A Rapid Method for Bio-Burden Testing in Live Attenuated Influenza Vaccine Intermediates From Concept to Implementation

Size: px

Start display at page:

Download "A Rapid Method for Bio-Burden Testing in Live Attenuated Influenza Vaccine Intermediates From Concept to Implementation"

Stanley McKinney
5 years ago
Views:

1 A Rapid Method for Bio-Burden Testing in Live Attenuated Influenza Vaccine Intermediates From Concept to Implementation Praful K. Bhusari Vaccine Analytical Sciences Medimmune, Inc.

2 Introduction Mission statement: MedImmune is committed to advancing science to develop better medicines that help people live healthier, longer and more satisfying lives. Primary focus is on respiratory viral diseases Currently two commercial products Synagis a monoclonal antibody used for the prevention of respiratory syncytial virus infection in neonates and young infants, and FluMist - a Live Attenuated Influenza Viral Vaccine Vaccine Analytical Sciences Department develops and validates assays says for pipeline & approved products 2

Live, cold-adapted, temperature-sensitive,

3 Live, cold-adapted, temperature-sensitive, attenuated influenza virus vaccine Trivalent (A/H1N1, A/H3N2, B) 10 7 FFU of each strain per dose Dose: 0.2 ml intranasal spray (0.1 ml per nostril) Storage: 2-8ºC 2 C (refrigerator) Contains no preservatives (e.g., no thimerosal) 3

4 Structure of Influenza Virus (Type A and B) Viral Genome 8 Negative-Sense RNA Segments Viral Non-Envelope Proteins: PB1, PB2, PA Polymerase Complex Viral Envelope Proteins: Hemagglutinin (HA) Neuraminidase (NA) M2 NP Matrix (M1) Non-Structural Proteins NS1, NEP 4

5 Derivation of New Master Virus Seed Master Donor Virus New Wild-Type Strain Co-infect Cells Six genes from MDV confer ca, ts, att phenotype Hemagglutinin and Neuraminidase Genes from Wild Type 6:2 Master Virus Seed (ca, ts, att) 5

6 The need for a rapid Microbiology method The live, attenuated influenza virus vaccine material is produced d by viral expansion in, specific pathogen-free (SPF) chicken eggs. Following incubation and propagation of the virus in the SPF eggs, the allantoic fluids are harvested, which have natural potential for harboring microbial contamination. To minimize the chances of contamination, viral allantoic fluid is harvested in smaller sub-lots One-liter sub-lots of allantoic fluids are collected and sequestered until bio- burden test results are obtained, at which time the sub-lots with satisfactory bio-burden burden results are pooled and further processed. The sub-lot quarantine and delay in bio-burden burden results prevents a continuous process, thus warranting the use of a rapid bio-burden burden detection method. 6

7 Why AATI s Micro PRO system? Desirable assay attributes Micro PROŖ System Rapid time to results Results in 3-5 minutes per sample Auto-sampler with 42 sample capacity High throughput Walk-away automation Results in familiar Counts/ml Quantitative results format Able to detect and enumerate wide Bacteria, Yeast and Molds range of microorganism Cell size ³ 0.2µm Ease of operation maintenance Simple operation and maintenance Software validation status CFR 21 Part 11 compliant Per test cost $ 2-5 per test Post-sale support Very responsive 7

Micro PRO detection system Laser-based excitation at 635 nm Fluorochromes bound to cells provide information on cell state Live/dead Spore/vegetative Light scattering provides relative size

8 Micro PRO detection system Laser-based excitation at 635 nm Fluorochromes bound to cells provide information on cell state Live/dead Spore/vegetative Light scattering provides relative size information Provides quantification of microorganisms per volume, yielding count/ml results System composed of fluidic, optic, and electronic components Can detect and enumerate bacteria, yeast and mold 8

9 Fluidic System Sample delivery Sheath delivery Sheath flow Labeled microbes Core flow Quantitative cell delivery Hydrodynamic Focusing Single File Passage through detection region

10 Optic System Labeled microorganism Scatter Detector Fluorescence Detector Scatter signal Fluorescence signal Laser Beam shaped and focused; 635 nm laser excitation High performance optical filters

11 Electronic System Signal processing component Triggers on fluorescence Fluorescent event above the threshold is processed, along with the corresponding scatter event, and is plotted and recorded as a count Detector output = 1 count Time = 0 count Fluorescence Threshold Level

12 TVO Labeling Protocol (Only Viable Cells are Stained) Micro PRO Viable Cell Count / ml

13 Example of Micro PRO intensity plot Cell size Amount of label

14 Example of Results page 14

15 bio-burden assay development and challenges Based on the results of initial feasibility study performed by AATI in July 07 instrument was brought in-house in late September 07 Enumeration of typical environmental isolates grown in normal allantoic lantoic fluid was demonstrated to be feasible Background interference caused by components of virus allantoic fluid posed a major challenge for assay development Virus allantoic fluid is a very complex matrix that contains chicken cken cells and cellular debris in addition to virus particles The cellular debris can non-specifically bind the nucleic acid dye and cause false positive fluorescent events and thus high background noise Assay development efforts were mainly focused on reducing background noise 15

16 Spiking Feasibility Studies at AATI Unspiked a = 6.4 x 10E+05 b = 5.9 x 10E+04 c = 5.1 x 10E+01 Unspiked a = 6.4 x 10E+05 b = 5.9 x 10E+04 c = 5.1 x 10E+01 16

Enumeration of egg grown environmental isolates Ralstonia pickettii Dilution MicroPRO Count MicroPRO CFU / ml Plate 10-3 410293 10-4 42058 4.

