17 RAPID STERILITY TESTING USING ATP BIOLUMINESCENCE BASED PALLCHEK RAPID MICROBIOLOGY SYSTEM

Claudio Denoya, Jennifer Reyes

Pfizer Global R&D Groton, CT USA

Maitry Ganatra and Daniel Eshete

Pall Corporation Port Washington, NY USA

INTRODUCTION
The quality attributes of manufactured pharmaceutical product include the physical, chemical, and microbiological characteristics of the raw materials, excipients, active pharmaceutical ingredient (API) as well as the final drug product (Table 17.1). Absence of microbiological contamination is considered a critical quality attribute due to its potential to dramatically impact, directly or indirectly, the safety and/or the efficacy of the drug product. Sterile, is a medieval word derived from the Latin sterilis (unfruitful), meaning, in modern terms, free from living germs or microorganisms. In pharmaceutical manufacturing, it is critical to 433

434

Rapid Sterility Testing

Table 17.1 Definitions of quality attributes Quality Attribute (QA) A physical, chemical or microbiological property or characteristic of a material Key Quality Attribute (KQA) potential to impact product quality or process effectiveness associated analytical method Critical Quality Attribute (CQA) directly or indirectly impacts the safety or efficacy of a drug product

assure sterility of sterile products in order to release aseptic and safe medicines to patients. Therefore, the importance of adequate and effective microbiological controls cannot be overstated enough. Sterility testing is performed to evaluate a finished pharmaceutical product as a batch release quality control test by following the requirements delineated in the compendia (USP <71>, 2011a; EP Section 2.1.6, 2010a; JP, 2006). The test is used to determine the presence or absence of viable, multiplying microorganisms (bacteria, yeast and fungi) under standardized growth conditions. As the sterility test is a very exacting procedure, it is performed by qualified personnel under tightly controlled environmental conditions where strict asepsis is ensured, maintained and monitored (USP, <71> 2011a; FDA, 2004). Current harmonized compendial sterility test methods using either membrane filtration or direct inoculation require at least 14 days of incubation. In cases where drug products either possess an intrinsic turbidity, or because of their formulation become opaque or cloudy during the incubation period, identification of microbial contamination based on visual confirmation of turbidity of growth media becomes difficult. In such instances, at the end of the 14-day incubation, a portion of the sample is sub-cultured into fresh medium for an additional 45 days to allow detection, further extending the incubation period. This and similar transfer/dilution

ATP Bioluminescence-based Pallchek

435

steps used for sterility testing of suspension products are time consuming and may compromise test integrity by introducing additional risk for contamination. In general, the complexity of the test procedure and required lengthy test period contribute to increased cost of manufacturing. The replacement or the supplement of the conventional sterility test by a rapid microbiology test will have significant benefits. A rapid method has the potential to produce test results much faster at enhanced sensitivity. In addition, a test based on current technologies can increase throughput and allow better data handling. The shortened time to results would allow the reduction of warehousing, a timely distribution of the drug product to the market, a better understanding and control of the manufacturing process and most importantly, a greater assurance of product safety. Currently, a number of alternative rapid microbiology methods that are based on various biophysical principles including ATP bioluminescence, nucleic acid amplification, vital dye/auto florescence, pH change and spectroscopic detection, are available. An ideal rapid method for sterility testing of a particular group of products, in addition to being equivalent and/or better than the traditional method, must be simple, economic and relatively easy to implement.

SELECTION OF A RAPID MICROBIOLOGICAL METHOD SUITABLE FOR STERILITY TESTING

In the selection process that led to this work, the following initial questions and supporting information were considered. How fast is the rapid method? How much effort is needed to implement? How much is the capital investment? How many samples (throughput) can be processed in one shift? References and previous submissions Vendor support Validation parameters.

436

Rapid Sterility Testing

A large number of alternative microbiological test platforms were evaluated, trying to identify the best match between QC and manufacturing process requirements and the capabilities of a particular rapid assay. Some of the characteristics considered in the rapid system selection process are summarized as follows. A method compatible with an initial membrane filtration step therefore allowing flexibility in the sample volume. Good references from previous customers and excellent vendor technical support. Relatively low maintenance and cost. Portable, requiring minimum laboratory bench space. Simple training and ease of use. Easy verification of proper system operation prior to use. Standard specific kit allowing corroboration of calibration and operation. Results in units that can be reasonably correlated to Colony Forming Units (CFU) of the traditional method. Previous validation experience. Detection of viable cells, including those that cannot grow in the compendia media or they could be injured or stressed as a result of manipulations or sanitization procedures applied during manufacturing (i.e., viable but non culturable (VBNC)), assuring better QC. A method compatible with the addition of a short growth step (enrichment) prior to assay to differentiate viable and culturable cells from those VBNC. Simple qualitative (presence/absence) procedure. Sensitivity down to one CFU-equivalent (see below) with enrichment. Minimal sample manipulation.

