Microbiological Method Validation: How Do We Prove that a Method is Fit for Purpose? Thomas Hammack Chief Microbial Methods Development Branch Division of Microbiology Office of Regulatory Science Center for Food Safety and Applied Nutrition U.S. Food and Drug Administration College Park, MD 20740
Equivalence? 24 h, 35ºC
Does
or
= 24 h, 35ºC and 42ºC
• Why validate methods? Benefits public health and world trade False negative results are unacceptable ISO 17025 lab accreditation demands the use of validated methods
Validation Programs • AOAC Microbiological Guidelines • ISO 16140:2003 • FDA’s Guidelines for the Validation of Analytical Methods for the Detection of Microbial Pathogens in Foods http://www.fda.gov/downloads/ScienceResearch/FieldScience/UCM273418.pdf
AOAC International •
1884—Association of Official and Analytical Chemists – –
•
Methods Validation for Fertilizers Membership Restricted to Regulatory Chemists
Supported by USDA –
–
Dr. Harvey Wiley -“Father of FDA” is also considered the “Father of AOAC” • 1892—Secretary of AOAC • 1898—Established AOAC’s Committee on Food Safety Housed within FDA
•
1939—Microbiological Sampling of Eggs and Egg products Official Method 939.14 still in use today
•
Many AOAC methods Official Methods were developed and validated in FDA Labs
•
1972—FDA published acceptance of AOAC Official Methods in the Federal Register –
•
Allows FDA to use proprietary rapid methods for the analysis of regulatory samples
1979—became an independent non-profit organization no longer tied to FDA, but many FDA employees serve as volunteers
International Organization for Standardization (ISO) •
International non-governmental organization – –
•
Started in 1926 as the International Federation of National Standardizing Associations (ISA) – – – –
•
Focused on mechanical engineering Disbanded in 1942 Reorganized as ISO in 1946 Mission statement is to provide “International Standards for Business, Government and Society”
1991 ISO and European Committee for Standardization (CEN) signed the Vienna Agreement – –
•
Based in Geneva, Switzerland Comprised of 163 national standards institutes including ANSI in the US
CEN is comprised of the 31 Member States of the European Union Vienna agreement stipulates that CEN and ISO Standards be identical to enable free trade within the EU
ISO Technical Committee 34 (TC 34) Subcommittee 9 (SC 9) – –
ISO exercises its role in the standardization of food microbiological testing methods through TC 34/SC 9 TC 34/SC 9 started standardizing “horizontal” reference methods for bacteria in foods in the 1970s
FDA’s Methods Validation Guidelines The Science and Research Steering Committee (SRSC), of the Office of Foods and Veterinary Medicine (OFVM), approved guidance to be used for validation of microbiological and chemical methods.
Guidelines for the Validation of Analytical Methods for the Detection of Microbial Pathogens in Foods http://www.fda.gov/downloads/ScienceResearch/FieldScience/UCM273418.pdf
Guidelines for the Validation of Chemical Methods for the FDA Foods Program http://www.fda.gov/downloads/ScienceResearch/FieldScience/UCM298730.pdf
Scope “These criteria apply to all FDA laboratories that develop and participate in the validation of analytical food methods for Agency-wide implementation in a regulatory capacity. This includes all research laboratories, and field labs where analytical methods may be developed or expanded for potential regulatory use. These documents will supersede all other intra-agency documents pertaining to foodrelated method validation criteria for microbial and chemical analytes. the SRSC will authorize the 18 formation of a Methods Validation Subcommittee (MVS) to serve as the governing body for all method validation concerns.”
FDA and Methods Validation Method validation is a process by which a laboratory confirms by examination, and provides objective evidence, that the particular requirements for specific uses of a method are fulfilled. It serves to demonstrate that the method can detect and identify an analyte or analytes: •
In one or more matrices to be analyzed
•
In one or more instruments or platforms
•
With a demonstrated sensitivity, specificity, accuracy, trueness, reproducibility, ruggedness and precision to ensure that results are meaningful and appropriate to make a decision.
•
Reliably for its intended purpose. Intended purpose categories include, but may not be limited to emergency/contingency operations; rapid screening and high throughput testing; and, confirmatory analyses.
•
After the method developer has conducted experiments to determine or verify a number of specific performance characteristics that serve to define and/or quantify method performance.
