AOAC CASP Meeting Book

810 Candidate methods are compared to a cultural reference method, where applicable. The 811 following are examples of sources of acceptable reference methods: AOAC OMA, U.S. Food 812 and Drug Administration Bacteriological Analytical Manual (BAM), U.S. Department of 813 Agriculture–Food Safety and Inspection Service Microbiology Laboratory Guidebook (MLG) (for 814 meat and poultry products), International Organization for Standardization (ISO) and Health 815 Canada Compendium of Analytical Methods, USP, Dairy Standard Methods. 816 817 Recognizing that there may be a lack of reference methods available for the cannabis matrix, 818 guidance in AOAC Standard Method Performance Requirements should be followed in 819 conjunction with reputable reference method sources. When appropriate, Method developers 820 should coordinate with the study director and/or consultant for best practices to be followed. 823 824 AOAC INTERNATIONAL recognizes claims for the range of specific cannabis matrices 825 successfully validated in the Method Developer Study, or the PCS and CS. The number of 826 different matrices required for testing depends on the applicability of the method. All 827 claimed matrices must be included in the Method Developer Study and the PCS. See section 7 828 for guidance on matrix categorization. 831 832 For the artificially contaminated food types, three inoculated levels (high, medium, and low) 833 and one uninoculated level are required. For naturally contaminated food, three 834 contamination levels (high, medium, and low) are required, and no uninoculated level. The 835 low level should be near the limit of detection, and the medium and high levels should cover 836 the analytical range of the candidate method. If the claimed range of the method is greater 837 than 4 logs, intermediate levels may be required at the discretion of the appropriate method 838 volunteer(s) in consultation with the Study Director. 839 If the method is intended to detect more than one target organism simultaneously from the 840 same test portion, the validation study should be designed so that target organisms are 841 inoculated into a common sample and the validation tests are performed in a simultaneous 842 manner. 845 846 For each level, analyze five test portions by the candidate method and five test portions by 847 the reference method (or confirmatory method). 850 851 Naturally contaminated matrix is preferred as a source of inoculum, if available. Inoculating 852 cultures are used only if the method is for a specific target analyte which may not routinely 853 821 822 5.1.3.2 Matrix Categories 829 830 5.1.3.3 Levels of Contamination 843 844 5.1.3.4 Number of Test Portions 848 849 5.1.3.5 Source of Contamination