ATP Bioluminescence-based Pallchek

437

Upon consideration, the Pallchek Rapid Microbiology system (Pall Life Sciences, Port Washington, NY) was selected because it matched most of the criteria described above. For our application, where very low levels of microbial contamination were expected in the samples submitted to the sterility test, the incubation of the membrane filter in liquid growth medium (enrichment) prior to the bioluminescence assay processing, mimics the traditional test (i.e., a growth based method using similar media and incubation conditions), making the comparability study easier to interpret. This enrichment step has the intention to allow detection down to one microbial cell or cell aggregate (i.e., the equivalent of 1 CFU detected in a conventional agar plate method) present in the original sample in a period of time shorter than the 14 days required in the traditional method for a broad range of microorganisms. The work detailed in the next sections was designed to demonstrate the previous assertion and, if verified, determine the shortest incubation time needed for detection of very low counts of a broad range of microorganisms.

DETECTION OF MICROBIAL CONTAMINATION USING ATP BIOLUMINESCENCE

The use of adenosine triphosphate (ATP) bioluminescence in the microbiological evaluation of pharmaceuticals is well established. The early acceptance of the technology by major regulatory agencies has led to the use of ATP bioluminescence based microbiological test applications on an increasingly diverse group of pharmaceutical samples (Stanley, 1989; Kramer et al., 2008; Nielsen and Van Dellen, 1989; Denoya et al., 2010). A luciferin-based bioluminescence reaction makes use of the firefly enzyme luciferase (luciferin 4-monooxygenase, EC 1.13.12.7), which catalyzes the oxidation of D-luciferin in the presence of ATP, magnesium ions and molecular oxygen (Figure 17.1). Adenosine triphosphate, which is the prime energy carrier in all living cells, constitutes the main driver of the bioluminescence reaction (Lehninger, 2008). The reaction yields a quantum of yellow light at 564 nm for each molecule of the luciferin substrate oxidized (Chappelle and Levin, 1968; White et al., 1961). Light produced from the luciferinluciferase reaction is proportional to the amount

438

Rapid Sterility Testing

of ATP used and can be measured in Relative Light Units (RLU) by a luminometer. In microbiological tests using bioluminescence, where microorganisms are the source of ATP in the reaction, the luminescence measured in RLU can be correlated to the number of microbial cells present in a sample.

Figure 17.1 The luciferin/luciferase bioluminescence reaction [Luciferase, Mg2+]

ATP + Luciferin + O2 AMP + Oxyluciferin + CO2 + PPi + Light (Bioluminescence)

PALLCHEK RAPID MICROBIOLOGY SYSTEM

The Pallchek Rapid Microbiology System selected for the application described in this chapter consists of a compact, portable luminometer and reagent kits. The system is versatile and allows measurement of microbial contamination on a membrane, in liquid samples as well as measurement of surface contamination using swabs (Pall Life Sciences, USTR 2358). Microbial detection by the system is based on the luciferin-luciferase enzyme/substrate system that uses an enzyme isolated from the firefly, Photinus pyralis (White et al., 1961). In the most common application, measurements are usually performed on filtered liquid samples. After filtration of the sample, the membrane filter can then be washed to remove any components present in the sample that could interfere with the ATP-based bioluminescence measurement. In addition, the membrane filtration step and subsequent wash eliminates any free ATP that could be present in the sample. The latter assures that only ATP located internally in the potential microbial contaminants is detected. A reagent is then directly added to the membrane to release the intracellular ATP from microorganisms and make it available for a second reagent containing the luciferinluciferase reaction mix. The resulting light is detected and measured with the Pallchek Luminometer, and reported as RLU. For applications like sterility testing, where very low levels of microbial contamination is expected, the membrane can be incubated in liquid growth medium

ATP Bioluminescence-based Pallchek

439

(enrichment step). This enrichment step allows detection down to one cell (the equivalent to one CFU as counted on a traditional agar plate) for a broad range of microorganisms.

VALIDATION OF AN ALTERNATIVE MICROBIOLOGICAL METHOD

The principal aim of the validation of an alternative microbiological method for the purpose of using it to test pharmaceutical products is the demonstration of its equivalence or superiority to the corresponding compendia test method. The validation process identifies risk areas in the test method and challenges them appropriately to evaluate and define the operating parameters and performance characteristics of the test system. Guidance on validation approaches can be found in the compendia (USP <1223>, 2011b; EP Section 5.1.6, 2010b) as well as industry guidance and vendor support documents (PDA, 2000; Pall Life Sciences USTR 2359). The objective of this chapter is to discuss an initial validation parameter study of a bioluminescence-based rapid sterility test method and the subsequent feasibility study conducted as part of the development of a rapid sterility assay for a specific drug product. The drug products under study posses a special challenge for the traditional sterility test method on account of its turbidity. The following sections describe the practical considerations that were taken into account in developing the rapid sterility test using Pallchek Rapid Microbiology System based on membrane filtration, as well as the results of the feasibility study.

