RAPID METHODS SUPPLEMENT


established by grading each critical factor in the two categories within a relative scale, adding the grade of each factor assigned in each group and positioning each candidate on a two-dimensional graph (technical reliability vs. business benefit).


Performance evaluation and qualification of a RMM based on science and regulatory standards Compared to traditional methods, the validation of an alternative method is much more complicated because of the complexity of a RMM system and heterogeneity of micro - organisms that are required to be detected. Although regulatory and industry guidance exists6,7,10


, each RMM system has challenges that


need to be confronted during evaluation and qualification. Although validation can be daunting, it is achievable given the scope of function of the system fits the intended use and the evaluation plan is scientifically based. The validation master plan for a RMM should be designed and prepared based on a risk assessment in order to demonstrate the alternative method is valid, reliable and stable. Performance evaluation of limit of detection


(LOD) is one of the most challenging in an alternative method validation. In the evaluation of a growth based RMM, the LOD is determined and demonstrated by the test organisms, incubation condition (medium, temperature and time) and instrument sensitivity in minimum detection. The minimum time set for incubation is critical to the success of LOD validation and should be based on sound science and as referenced in the applicable standards. In conventional sterility tests, for instance, the incubation time is required to be 14 days8,11


. The reasonable equivalent time in the


RMM for a sterility test can be determined and demonstrated by both practical experimental data and theoretical growth rate calculations. In addition to the reference organisms listed in the pharmacopeia for conventional method validation, some of the slow growers such as Propionibacterium acnes, Streptomyces spp., Streptococcus spp., andBacteroids spp. should be used in RMM evaluation and qualification. The bacteria cells with stress or injured status in a sample matrix should be purposely prepared to mimic the physiological status of contaminants in real samples because a stressed cell exhibits a prolonged lag phase in culture medium3


. The cell concentration in the replicate 12


European Pharmaceutical Review Volume 16 | Issue 5 | 2011


samples spiked with stressed cells should be in the quantum range allowing minimum number of cells survived in certain fraction of the replicate samples. If a test organism does not show its growth (positive result) until 10 days later in the conventional method, the minimum time needed for its growth that can be detected by the alternative method can be considered as the equivalent incubation time to the 14 days required in conventional sterility tests. On the other hand, the equivalent incubation time in a RMM can also be theoretically estimated with a growth rate calculation. Generally, bacterial growth follows the .2


principle Nt= No formed to


n5 , which can be trans-


traditional method if the incubation conditions used (temperature and media) are different from each other. If a RMM system is non-growth based,


Applied to the traditional sterility test, Nt is the microbial population that can be detected at a time point t, which typically is 10^(6-7) cells/ml; N0


= 1, representing microbial population in a


certain volume of growth medium at the starting point ; n represents the number of


“Due to the heterogeneity nature of microorganisms, the


sensitivity of a RMM varies with different species of bacteria or


even the physiological status of a single strain”


generations needed when an organism grows from 1 to Nt


. To have a positive detection in a


100 millilitre growth medium (initial cell concentration is 1/100ml = 10^(-2) cells/ml) in a traditional sterility test, it generally takes at least 27 generations to become visible (10^6 cells/ml) by unaided eyes. The maximum generation time for the slowest grower which growth can be detected with the traditional method should be less than 0.52 days. The maximum generation time calculated for the traditional method could be used as a reference to calculate the equivalent incubation time for the alternative method according to the sensitivity (minimum detectible cell population) of the RMM instrument. Due to the nature of heterogeneity in physiological status of stressed bacteria cells, variable lag phases are predicted in replicate samples. It is difficult to have the lag phase included in the equivalent time calculation. However, one should demonstrate that the lag phase in the RMM should be equivalent or shorter than that in the corresponding


challenges that typically occur in performance evaluation are the risks of false positive/negative detection, difficulty in setting threshold, difficulty in identification of positive objects, and insufficiency of data for statistical analysis. The prominent advantages of non-growth based RMMs are their speed to results and their ability to detect organisms which are fastidious or in a quiescent state (e.g. viable but non- culturable (VBNC) organisms). However, presence of fluorescent particles could render a false positive or over estimate the result. Therefore, the result generated from a non- growth based RMM is generally expected to be higher than (or at least equivalent to) that from a parallel traditional growth-based method. Risks of false positive detection could be assessed by running sufficient numbers of negative controls using sample substrate without contamination. On the other hand, a false negative result could be generated from the viable particulates with weak and indistinctive signal which is obscured in the background noise of the system. Risks of false negative detection could be assessed by running sufficient samples spiked with a wide spectrum of organisms with different species and genus under known stress conditions as well as bacterial spores. In a RMM with fluorescence staining, studies on optimal staining conditions might be needed to minimise background due to possible staining interference by sample residues or matrix, which could streamline the process of threshold setting during performance evaluation. Identification of positive objects could be achieved by incubation followed by staining and scanning given that the lethal effect of staining and the scanning procedure to the target subjects is minimised. In the evaluation of a RMM for air monitoring, running parallel testing with growth based traditional method for a certain period of time would help to establish a microbial profile in a defined clean zone or area. Inconsistency and variable results from


replicate and repeated tests may be encountered, so it is imperative to ensure that enough data is collected for valid statistical analysis. Due to the challenge of inoculation of an exact number of bacteria cells for LOD assessment, an appropriate statistical model


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13