The Office of Foods and Veterinary Medicine & the SRSC “Roadmap for Microbiological Method Development and Validation”
RESEARCH
Organizational Partnerships •BAM Council •FERN MCC •IFSH •ORA “Micronauts”
VALIDATION
Micro Method Validation Sub-group
VERIFICATION
IMPLEMENTATION 9
The Method Validation Sub-Group (Microbiology) ESTABLISHES validation needs and priorities in consultation with the SRSC-Micro Super-group, FDA Bacteriological Analytical Manual Council (BAM Council), FERN Method Coordinating Committee, ORA “micronauts” inter-center working groups and others as appropriate
ADOPTS procedures to govern all administrative processes needed for emergency and nonemergency method validation proposals and studies.
PROVIDES planning, guidance, oversight, and resources to participating laboratories during the method development and validation process; will be the responsible authority for recommendations, evaluations and final approval of all validation studies from planning through field implementation.
CONSULTS with other governmental, and independent (commercial, and international) validation bodies to harmonize validation standards where possible and practices
10
The Method Validation Sub-Group (Microbiology) BROAD REPRESENTATION from CFSAN, ORA, CVM, and NCTR with additional expertise from biostatisticians and QA/QC managers
CURRENT MICRO MVS COMPOSITION: ORA Palmer Orlandi (co-Chair), Cathy Burns CFSAN Thomas Hammack (co-Chair), William Burkhardt, Darcy Hanes CVM Beilei Ge NCTR Steven Foley FERN Don Burr NCFST (CFSAN Moffett) Ravinder Reddy
11
METHOD VALIDATION SCOPE OF RESPONSIBILITY Pathogens, Genetic Material, Toxins and Antigens
Qualitative assays
Analyte
Bacteriological, e.g.
Salmonella spp.
Pathogenic Escherichia coli
Listeria monocytogenes
Shigella spp
Vibrio spp
Campylobacter spp
Applications
Pre- and selective enrichment
Microbial analyte recovery and concentration
Screening, high-throughput, confirmation
Procedures
Phenotypic, e.g.
Biochemical characterization
Antibiotic resistance traits Antigenic characterization Genetic, e.g.
Microbial toxins
Nucleic acid isolation/concentration
Viral pathogens, e.g.
Polymerase Chain Reaction
Hepatitis A virus
Conventional, Real-time
Norovirus
Reverse transcription
Enterovirus
Parasitic protozoan pathogens, e.g.
Cryptosporidium
Cyclospora cayetanensis
Bioengineered analytes, e.g.
Genetically-modified foods (GMOs)
Sequencing, e.g.
Whole genome
Selective sequencing
Single nucleotide polymorphism (SNP) analysis
Strain-typing applications
Method Validation is Required for… 1.
Submission of a new or original method, OR,
2.
Any significant modification of a method that may alter its performance specifications or changes to the fundamental science of an existing method. Categories include: •
Substitutions of reagents/apparatus
•
Expansion of the scope of an existing method to include additional analytes.
•
Changes in intended use i.e. screening or confirmatory.
•
Platform extensions or significant parameter changes e.g. adaptation to another real-time PCR thermal cycler.
•
Matrix extensions.
•
Changes to time/temperature incubation periods, or enrichment media.
•
In cases where the sample preparation and/or the extraction procedure/analytical method is modified from the existing test procedure and protocol, i.e the new method should demonstrate that the modifications do not adversely affect the precision and accuracy or bias of the data obtained.
•
Modification of a method’s performance range e.g. specificity, sensitivity beyond previously validated levels.
Levels of Validation Four levels of performance are defined. The hierarchy of scrutiny will provide general characteristics on the method’s utility and insights for its intended use, the assessed risk, and the food-borne illness potential for an analyte-matrix pairing.
Not all methods will or should be validated to meet the requirements of a Level 4: full collaborative study.
Method Validation Criteria Level One The lowest level of validation, with all the work done by one lab. Inclusivity and exclusivity testing has been tested, but by a limited number of strains. The analyte was tested at a level based on the intended use of the method, with just normal background flora. There is no aging of the artificially-inoculated samples and no comparison to an existing reference culture method. The expectation would be for the originating lab to continue to conduct further testing to eventually elevate the method to a higher level of validation.