CHALLENGE MICROORGANISMS, MEDIA AND GROWTH CONDITION

Pharmacopoeia referenced strains derived from reconstituted BioBalls (bioMrieux, Hazelwood, MO) as well as environmental isolates from a number of manufacturing sites that represent known contaminants were used as challenge organisms. The range of microorganisms consisted of yeast, mold, Gram positive and Gram

440
Organism Escherichia coli Aspergillus brasiliensis Bacillus subtilis Candida albicans Clostridium sporogenes Pseudomonas aeruginosa Staphylococcus aureus Bacillus subtilis Acinetobacter baumannii Aspergillus fumigatus Methylobacterium rhodesianum Micrococcus luteus Paenibacillus macerans Penicillium chrysogenum Ralstonia pickettii Propionibacterium acnes

Rapid Sterility Testing

Description Gram negative Mold Gram positive Yeast Gram Gram Gram Gram positive negative positive positive Strain ATCC 8739 ATCC 16404 ATCC 6633 ATCC 10231 ATCC ATCC ATCC ATCC 11437 9027 6538 6633 Media TSB TSB Incubation Temp. (C) 3035

Table 17.2 List of microorganisms and growth conditions

2025

FTM

3035

Gram negative Mold Gram negative slow grower Gram positive Gram positive spore former Mold Gram negative Gram positive, anaerobic, slow grower Gram negative, anaerobic

Environmental

TSB

2025

Environmental

FTM

3035

Bacteroides vulgatus

negative bacteria (aerobic and anaerobic) to represent a wide spectrum of possible microbial contaminants. The use of reference microorganisms offers various advantages, including comparability of data and commercial availability in convenient formats. However, it is important that studies also include organisms that reflect the typical bioburden of manufacturing environment and the test material to allow proper interpretation of test data. In total 16 strains were used in these studies (Table 17.2). The cultures were diluted to

ATP Bioluminescence-based Pallchek

441

<1, 5 or 50 CFU per 100 L as required. Growth promotion of nutrient media, as well as identification of challenge organisms were performed as appropriate. In principle, any growth media and incubation condition may be used and validated for a method which utilizes the Pallchek Rapid Microbiology System. However, it is advisable to restrict growth conditions to those that are well established and recognized by regulatory authorities, unless a different growth condition offers a special advantage. The following commonly used media and rinse agents were used Tryptone Soy Agar (TSA), Sabouraud Dextrose Agar (SDA), Fluid Thioglycollate Medium (FTM), Fluid A and Sodium Chloride 0.9% w/v Solution. Samples were filtered using GN-6 membrane (0.45 m) (Pall Life Sciences PN 4800). The use of disposable MicroFunnel filter funnels allows the filtration of sample, washing of filter membrane and incubation of the membrane in growth media all in one filter unit, without the need to transfer the filter membrane. MicroFunnel filter funnels are convenient to use and also minimize the risk of contamination.

COMPONENTS OF PALLCHEK RAPID MICROBIOLOGY SYSTEM

Pallchek Rapid Microbiology System consists of the following: Pallchek Luminometer (PN13673) Aluminium Test Plate for Pallchek Luminometer (PN 13679) High Sensitivity Bioluminescent Kit (PN 7142) High Sensitivity ATP Correlation Kit (PN 7150) Disposable Sample Holders for Liquid Samples (PN 7147) Spreaders (PN 7149).

442

Rapid Sterility Testing

BIOLUMINESCENCE ASSAY REQUIREMENTS OF TEST ENVIRONMENT

Detection of light by the Pallchek Luminometer from exogenous sources may contribute to a high background signal that might interfere with measurement. In order to reduce background signal and reagent decay, the luminometer was operated under subdued light. Tests were run under aseptic conditions (gloves and laminar flow cabinet) to eliminate external microbial contamination, and were performed at room temperature (1822 C). A standard suitability test performed at the beginning of each test session verifies that the ambient light is adequately controlled and that test components do not generate significant background signal. Tests were conducted as described by the manufacturer (Pall Life Sciences USTR 2358).

DRUG PRODUCT SAMPLE

Product samples from a stability study program were used in this study. Analytical studies conducted on the product indicated that its chemical composition was acceptable, and samples were sterile. Drug Product (1 mL vial) is an aqueous, white and cloudy suspension.

PRESENCEABSENCE TEST WITH ENRICHMENT

Sample preparation included a filter pre-rinse, sample filtering of up to 100 mL volume and a post-filtering wash with Sodium Chloride 0.9% w/v Solution. Following the last wash, a volume of sterile media (usually 10 mL) was added to the MicroFunnel filter funnel. The container was then incubated (enriched) at an appropriate temperature for a time period that would allow the detection of a very low count of microbial contamination by the bioluminescence assay. Negative controls were processed in parallel. Following the enrichment period, an aliquot of 8 mL was filtered in a final volume of up to 100 mL with sterile water. The remnant 2 mL was saved for microbial identification. An additional 100 mL wash was occasionally performed to remove any excess components of the filtrate. The filter membrane was removed from the MicroFunnel filter funnel aseptically and placed on a sample

ATP Bioluminescence-based Pallchek

443

holder, and bioluminescence measured as described by the manufacturer (Pall Life Sciences USTR 2358). The output in RLU was recorded. The time it took to obtain a reading was less than one minute including the reagent addition. The presence of microorganisms is determined when the RLU value is greater than the background RLU value as determined, based on a predefined threshold value. For studies that required multiple sampling from an enrichment culture, a proportionately larger volume of sample and media were used for incubation.