17 Enumeration of egg grown environmental isolates Ralstonia pickettii Dilution MicroPRO Count MicroPRO CFU / ml Plate x x Staphylococcus epidermidis Dilution MicroPRO Count MicroPRO CFU / ml Plate x x

18 Reduction of background noise 18

19 Assay development Background noise was the major challenge to overcome Complex sample matrix leading to non-specific binding of nucleic acid dye Assay development required significant trial and error for screening various means to reduce or eliminate the background Experiments were designed to assess the effect of various sample treatments on background and on spiked bacteria Several sample pre-treatments were able to reduce the background noise, but not to the extent that they could be used alone A combination of several of these treatments was required to lower the background to an acceptable level Further improvements are required to consistently reduce the background in all virus harvest types 19

Reduction of background noise Filter clarification Using 20-40 micron commercially available filters

permeablize membrane fragments and to make larger micelles Moderate heat Incubate at 37 to 50 C

20 Reduction of background noise Filter clarification Using micron commercially available filters / strainers to remove large cellular debris Use of detergents in very low concentrations to permeablize membrane fragments and to make larger micelles Moderate heat Incubate at 37 to 50 C Refinements to the area definition Square vs. Oval gate Counting algorithm Literal vs. Centre Association Algorithm

21 Center association algorithm Center association works under the assumption that the main concentrations of events of interest are located in the center of the area If there are a large number of events in the fringe areas, some may be associated with the center area, but many will not This algorithm will not count events in the fringe area that exceed the counts in the center The area in the center of an area, one half the size of the main area, is calculated. The events counted in the fringe that exceed the count in the center area will be ignored 21

22 Method prototype 300µl sample 20µm strainer to remove large debris Test 2.7 ml phosphate buffer containing detergent, pre-warmed at 37 C 22

23 Method optimization and comparability testing Micro PRO system was installed in the manufacturing QC lab to continue method development and optimization Experiments are being performed towards the confirmation of the proof of concept and for assessment of the critical validation parameters After the method development is completed, Micro PRO method will be performed in parallel with the current bioburden method during the 2008 production campaign Evaluation of the comparability data would allow us to assess the need for a new acceptance criteria. As of now representative set of rejected (high bioburden) ) and passed samples (Low or no bioburden) ) by the current method were tested in parallel using the Micro PRO method

Example of Results Bioburden reject samples MicroPRO Count per ml Current method TSA plate cfu/ml Sub-lot - 24 1957110

39 33100 Fail 52000 Sub-lot - 137 91900 Fail 500000 Bioburden free accepted samples MicroPRO Count per ml Current method

24 Example of Results Bioburden reject samples MicroPRO Count per ml Current method TSA plate cfu/ml Sub-lot Fail Sub-lot Fail Sub-lot Fail Sub-lot Fail Sub-lot Fail Sub-lot Fail Bioburden free accepted samples MicroPRO Count per ml Current method TSA plate cfu/ml Sub-lot Pass <100 Sub-lot Pass <100 Sub-lot Pass <100 Sub-lot Pass <100 24

25 Screening of contaminated sub-lots Screening of high bioburden containing sub-lo using the Micro PRO method Sub-lots

26 Key findings from the screening study Micro PRO Micro PRO was able to accurately detect and provide relative contamination levels in virus harvest samples Based the data the limit of detection appears to be around 10 3 CFU/mL A screening criteria or cut-off between 5000 to 9000 CFU/ml is feasible Micro PRO system has demonstrated excellent correlation to plate counts in pure culture studies Treatment with detergents to reduce background noise may yield lower recoveries of stressed organisms and thus impair correlation on with plate counts.

27 Conclusion Micro PRO system was acquired and evaluated Demonstrated feasibility for enumeration of allantoic fluid grown typical environmental organisms grown in eggs Screened methods for background noise reduction Successfully demonstrated background noise reduction to an acceptable level Developed a prototype assay Demonstrated the feasibility of Micro PRO system as a screening method to identify contaminated sub-lots of viral harvest bulk 27

28 Next steps Finalize the method development Plackett-Burman DOE study to optimize method Generate data to demonstrate validation parameters such as accuracy, precision, linearity, LOD / LOQ Validation IOQ is nearing completion SOPs and PQ document are in progress Implementation 28

29 Acknowledgements Medimmune -CA Mark Galinski Dave Tabor Alfred Pan Kuldip Sra Lance Wong Thor Grespan Medimmune -UK James Wong Dennis Wong Steve Griffiths Kay Gildart Mark Carruthers John Schober AATI Ann Steger Dylan Baker Bret Brooks

Evaluation of the feasibility of the VACUETTE Urine CCM tube for microbial testing of urine samples

Evaluation of the feasibility of the VACUETTE Urine CCM tube for microbial testing of urine samples Background The VACUETTE Urine CCM tube is for the collection, transport and storage of urine samples