Intended Use: Emergency needs. A method developed for the detection of an analyte, or a matrix not previously recognized or identified as a threat to food safety or public health. As the first level in the development of any method designed for regulatory use; performance of the method at this level of scrutiny will determine, in part, whether further validation is useful or warranted.
NOTE: Under emergency situations where the rapid development and deployment of a method is needed to immediately address an outbreak event, Level 1 criteria should be followed as closely as the situation will allow. Representatives of the MVC and Agency subject matter experts (SMEs) should be in close consultation with the originating laboratory. Once the crisis has past and it has been determined that there is a need for further validation, procedures outline in this document will be followed.
Method Validation Criteria Level Two This is a more robust study, with the possibility of regulatory strength depending on the circumstances. The originating lab has done a more comprehensive initial study, with inclusivity/exclusivity levels at the AOAC Collaborative Study level. If possible, a comparison has been done to an existing reference culture method. One other independent laboratory has participated in the collaborative study. Some of the criteria of the study are at a lower level than the full AOAC study, but still appropriate for the developing method at this stage.
Intended Use: Emergency needs. Slightly higher false-positive rates are acceptable as all samples analyzed with methods validated to this level will require confirmatory testing.
Method Validation Criteria Level Three This is a validation level that should have full regulatory strength. Most of the criteria followed by the originating lab are at the AOAC level, including inclusivity/exclusivity, analyte levels, competitor strains, aging, and comparison to existing method when available. The additional collaborating labs follow many of the criteria of an AOAC collaborative study. Intended Use: All methods validated to this level of scrutiny are acceptable for use in any and all circumstances e.g. confirmatory analyses; regulatory sampling, and compliance support.
Level Four This validation level has criteria equivalent to the AOAC Collaborative Study. Any method reaching this level of validation should be able to be submitted for adoption by the AOAC as a fully collaborated method.
Microbiology Validation Category Examples I.
General Qualitative Guidelines for Microbial Analytes
Table 1. ORIGINATING Laboratory Requirements †AOAC
Originating Laboratory Criteria
Level One: Urgent usage
Level Two: Independent lab validation
Level Three: Multiple lab collaborative
Level Four: Full collaborative study
# of target organism (inclusivity)
10 (20 for Salmonella)
50 (unless 50 aren't available)a,b
50 (unless 50 aren't available)a,b
50 (unless 50 aren't available)a,b
50a,b
# of non-target organism (exclusivity)
10 strains
30 strainsc
30 strainsc
30 strainsc
30 strainsc
# of foods
1 or mored
1 or mored
1 or mored
1 or mored
Up to 20 foodse
# of analyte levels/food matrixf
set level based on intended use and negative control
One inoculated levelf and uninoculated level
One inoculated levelf and uninoculated level
One inoculated levelf and uninoculated level
One inoculated levelf and uninoculated level
Replicates per food at each level tested
20
20
20
20
20
Aging of inoculated samples prior to testing
No
Yes
Yesg
Yesg
Yesg
Addition of competitor strain
Normal background flora
In 1 food at +1 log>analyte at fractional positivef analyte level
In 1 food at +1 log>analyte at fractional positivef analyte level
In 1 food at +1 log>analyte at fractional positivef analyte level
In 1 food at +1 log>analyte at fractional positivef analyte level
No
Yes, if available
Yes, if available
Yes, if available
Yes, if available
Comparison to recognized method
h
Collaborative Study
Microbiology Validation Category Examples, continued Table 2 - General Qualitative Guidelines for Microbial Analytes-Collaborating Laboratory Requirements