VALIDATION STRATEGY OF RAPID STERILITY TEST

The overall strategy to validate the rapid sterility method using bioluminescence included: studies to determine the system suitability of the luminometer, reagents and media the establishment of background values for the different experimental conditions to be used in the protocol the validation tests using either ATP standard solution, reference and environmental microbial isolates and pharmaceutical samples.

SYSTEM SUITABILITY TESTING

A system suitability test is generally performed before each reading session to ascertain that the test environment, as well as equipment and reagents are functioning as intended. These tests include the determination of the level of background noise, negative and positive control tests for reagents, and negative control for the measurement of a filtered liquid sample. A typical suitability test result is shown in Table 17.3.

444

Rapid Sterility Testing

Table 17.3 System Suitability Test Sample Aluminum Plate Aluminum Plate + Sample Holder Aluminum Plate + Sample Holder + Pallchek Reagents Enzyme activity; positive control: Aluminium plate + sample holder +100L ATP (109 M)+ 100L Enzyme TSB Negative Control (membrane filtrated, two washes of saline) RLU 11 (<20) 11 (<20) 14 (<80) 105 (105) 220 ( 1000)

Measurements are the average of duplicates. Acceptance criteria are shown in parenthesis

ESTABLISHMENT OF BACKGROUND VALUES

Interference from background luminescence should be minimized as much as possible. Type of filter membrane, media and rinse fluid and amount must be investigated and selected accordingly to define conditions that provide the lowest background. Pall GN-6 Metricel MCE membrane and saline washes generally give low background signals (<100 RLU). Once the optimal conditions that give low background were identified, a control sterility test was conducted, using either TSB alone or in combination with sterile pharmaceutical samples incubated for not less than 24 hours at 35C (true negative) to determine the background in conditions close to those used during actual test. These conditions were similar to those employed during the enrichment phase of the rapid sterility method (see below). The establishment of this value is critical since the threshold value is set based on the background value. The binary designation of samples as positive or negative is based on whether an RLU reading obtained is above or below this threshold value. This value is determined by considering the acceptable level of risk of false positive, primarily a business risk, and false negative, a potential health risk. The threshold

ATP Bioluminescence-based Pallchek

445

value was established at a RLU value higher than eight times the average background level (Table 17.4). A value above this threshold value indicates presence of microorganism.

Table 17.4 Background Luminescence value Mean Standard Deviation 95% Confidence Interval Threshold RLU value 131.4 60.5 103.1159.7 1000

Sterile TBS samples were incubated for 24hrs at 35C, membrane filtered, washed with sterile saline and assayed.Values are expressed in RLU, n= 20.

INITIAL VALIDATION PARAMETERS OF THE QUALITATIVE RAPID METHOD

Validation parameters were identified and evaluated based on compendia requirements (USP <1223>, 2011b; EP Section 5.1.6, 2010b), and the recommendation of the Parenteral Drug Association (PDA, 2000). The initial validation experiments were carried out using the ATP standard solution provided in the Pallchek kit. These tests did not include the use of microorganisms or pharmaceutical samples.

Linearity
The bioluminescence based sterility test as outlined here is a qualitative presenceabsence test. Test for linearity is a validation parameter generally reserved for a quantitative assay (USP <1223>, 2011b). However, the designation of presence or absence using Pallchek depends on an initial quantitative determination of RLU by the luminometer. It is therefore important to determine if within the range of the detection limits of equipment, a predictable relationship exists between RLU and concentration of ATP. Figure 17.2 shows a typical RLU/ATP correlation curve where a very good linear relationship is observed (R2 > 0.95).

446
Figure 17.2 Linearity

Rapid Sterility Testing

ATP Correlation Curve by single user

Ruggedness and robustness

The ruggedness and robustness of an ATP bioluminescence based test in part can be evaluated using ATP standards and generating RLU/ATP correlation curves under different test conditions. RLU/ATP correlation curves performed by different operators on different dates produced comparable results showing the tests rugged nature (Figure 17.3). The test for robustness involves the introduction of small but deliberate changes in test parameters to determine the effect on the assay performance. ATP standard and reagent solutions are recommended to be kept refrigerated at all times. During a routine test both ATP and reagent preparations are exposed to room temperature for various amounts of time. Figure 17.4 shows the RLU/ATP correlation curve performed using ATP solution and luciferase enzyme deliberately left out at room temperature for varied periods of time. The robustness of the test is suggested by the comparable correlation coefficients obtained under different test conditions (Figure 17.4). At room temperature, the diluted ATP standard solutions, as well as the luciferase enzyme, remained stable up to at least four hours, allowing flexibility in the set up of the sterility test. When properly kept refrigerated, reconstituted reagent is stable for at least one week (not shown).

