Introduction
The intent of this article is to provide practical advice based upon gaps that the author has observed during investigations into numerous sterility test failures. The approach described will also apply to other types of viable microbial contamination events, such as process simulation test (media fill) failures. The author hopes that the reader will be able to avoid some of the common pitfalls that prevent “solving the puzzle,” ie, arriving at the root cause of the microbial contamination observed.
Sterility Testing Requirements
Under most circumstances, a sterility test must be performed to demonstrate that pharmaceutical products labeled as “STERILE” are not grossly contaminated with viable microorganisms (that are detectable using the media employed). It is possible to avoid performance of sterility testing altogether. If one does not have to perform sterility tests, then, to state the obvious, there is no possibility of having a test failure (contamination event) that will have to be investigated. There are only 2 circumstances in which sterility testing is not required for a sterile pharmaceutical product.
The FDA and other regulatory authorities allow parametric release instead of sterility testing for products that are terminally sterilized in the final container. In fact, the FDA Review Microbiologists prefer parametric release to sterility testing for release of terminally sterilized products to the marketplace. In addition, sterility testing does not need to be performed on stability test samples at time points (test stations) on the commercial stability protocol. Product is released using the sterility test, which is the time zero test station on the stability protocol. Thereafter, container/closure integrity testing (CCIT) can be performed in lieu of sterility testing. Maintenance of sterility is then demonstrated using CCIT (physical method) over the shelf life of the product. This makes perfect sense, because some sort of seal integrity failure would be the only way that a product can become non-sterile over its shelf life. The author has helped investigate >70 sterility test failures for stability samples. Those investigations revealed that all results were probable or proven false-positives.
Investigating Sterility Test Failures
If sterility testing is performed for one’s products, the odds are that sooner or later one will be called upon to investigate a sterility test failure (either real or false-positive). One is faced with the following question: How should a sterility test failure be investigated? Many facilities have no real idea how to begin an investigation into a sterility test growth-positive (failure) because they do not have prior experience. There is no formal plan, ie, a specific SOP with a decision tree that describes how a sterility test failure should be investigated. Many facilities use their QC OOS SOP that describes what to do for testing deviations. But that SOP is typically chemistry test oriented and usually does not provide sufficient guidance on conducting sterility test failure investigations.
In the author’s experience, sterility test failure investigations are typically flawed to some extent. For example, only negative findings are documented. Often the scope (breadth and depth) of the investigation is not sufficient to detect the root cause. Documentation does not reflect all of the efforts expended, so one cannot tell which areas were investigated by reading the investigation summary report. Often assumptions are made that preclude finding the root cause of the contaminated sterility test. Also, conclusions regarding the root cause for the sterility test contamination are made that are not supported by data. So, those “conclusions” are really speculation that could lead to performance of corrective and preventive measures that do not solve the problem (mitigate the root cause for viable microbial contamination).
Testing Laboratory Environments
Section XI “Sterility Testing” of the FDA’s 2004 Aseptic Processing Guidance states that “the testing laboratory environment should employ facilities and controls comparable to those used for aseptic filling operations. Poor or deficient sterility test facilities can result in test failure [false-positive results].” Yet many facilities still employ laminar flow hoods located in clean room suites for sterility testing of products manufactured using advanced aseptic processing, such as isolator technology. The same can be said for products that are terminally sterilized using a qualified steam cycle. In both scenarios, there is a much higher possibility of having a false-positive sterility test result than of detecting a real sterile batch failure (non-sterility). The author observed one scenario recently in which a product was manufactured using aseptic processing into glass ampules in an isolator followed by steam sterilization using a qualified cycle. The product was then tested for sterility in a laminar flow hood and the test was growth-positive for a Paenibacillus species. The investigation demonstrated that the result was a false positive due to inadequate decontamination of sterility test samples and testing materials. Environmental monitoring of the sterility-testing suite during that particular testing session recovered the same bacterium from multiple sample sites.
In the author’s experience, performance of sterility testing in clean room suites is problematic for many reasons. Sterility testing suites are often located directly adjacent to micro testing labs where numerous cultures of viable microbes are manipulated. Some clean room suites do not have an adequate air cleanliness cascade to prevent microbial ingress into the testing area. Also, sterility test samples are typically brought to the micro lab and stored until testing, creating the opportunity for superficial contamination with viable microbes.