Collaborating Laboratory Criteria
Level Two: Independent lab validation
Level Three: Multiple lab collaborative
Level Four: Full collaborative study
AOAC Collaborative Study
# of laboratories providing usable data
2
5
10
10
# of foods
1 or morea
1 or morea
1 or morea
1 to 6 foodsb
# of strains of organism
1 per food
1 per food
1 per food
1 per food
# of analyte levels/food matrixc
2 levels: One inoculated levelc and uninoculated level
3 levels: One inoculated levelc one at 1 log higher and uninoculated level
3 levels: One inoculated levelc, one at 1 log higher and uninoculated level
3 levels: One inoculated levelc one at 1 log higher and uninoculated level
# of replicate samples/food
8 per analyte level
8 per analyte level
8 per analyte level
6 per analyte level †
Aging of inoculated samples prior to testing
Noe
Yesd
Yesd
Yesd
Comparison to Recognized Method
Yes, if available
Yes, if available
Yes, if available
Yes, if available
Microbiology Validation Category Examples, continued II. General Qualitative Guidelines for Food-borne Microbial Pathogens That Present Unique Isolation and/or Enrichment Challenges† Table 1. ORIGINATING Laboratory Requirements Originating Laboratory Criteria
Category One: Urgent usage
Category Two: Independent lab validation
Category Three: Multiple lab collaborativ e
Category Four: Full collaborativ e study
Replicates per strain
3
3
8
8
Comparison to recognized methoda
No
Yes, if available
Yes, if available
Yes, if available
Table 2. COLLABORATING Laboratory Requirements Collaborating Laboratory Criteria # of laboratories providing usable datab Replicates per strain Comparison to Recognized Methoda
Level One: Urgent usage
Level Two: Independent lab validation
Level Three: Multiple lab collaborative
Level Four: Full collaborative study
n/a
2
5
10
n/a
3
6
n/a
Yes, if available
8 Yes, if available
Yes, if available
Microbiology Validation Category Examples, continued III. CRITERIA AND GUIDANCE FOR THE VALIDATION OF FDADEVELOPED MOLECULAR-BASED ASSAYS •Inclusivity and exclusivity •Target gene(s) and controls (positive and negative). •Comparison to the Reference Method
IV. FOOD MATRIX EXTENSION FOR VALIDATED MICROBIOLOGY METHODS The validation of a method for a food matrix not previously included in a validation study is necessary to assure that the new matrix will produce neither high false positive (matrix is free from cross reactive substances) no high false negative rates (matrix is free of inhibitory substances)
•Guidance to Support Field Laboratory Expedience •Guidance for the Formal Expansion of Validated Food Matrices V. PLATFORM EXTENSION
Microbiology Validation Category Examples, continued VI.
CRITERIA AND GUIDANCE FOR THE VERIFICATION AND VALIDATION OF COMMERCIALLY- AVAILABLE MICROBIOLOGICAL DIAGNOSTIC KITS AND PLATFORMS Validation of an Alternative method: Demonstration that adequate confidence is provided when the results obtained by the alternative method i.e. the commercially-available kit, are comparable to or exceed those obtained using the reference method using the statistical criteria contained in the approved validation protocol. Verification: The confirmation by examination and the provision of objective evidence that specified requirements have been fulfilled. Criteria For Kits Fully Validated in a Collaborative Study Monitored by an Independent Accrediting Body 1.
For commercially-available microbiological diagnostic kits whose performance parameters have been fully validated in a collaborative study monitored and evaluated by an independent accrediting body e.g. AOAC-OMA, AFNOR, etc.
2.
For commercially-available microbiological diagnostic kits whose performance parameters are supported by data obtained through an abbreviated validation protocol and evaluated by an independent accrediting body e.g. AOACRI.
Current Micro MVS Priorities I. Hepatitis A Virus a. Real-time RT-qPCR assay to detect Hepatitis A virus b. In a food matrix (green onions) Status: Final report nearing completion
2.
†Non-O157:H7
STEC
Screening method to detect non-O157:H7 STECs using the BioPlex Status: Multi-lab collaborative study has been completed..
3.