ATP Bioluminescence-based Pallchek

Figure 17.3 Ruggedness

447

ATP correlation curves by multiple users, R2 shown in parenthesis

Figure 17.4 Robustness

ATP and enzyme stability: correlation curve using ATP and enzyme preparation exposed to room temperature for various times, R2 is shown in parenthesis

448

Rapid Sterility Testing

Bioluminescence undergoes decay with time. The signal is more stable at higher ATP concentration and starts to diminish after 30 seconds particularly at lower ATP concentrations. The data are consistent with those obtained by Pall (Pall Life Sciences). It is therefore very important to perform the detection within 10 seconds of the initiation of the reaction to assure accurate and repeatable measurements (Pall Life Sciences USTR 2359) (Figure 17.5).

Figure 17.5 Robustness: bioluminescence decay

Specificity, limit of detection and repeatability

Additional validation parameters were investigated in the presence of reference and environmental isolates. The objective of this part of the study was to determine the time required to detect low microbial counts by the bioluminescence assay. A typical test design and results are shown in Table 17.5 for Ralstonia pickettii. As seen in Table 17.5, the rapid sterility test method was able to detect a starting count of 3 cell of R. pickettii within 24 hrs.

ATP Bioluminescence-based Pallchek

Table 17.5 Detection of Ralstonia pickettii
Dilution Spiked (1 mL) Time 0 cell count (CFU) 1 day enrichment Pallchek (RLU) 1 day enrichment plate count (CFU) 14 day enrichment Pallchek (RLU) 3.2 102 NP NP NP NP 2.7 105 3.4 102 3.2 102 14 day enrichment plate count (CFU)

449

Visual growth observed T 1/T 14

TSB negative control 105 106 107 108 109 1010 1011

0 3 x 104 3 103 3 102 3 102 3 0.3 0.03

7.3 1.4 1.3 9.3 9.7 1.6 9.2 9.3

102 106 106 105 105 106 102 102

0 TNTC TNTC TNTC TNTC 7.5 107 0 0

0 TNTC TNTC TNTC TNTC TNTC 0 0

No/No Yes/Yes Yes/Yes Yes/Yes Yes/Yes Yes/Yes No/No No/No

NP: Not Processed TNTC:Too Numerous To Count

The results shown in Table 17.6 correspond to a study using the Acinetobacter baumannii culture. The rapid sterility test method was able to detect 5 cells of A. baumannii within 17 hrs, which was faster than the time to visual detection (Table 17.6). No visual growth was observed at 17 hrs in the enrichment sample spiked with 5 cells of A. baumannii (time to visual observation of growth was within 48 hrs). However, the sample was confirmed to be positive by Pallchek Rapid Microbiology System at that earlier time. In summary, the Pallchek Rapid Microbiology System provided faster sterility results for A. baumannii than the traditional sterility method of visual observation for microbial growth. Contamination by all challenge organisms was detected in all instances (specificity) at a very low microbial inoculum level. Table 17.7 summarizes the time to detection for various microorganisms. The length of incubation time (enrichment) for a starting count of 15 cells of any of the listed microorganisms required to achieve a detection using the bioluminescence assay was 48 hours. Under similar conditions, the time to visual detection for the traditional method was 96 hours.

450

Rapid Sterility Testing

Table 17.6 Detection of Acinetobacter baumannii

Dilution Spiked (1 mL) 0 hour cell count (CFU) 17 hr enrichment Pallchek (RLU) 7.1 102 1.9 106 1.0 . 107 2.7 106 1.3 107 4.8 102 7.0 102 17 hr enrichment plate count (CFU) 48 hr enrichment Pallchek (RLU) 2.4 102 NP NP 4.0 x 106 3.2 x 106 2.3 102 4.8 102 48 hr enrichment plate count (CFU) Visual growth observed 17 hr/48 hr

TSB negative control 106 107 108 109 1010 1011

0 4200 420 42 4.2 0.42 0.042

0, 0 TNTC TNTC TNTC TNTC 0, 0 0, 0

0, 0 TNTC TNTC TNTC TNTC 0, 0 0, 0

No/No Yes/Yes No/Yes No/Yes No/Yes No/No No/No

NP: Not Processed,TNTC:Too Numerous To Count

Table 17.7 Time to detection of samples spiked at time 0 with 15 cells Organism Time to detection (hrs) Pallchek 24 20 20 20 48 48 28 48 17 24 22 22 24 18 22 17 Time to detection (hrs) Visual 24 20 20 20 96 48 28 96 48 24 24 48 24 48 72 48

Escherichia coli Bacillus subtilis Staphylococcus aureus Pseudomonas aeruginosa Aspergillus brasiliensis Candida albicans Bacteroides vulgatus Propionibacterium acnes Paenibacillus macerans Ralstonia pickettii Micrococcus luteus Methylobacterium rhodesianum Bacillus pumilus Penicillium chrysogenum Aspergillus fumigatus Acinetobacter baumannii

ATP Bioluminescence-based Pallchek

451

The next two sections describe the feasibility studies of the rapid sterility test using pharmaceutical samples (eight commonly-used excipients and one drug product). Sterile samples were spiked with inocula containing a low count (<5 cells) of a number of compendial reference and environmental strains, and tested after various times of incubation (enrichment) either by a traditional method (membrane filtration and/or plate count) or the rapid method (bioluminescence).