Section XI of the FDA 2004 guidance also states that “if production facilities and controls are significantly better than those for sterility testing, the danger exists of mistakenly attributing a positive sterility test result to a faulty laboratory even when the product tested could have, in fact, been non-sterile. Therefore, a manufacturing deficiency may go undetected. The use of isolators for sterility testing minimizes the chance of a false positive test result.” The author is in total agreement with this statement.
Investigation Approach
It is difficult to support invalidation of a positive sterility test. One must have conclusive and documented evidence that clearly shows that the contamination occurred due to the testing that was performed. Key Elements of the Investigation as described in Section XI.C.1 are as follows:
- Identification (speciation) of the organism isolated from the sterility test (a strain level ID is desirable for such investigations)
- Record of laboratory tests and deviations
- Monitoring of production area environment
- Personnel monitoring
- Product pre-sterilization bioburden
- Production record review
- Manufacturing history
When performing any investigation, including one for a sterility test contamination, one must have an open mind to all possible causes for that failure. One cannot jump to conclusions, because that typically leads the investigation in the wrong direction. All evidence should be evaluated equally. One needs to document everything that was reviewed and its status, good or bad. Investigations should be viewed as opportunities to discover what is being done correctly as well as procedures/practices that are not the best.
In the author’s opinion, the following 3 simultaneous investigations should be conducted whenever a sterility test failure occurs:
- Microbiology investigation
- Manufacturing investigation
- Validation investigation
Often firms wait until the microbiology investigation has been performed to a great extent before commencing an investigation into the manufacturing areas. People are convinced that a laboratory error of some sort has occurred and that there is no issue with the aseptic filling operation. After all, the most recent media fill “passed.” Delaying the manufacturing area investigation can result in spread of viable microbes to other areas, including aseptic filling lines. Therefore, it is imperative to begin the manufacturing investigation without delay once a sterility test growth-positive has been identified. Most facilities include cursory review of re-validation experiments in the manufacturing investigation. The author recommends a separate targeted in-depth review instead. Firms should review all recent sterilization, depyrogenation, and decontamination cycle re-validations and compare them to the original qualification experiments performed. It is the author’s experience that loading patterns for sterility test isolators change over time and typically become more crowded. The increased material load within the isolator becomes a barrier to the flow of vaporized hydrogen peroxide (VHP). Also, the possibility of mated (touching) surfaces of testing materials is increased due to the crowding of materials when the load size increased. The author has seen such a scenario be the root cause for false-positive sterility tests several times around the world. So, it is very important to look for changes in production cycles or loads.
Investigative (Extraordinary) Environmental Monitoring (EM)
Review! Review! Review! When a sterility test growth-positive is discovered, most firms do a great job of reviewing historical EM data that they have collected for a particular area. However, in the author’s experience, they almost never perform any meaningful investigative EM to look for the source of the contaminating microbe. Without knowing the source(s) of the contaminating microbe it is almost impossible to arrive at a probable or definitive root cause. Furthermore, many people confuse the source of the microbe with the root cause. The root cause is how the microbe got into the product; it is not the source of the microbe. But, you really cannot have one without the other.
Investigative EM is defined as additional environmental monitoring performed during sterility test failure or other microbial contamination investigations. Samples are typically taken using swabs, because irregular surfaces and hard-to-get-to sites (nooks and crannies) need to be sampled. Samples are taken at non-routine sites, which may not have been cleaned and/or sanitized effectively. An increased sampling frequency at the non-routine sites is also required. A one-off sampling is not sufficient! In most cases one needs to perform aggressive sampling multiple times to have a realistic chance of finding the source of the sterility test contaminant. A check-the-box investigative EM of 20 samples is unlikely to be helpful and may send the wrong message to regulatory authorities; they may conclude that sufficient due diligence has not been exerted to find the source of the microbial contamination. In reality it may take hundreds of samples to locate the source of the microbial contaminant.