‡Salmonella
Enteritidis
Isolation and Detection of Salmonella Enteritidis (SE) from Whole Shell Eggs-Cultural and Molecular Applications
4. Norovirus a. Real-time RT-qPCR assay to detect Noroviruses b. In a food matrix (molluscan shellfish) Status: planning phase underway in collaboration with the ORA-CFSAN virology working group
23
Validation of an Identification Method for Shigatoxigenic E. coli Somatic (O) Groups using the BioPlex Suspension Array System Method Authors Andrew Lin Teresa Lee Laurie Clotilde Julie A. Kase Insook Son J. Mark Carter Carol R Lauzon
24
Validation of a STEC Molecular Serotyping Method •
STECs are a significant public health concern –
Non-O157 STECs are responsible for over 60% of STEC infections or an estimated 112,000 illnesses in the U.S. each year
–
Over 74.2% of STEC infections in the U.S. are caused by serogroups O26 (23.9%), O45 (7.8%), O91 (2.3%), O103 (16.7%), O111 (12.6%), O121 (7.5%), and O145 (3.4%)
–
O26, O103, O111, O121, and O145 are known to cause HC and HUS, and O45 is associated with HC
–
Other serogroups that may cause HC and HUS, but are less commonly isolated, are O91, O113, and O128
25
Validation of a STEC Molecular Serotyping Method BioPlex Assay Overview • •
Must be performed on Pure Cultures Bead-based assay that can be multiplexed –
• • • •
• •
Conventional and real-time PCR assays for STEC O serogroup - limited by resolution of band sizes on a gel, or the # of fluorescent channels
96-well plates Targets = nucleic acid, proteins – Ab, antigens, cytokines Each bead = a separate assay 100 different color-coded beads (magnetic or polystyrene) – Unique color comes from different ratios of two distinct flourophores Two lasers, fluidics, optics, detectors Automated, High-Throughput, Fast, Expandable
26
Validation of a STEC Molecular Serotyping Method Method Status • In-House Validation Study Completed – Paper – LIB • 5 Lab Collaborative Study Successfully Completed • Approved by the BAM Council 27
Validation of a STEC Molecular Serotyping Method In-house Validation Results Inclusivity Panel
28
Validation of a STEC Molecular Serotyping method Exclusivity Panel
40
BioPlex Results •
Fluorescence signals are quantified for each micro bead
•
Signal to background ratios are calculated, where background is measured using all no template reactions
•
Samples were considered to be positive when signal to background ratio was greater than 5.0 30
Validation of a STEC Molecular Serotyping method BioPlex Results
31
Validation of a Method for Salmonella in Whole Shell Eggs Development and Validation of a Method for the Detection and Isolation of Salmonella Enteritidis (SE) from Whole Shell Eggs Guodong Zhang Eve Thau Eric W. Brown Thomas Hammack 32
Validation of a Method for Salmonella in Whole Shell Eggs Background • SE is the second most commonly isolated Salmonella serotype in the United States – 14.18% (1968-1998) • SE is most commonly associated with whole shell eggs and egg products • FDA’s egg rule (74 FR 33030) recommends various preventive control measures including sampling and sample analysis • FDA’s BAM reference culture method for the detection of SE in whole shell eggs takes 9 days to complete • There are no AOAC Official Methods of Analysis methods for the detection of SE in whole shell eggs 33
Validation of a Method for Salmonella in Whole Shell Eggs Clean surface Surface disinfection
20 eggs to a container Hand homogenize sample
Hold at room Temp (20-24oC) for 96 h
Clean surface Surface disinfection 20 eggs + 2L TSB 24 h at 35oC 1 ml to 10 ml TT 24 h at 35oC
0.1 ml to 10 ml RV 24 h at 42oC
Streak on XLD, BS, HE 24 h at 35oC
Streak on XLD, BS, HE 24 h at 35oC
TSI, LIA 24 h at 35oC
TSI, LIA 24 h at 35oC
Serological Test
Serological Test
25 ml eggs to 225 ml TSB + ferrous sulfate; 24 h at 35oC
1 ml to 10 ml TT 24 h at 35oC
0.1 ml to 10 ml RV 24 h at 42oC
Streak on XLD, BS, HE 24 h at 35oC
Streak on XLD, BS, HE 24 h at 42oC
TSI, LIA 24 h at 35oC
TSI, LIA 24 h at 35oC
Serological Test
Serological Test
CURRENT BAM METHOD
PROPOSED BAM METHOD
34
Validation of a Method for Salmonella in Whole Shell Eggs
35
Validation of a Method for Salmonella in Whole Shell Eggs
36
Validation of a Method for Salmonella in Whole Shell Eggs Method Status • In-House Validation Study of Culture Method Complete – Manuscript in preparation – Report to BAM Council in preparation
• In-House Validation Study of qPCR Methods in Progress – ABI SE qPCR Method Promising – Evaluation of an internal qPCR method in progress 37
Validation of a Method for Salmonella in Whole Shell Eggs Collaborative Studies Tentative Schedule •Culture Method Alone Fall 2012
•Culture Method Plus qPCR Method Spring 2013 38