EVALUATION OF THE RAPID BIOLUMINESCENCE TEST IN THE PRESENCE OF EXCIPIENTS

Excipients are used in a wide variety of drug formulations. Their compatibility with the test system widens the spectrum of the applicability of the Pallchek system. The presence of any inhibitory activity present in commonly-used excipients that could affect either the growth of microorganisms or the bioluminescence assay in the rapid sterility procedure was evaluated using Gram positive and Gram negative microorganisms, yeast and mold. Microbial cultures were exposed to each excipient (1% in TSB) for 30 min at room temperature. After incubation, samples were processed as described earlier. No significant deleterious effect was observed on microbial growth or the bioluminescence reaction (Table 17.8).

Product specific feasibility study

Drug product was prepared by pooling samples to a total of 10 mL. An aliquot of DP was added to a sterile bottle (250 mL) containing 150 mL of either FTM or TSB. The preparation was spiked with <5 microbial cells and incubated for up to five days at the appropriate temperature. A 10 mL aliquot was taken at various time points and was analyzed using Pallchek Rapid Microbiology System. Most of the challenge organisms were detected by the end of 48 hrs. Isolates of M. luteus, and R. pickettii though could be detected at 48 hrs, the peak RLU reading was reached later. A. fumigatus was detected at 120 hrs which was the only time point the culture was sampled after 17 hrs. However data from other experiments have shown that the organism can be detected as early as 72 hrs (not

452

Rapid Sterility Testing

Table 17.8 Bioluminescence assays conducted in the presence of excipients (1% in TSB)
Excipient Gram Positive B. subtilis S. aureus Gram Negative E. coli P. aeruginosa S. enterica         Yeast/Mold C. albicans A. brasiliensis

Microcrystalline Cellulose Lactose Gelatin Magnesium Stearate Ethanol Polypropylene Glycol Mannitol Starch

       

       

Checkmark represents successful detection comparable to parallel controls run in the absence of excipient

Figure 17.6 Detection of Microbial Contamination in DP1

Test samples were spiked with 5 cells of challenge microorganism

ATP Bioluminescence-based Pallchek

453

shown). These results were reproduced over several experiments. In the presence of DP, M. rhodesianum could not be detected at the end of 120 hrs of incubation (Figure 17.6). An additional attempt to detect M. rhodesianum in the presence of DP was done by running parallel samples incubated either in TSB or in FTM. A weak value barely above the threshold value was obtained after 120 hours of incubation in FTM. A control culture of M. rhodesianum in TSB alone was also run and confirmed the identity of the isolate by colony morphology and Gram stain. All suitability controls and confirming plate counts were within expected ranges for all of the above experiments.

SUMMARY
Release sterility testing is the critical quality control test that defines the acceptability of a manufactured drug product as aseptically safe to be administered to a patient. This assay, therefore, must be designed, validated and executed following the most stringent QC guidance and aseptic techniques. Technicians competence, the ability of media used in sterility test to support microbial growth, the conformity of the test environment to the requirements of the Pharmacopoeia in terms of viable microbial air and surface counts must be demonstrated and documented. In addition, procedures for sampling, testing and follow-up must be defined in the validation procedures. The validation and implementation of a rapid method must also comply with the previous requirements and new parameters must be validated so that they relate more directly to the characteristics inherent to the mechanism of detection of the new assay. In addition, important consideration must be assigned to demonstrate that the new procedure does not have any chance of producing either false positives or, very importantly, false negatives (FDA, 2004). Based on the data presented and discussed in this chapter, the Pallchek system constitutes a reliable rapid method for sterility testing that is equivalent or better than the traditional assay. Overview schematics of a rapid sterility test using the Pallchek Microbiology System is shown in Figure 17.7.

454

Rapid Sterility Testing

Figure 17.7 Overview of sterility test using Pallchek Rapid Microbiology System

The results of the feasibility study on the application of the rapid microbiological Pallchek Rapid Microbiology System to a rapid sterility test for a selected drug product were shown to be very encouraging. This drug product is a suspension which confers turbidity to the culture media and prevents visual inspection at the end of the 14-day incubation period. According to the traditional procedure, an aliquot of the incubated sample is inoculated into fresh, sterile media at day 14 and further tested for an additional five days. Since the inherent turbidity of the product does not interfere with the bioluminescence assay, the use of the Pallchek Rapid Microbiology System allows readings in samples from the original incubation container and detects contaminants in a period of time as short as five days, instead of the required 14+5 days prescribed in the traditional standard operating procedure (SOP) developed for this drug product. Of the 16 reference and environmental isolate microorganisms tested all but one (M. rhodesianum) were detected by 120 hr. One possible explanation may be that DP1 is not amenable to growth of M. rhodesianum. This particular isolate was not available at the time of the original validation studies for the traditional test. Therefore, there is no data

