was never finalized for a variety of excellent reasons. The topic of container/closure testing of packaging is a complex one, and will be discussed in a future newsletter.
Stability of non-sterile products is based on several considerations, and so is much more complex from a microbiological perspective than stability studies of sterile products. The first (and easiest) consideration is bioburden. There are regulatory differences in the various regions as to what specifications mean in this situation. For example, an observed value of 154 CFU/g would fail a specification of “100 CFU/g” in the USA, while passing it easily in the EU. A scientific issue also exists with bioburden numbers. It is assumed that the distribution of microbial contamination in a sample is homogeneous throughout the product. This allows sampling of 10 g from a 100 Kg batch as “representative.” However in practice it is well-established that microbial contamination is anything but homogeneous. Therefore, trending of bioburden data becomes somewhat imprecise as it is never clear whether an elevate count is due to growth of the contaminant or that the sampling event occurred in a “hot spot” of the larger batch.
A more difficult issue is the one of “absence of objectionable organisms.” This is a GMP issue not addressed by the compendia. The Microbial Limits Tests evaluate “absence of specified” organisms, rather than objectionable. This disagreement has been reviewed in the previous issues of the PMF Newsletter listed below and will not be discussed further.
For the moment, then, let’s assume that we are going to agree that “absence of objectionable” microorganism studies will require the lab to identify every unique colony type from the bioburden study and determine its identity. What value is this information in a stability study? The obvious answer is none whatsoever. A strong case can be made for determining the bioburden of the sample over time, demonstrating that it does not support the growth of microorganisms (ignoring sampling issues of a heterogeneous mixture for the moment). However, the identity of the organisms is of less value. Of course, it is possible to conceive of a situation where a pathogenic organism is present in low numbers and the product provides a selective pressure in favor of the pathogen’s growth in preference to all other organisms yielding a horribly contaminated product after a period of time. However, following the bioburden provides a measure of control over this situation, and little additional protection is likely to be offered by the dramatic increase in expense involved in identifying every organism on stability.
The final consideration for non-sterile stability studies is that of the product’s water activity. This characteristic is well-known in the food industry for its effect on stability (6), it is becoming more established in the pharmaceutical arena (7, 8). In fact, a new chapter in USP was released (second supplement to USP 29, 2006) on the topic under the title “<1112> Application of Water Activity Determination to Nonsterile Pharmaceutical Products.” The water activity of a product is not, in and of itself, a microbiological parameter, but it is a strong indicator of the ability of that formulation to support the proliferation of microorganisms and should be considered in designing stability test schedules (9). It is important to remember in this analysis that low water activity does not necessarily result in cell death, but can be effective in the prevention of growth.
Multiple use products must be protected from proliferation of adventitious contamination. That is, they must be preserved. The standard method to demonstrate preservation of a formulation is the antimicrobial efficacy test (AET). This test is a suspension test, where challenge organisms are suspended in the product to be tested and their survival determined with time. A standard format for this test is to individually suspend 4 or 5 challenge organisms to a final concentration of approximately 106 CFU/mL and check for survivors at 6 hours, 24 hours, 48 hours, 7 days, 14 days and 28 days. The multiple time points allow for determination of kill rate against the organism, and the organisms are selected to provide a range of responses to the preservative system. The use of these intervals may allow the data to be used in all regulatory regions.
The antimicrobial effectiveness test is one that can provide a great deal of information on a stability program. While each organism’s response might not be illuminating, it is likely that at least one of the organisms will provide useful information. Unlike the chemical assay for identity and for concentration of the preservative, the AET evaluates the biological activity of the entire formulation. It is clearly a superior test for preservative activity to those available by HPLC. Unfortunately there seems to be a perception that chemical stability of the preservative moiety is directly related to the microbial performance. While this is generally true, the exceptions can lead to spectacular situations for the head of the microbiology group.
Excipient and API Stability
While it seems obvious, the storage conditions and expiry dating assigned to raw materials must be supported. Raw materials for product formulation are normally in an non-sterile condition, and the concern is always present that the microbial population could cause spoilage of the material. Depending on the material, it might be prudent to check the Total Aerobic Count and Total Yeast and Mold Count on an annual basis as part of the raw material stability program.
Of course, not all raw materials need to be tested. Some are inhospitable to life due to extremely low water activity – some by extremes of pH. These types of materials may confidently be assayed by purely chemical analysis as long as the rationale is documented and is scientifically justifiable.
Do You Want to Learn More?
Experts at the Microbiology Network are ready to assist with consulting or training to meet your needs. Have customized, in-house assistance with your questions from our recognized experts at your schedule either through consulting agreements, in-house training or customized webinars – contact us using the “Let’s Talk” communication found in the upper right of this page.