The following statement assumed that all aspects of the cleaning and sanitization program were properly performed, which may not have been the case:
“But, the area has been cleaned and sanitized multiple times since the batch was filled. So, we have no chance of isolating anything with the extraordinary EM that you propose.” (They were wrong; a filamentous mold was isolated within the “guts” of the filling machine where no one thought to look.)
The rate of growth of sterility test contaminants may be very slow and some types of microorganisms will never be seen during routine EM, eg, Propionibacterium acnes (microaerophillic or anaerobic) and Cladosporium species (dematiaceous mold).
Trending of EM data should help prevent product contamination. The following are negative EM trends that can contribute to a batch sterility failure if they are ignored:
- Increased numbers of viable microorganisms in critical areas—one does not have to exceed alert or action levels to have a batch failure
- New or unusual isolate(s) in the facility
- Increase in baseline microbial “load” over time
- Increase in bioburden of raw materials
- Presence of a microorganism resistant to disinfectant used in the facility
Sterility Test Isolators
Section XI of the FDA’s 2004 guidance also states, “The use of isolators for sterility testing minimizes the chance of a false positive test result.” In principle the author agrees with that statement, but isolators should not be viewed as foolproof. The operating principles of isolators can be bypassed and lead to false-positive sterility test results. Once sterility test isolators are qualified, they are often largely ignored in terms of cleaning and maintenance. Each of the following questions should be answered during an investigation into a failure for a sterility test performed in an isolator:
- How often is the isolator cleaned and sanitized?
- How often is the room in which the sterility test isolator is located cleaned and sanitized? Is a sporicidal disinfectant used?
- Who performs cleaning and sanitization of the sterility test isolator? Do they know what they are doing?
- How often is leak testing of the isolator and isolator gloves performed?
- Is EM performed for the room surrounding the isolator? How often is that done?
Proper decontamination and transfer of sterility test samples into the isolator is essential to avoid false-positive test results. Often not much thought is given to decontamination required for test samples and testing materials before they are placed into the sterility test isolator. The assumption is made that VHP by itself is adequate for decontamination of test samples. Typically 70% IPA is used for decontamination of test samples prior to placing them in isolators. However, in the author’s opinion, it is necessary to use a sporicidal agent for that purpose instead. Seventy percent IPA is not sporicidal, ie, will not destroy Bacillus spp. or filamentous mold spores. Many of the isolates from false-positive sterility tests performed in isolators were identified as spore-forming microbes that were not destroyed by 70% IPA. It is fine to use 70% IPA as a final decontamination step as the samples and materials are passed into the isolator, if those have first been decontaminated with a sporicide. In the author’s experience a 2-step procedure, ie, samples are decontaminated using a sporicide outside the isolator and are then exposed to the sterilant used for isolator decontamination, eg, VHP, is very effective in preventing false-positive sterility test results.
Summary
- For successful investigation of a failed sterility test result, one should:
- Perform aggressive extraordinary (investigative) environmental monitoring to find the source of the sterility test failure isolates
- Make no assumptions and keep an open mind
- Document everything
Citations
- 2004 FDA Guidance: “Guidance for Industry: Sterile Drug Products Produced by Aseptic Processing— Current Good Manufacturing Practice.”
- 2008 FDA Guidance: “Container and Closure System Integrity Testing in Lieu of Sterility Testing as a Component of the Stability Protocol for Sterile Products.”
- 2010 FDA Guidance: “Submission of Documentation in Applications for Parametric Release of Human and Veterinary Drug Products Terminally Sterilized by Moist Heat Processes.”
Author Biography
Dr. Ken Muhvich is the Principal Consultant for Micro-Reliance LLC, which specializes in Sterility Assurance and Regulatory Compliance Consulting. He has conducted numerous mock Prior-approval audits of sterile manufacturing facilities, including their microbiology laboratories. He is frequently involved in guiding companies in sterile process design and validation. Ken is often called upon to help lead investigations into batch sterility failures and/or to review completed sterility failure investigations to look for possible gaps. From 1992 to 1997 he was a Review Microbiologist at the U.S. Food & Drug Administration’s Office of Generic Drugs. Ken is a recognized expert in aseptic processing of sterile drug products. He holds a Master’s degree in Medical Microbiology from West Virginia University and a Doctor of Philosophy degree in Experimental Pathology from the University of Maryland.