ATP Bioluminescence-based Pallchek

455

precedent to this study. Additional studies will be required to confirm this initial observation. Rapid microbiological methods can be applied to a broad range of quality control operations, such as purified and process water testing, raw materials and excipients testing, environmental monitoring, in-process monitoring, manufacturing process design, investigations and final products release testing. The latter is arguably the most critical test because of the immediate impact that any failure or undetected contamination could have in the patient population. The data discussed throughout the chapter strongly suggest that a rapid sterility test using the Pallchek Rapid Microbiology System provides results in a timely manner that are equivalent to or better than those obtained in the traditional method. The use of an incubation (enrichment) phase during a period of time significantly shorter that the 14 days of the conventional method contribute to an easier interpretation and comparability of the results since both the rapid and the traditional assays rely on growth of the microorganisms under conditions similar to those described in the pharmacopoeia (USP <71>, 2011a; EP Section 2.1.6, 2010a; JP, 2006). In addition, the implementation of the rapid method described here potentially presents the following advantages: Reduced warehouse space and costs for raw materials Intermediates and final products fast final-product release shorter product release cycle times time and labor savings in the lab, during manufacturing Decreased plant downtime Reduced cycle times reduction of backorders reduction or elimination of product losses increased manufacturing capabilities Risk reduction in manufacturing Increased business and production flexibility

456

Rapid Sterility Testing Increased product development capabilities Robust understanding of manufacturing processes Proactive control: move from QC to QA procedures Immediate detection and correction of contamination Immediate cleaning validation Better protection of customers and company image.

In the decision making for the implementation of a rapid sterility test, a translation from the potential advantages delineated above to actual return of investment (ROI) dollars is essential. The ROI exercise should comprise calculations on the cost of the traditional test, the cost of implementing the rapid method, and a detailed analysis of the savings generated by the RMM (Yvon, 2008; Gadal and Yvon, 2009). The analysis should apply to a period of time from one to five years and should include comparative data between conventional and rapid methods, including: number of tests per year and price per test total testing time (hours) and labor cost (per hour) equipment (investment in new equipment, depreciation, calibration, qualification) lab space and environment test requirements disposal of used plates, reagents, etc. cleaning, preparation and downtime operation time time to results validation documents training maintenance contracts.

ATP Bioluminescence-based Pallchek

457

We have found that the relatively low cost and simplicity of the Pallchek luminometer, along with the time and cost savings from early contaminant detection, significantly facilitates the required calculations, showing ROI figures that can be realized in just 12 years, depending on test volumes, cost of warehousing and cost avoidances on additional investigations. Overall, the results described in this chapter demonstrate that the Pallchek Rapid Microbiology System provides a significant time saving for the detection of microorganisms within media that are visually occluded due to the cloudy suspension characteristics of the drug product studied in the work reported here. Based on the encouraging results that the Pallchek Microbiology System provided with this drug product, plans are being considered to running a comparability study of this method in parallel to the harmonized compendial sterility test method at a manufacturing site for a period of three to six months, using actual batch samples of manufactured product.

Acknowledgments
We would like to thank Michael Baumstein, Amber Dellar, Karen Boeve and John Shabushnig for their support and insightful information and advice during the development of this work. In addition, we are grateful to Michael Boquet for his excellent technical assistance during the early stage of these studies.

REFERENCES
Chappelle E.W., Levin, G.V. (1968) Use of the Firefly Bioluminescence Reaction for Rapid Detection of Counting Bacteria. Biochemical Medicine 2: 4152. Denoya, C., Reyes, J., Dawson, E., Sessoms, D., Baumstein, M., Shabushnig, J. (2010) A Rapid Microbiological Assay to Monitor the Effectiveness of a Vaccine Injector Sanitization Following a Microbial Challenge Procedure. American Pharmaceutical Review 13(4): 5461.

458

Rapid Sterility Testing

European Pharmacopoeia (EP) (2010a) Chapter 2.1.6 Sterility. 7th edition. Council of Europe, Strasbourg, France. European Pharmacopeia (EP) (2010b) Chapter 5.1.6 Alternative Methods for Control of Microbiological Quality. 7th edition. Council of Europe: Strasbourg, France. FDA (2004) Guidance for Industry. Sterile Drug Products Produced by Aseptic Processing Current Good Manufacturing Practice U.S. Department of Health and Human Services, Food and Drug Administration, Rockville, MD. Gadal, P., Yvon, P. (2009) Rapid Microbiology ROI: Calculating scientific benefits as return on investment dollars. Pharmaquality. www.pharmaquality.com Japanese Pharmacopoeia (JP) (2006) Chapter 4.06 Sterility Test. 15th edition, the Ministry of Health, Labor and Welfare, Japan. Kramer, M., Suklje-Debeljak, H., and Kmetec, V. (2008) Preservative Efficacy Screening of Pharmaceutical Formulations using ATP Bioluminescence. Drug Dev and Industrial Pharmacy, 34: 547557. Lehninger (2008) Principles of Biochemistry. Fifth edition Eds. David L. Nelson, Michael. M. Cox. WH Freeman Publishers. New York, NY. Nielsen, P. and Van Dellen E. (1989) Rapid Bacteriological screening of cosmetic raw materials by using bioluminescence. Journal of the Association of Analytic Chemists 72(5): 708711. PDA (2000) Technical Report No. 33. Evaluation, Validation and Implementation of New Microbiological Testing Methods. PDA Journal of Pharmaceutical Science and Technology. Supplement 54(3): 139. Parenteral Drug Association, Bethesda, MD. Stanley, P.E. (1989) A concise beginners guide to rapid microbiology using adenosine triphosphate (ATP) and luminescence. In ATP Luminescence: Rapid Methods in

ATP Bioluminescence-based Pallchek

459

Microbiology, P. E. Stanley, B. J. McCarthy, and R. Smither, Eds., Blackwell Scientific Publications, Oxford, England, 111. Pall Life Sciences. Testing Procedures and Applications for the Pallchek Rapid Microbiology System, USTR 2358, Pall Life Sciences, Port Washington, New York. Pall Life Sciences. Validation Guide for the Pallchek Rapid Microbiology System. USTR 2359. Pall Life Sciences, Port Washington, New York. USP (2011a) Chapter <71> Sterility Tests. USP 34-NF 29. The United States Pharmacopeial Convention/National Formulary, Rockville, MD. USP (2011b) Chapter <1223> Validation of Alternative Microbiological Methods. USP 34-NF 29. The United States Pharmacopeial Convention/National Formulary, Rockville, MD. White, E.W., McCapara, F., Field, G.F., McElroy W.D. (1961) The Structure and Synthesis of Firefly Luciferin. Journal of American Chemical Society 83: 24022403. Yvon, P. (2008) Rapid Methods: Return of Investments. Podium Presentation at Plenary Session 4 Rapid Methods. PDA 3rd Global Conference on Pharmaceutical Microbiology, Chicago, IL.

ABOUT THE AUTHORS

Claudio Denoya, Ph.D., is a Research Fellow and Group Leader of the Microbiological Technology Assessment group at Pfizer Global R&D. He is a co-chair of the Global RMM Steering Team. He is also an Adjunct Professor at the Department of Molecular and Cell Biology, Univ. of Connecticut. At Pfizer Dr. Denoya held several lead positions and led many molecular and microbiology projects. Dr. Denoya has received several recognitions, including the Pfizer Global R&D Outstanding

460

Rapid Sterility Testing

Team Achievement Award, Pfizer Global R&D Individual Achievement Award, two Pfizer Individual Performance Awards, two Supply Chain Recognition Awards, and the United States National Hispanic Corporate Achiever Award. He has received distinguished fellowships and visiting investigator positions from the Public Health Research Institute of the City of New York, the University of Sao Paulo, the Autnoma University of Madrid, and the PNUD-UNESCO. He has authored over 250 patents, book chapters, journal articles and technical presentations in the areas of biochemistry, microbiology, cell and molecular biology, and pharmaceutical sciences. Dr. Denoya holds a Ph.D. in Biochemistry and Molecular Genetics of Animal Viruses, a M.S. in Biochemistry and Microbiology, and a B.S. in Clinical Biochemistry from the University of Buenos Aires.

Jennifer Reyes is a Microbiologist who joined Pfizer in 2006 and has worked on the development of Rapid Microbiological Methods (RMMs) in the area of detection of microbial contaminants in pharmaceutical products. Prior to Pfizer, Jennifer worked as a Microbiologist for Amgen and before that for Monsanto in the area of Quality Control Assay Development. Jennifer received her B.S. and M.S. degrees in Microbiology from the University of Rhode Island.

Maitry Ganatra is Global Product Manager with Pall Life Sciences and is responsible for managing the process monitoring product portfolio and directing global cross functional team on new products. She has more than 10 years of business development experience with Life Sciences products. Prior to joining Pall, she led Microbiology Validation Program at Claris Life Sciences. She has authored many scientific publications and is committee member of PDA Task force for Revision of Technical Report No. 13 on Environmental Monitoring. Dr. Ganatra holds a Diploma in Pharmacy, Ph.D. in Microbiology (Gujarat University) and currently pursuing MBA at Long Island University.

ATP Bioluminescence-based Pallchek

461

Daniel A. Eshete M,D. PhD., holds a Doctor of Medicine (Addis Ababa University), M.Sc. in Biochemistry (Addis Ababa University/KI) and Ph.D. in Chemical Pathology (University of Cape Town) with Post-Doctoral studies in Microbiology/ Immunology at Karoliniska Institute, Microbiology and Tumour Biology Centre (Stockholm) and St. Louis University Division of Infectious Diseases (St Louis, MO). Prior to moving to the Biomedical Industry he worked in academia teaching Biochemistry, Clinical Chemistry and Instrumentation. His research works have been focused on cellular signal transduction, and the molecular basis of host parasite interaction and microbial infection. Since joining the industry his focus has mainly been on Pharmaceutical Quality Control with particular emphasis in rapid microbiological methods development and validation. Dr. Eshete is currently a staff scientist and a member of the Process Monitoring and Pharmaceutical QC team at Pall Life Sciences, Scientific & Laboratory Services.

Reprinted from Rapid Sterility Testing, edited by Jeanne Moldenhauer. Copyright 2011, co-published by PDA and DHI. All rights reserved.