AOAC SPDS ERP DECEMBER 15, 2017

6. Based on the supporting information, what are the cons/weaknesses of the method?

The method does not meet all of the SMPRs as discussed above.

The validation guidance around homogeneity is clear that grinding, while a reasonable option for homogenizing samples, may result in degradation and clumping of resin resulting in less accurate results and highly variable samples. This method employs grinding of a 5g sample, with 200mg used for extraction, but does not address how to ensure the 200mg sample is representative. Some alternative methods successfully achieve homogeneity by “wet-grinding” a sample in the extraction solvent. The use of liquid nitrogen for grinding could necessitate more rigorous safety requirements. Additionally, incorporating a major material change to the sample preparation in the supplemental info could be quite confusing to potential adopters. 5g is a large sample requirement for a crop of this value. One of the key challenges for the adoption of a method will be sample size. While it may be that 5g does produce a far more accurate and precise sample for analysis, it would be valuable to prove that this sample size is an improvement beyond the current status quo of 0.5-2g samples. The separation time is quite long and could represent a barrier to adoption. This is exacerbated by the use of a gradient method and a 7 minute equilibration. Shorter isocratic methods have been employed successfully for the separation of cannabinoids with similar columns and mobile phases.

The addition of vortexing every 5 minutes during the sonication adds a fair bit of labor without showing a statistically significant improvement.

7. Any general comments about the method?

This method comes close to achieving most of the SMPR specifications. The submission represents a published article with some additional supporting information. This format has made reading and review challenging.

Recommendation for the Method

V. Final Recommendation Do you recommend this method be adopted as a First Action and published in the Official Methods of Analysis of AOAC INTERNATIONAL? Please specify rationale.

The method does not meet SMPR specifications for all analytes. The method is quite close to achieving SMPR specifications in many cases, though no among- laboratory reproducibility data is provided. Additionally, the submission as presented is challenging to read due to being a previously published paper with nonessential material (i.e. oil method) remaining as well as supplemental info added, but not incorporated. As previously discussed the clarity challenges in my opinion would represent a barrier to adoption. I firmly believe with some minor optimization this method could achieve the SMPR specifications and better serve the stakeholder community. Additionally, with some minor editing the method could be clearer, more concise, and ideal for First Action. I hope there is an opportunity for this optimization and editing to take place as the method has many strengths that would serve the stakeholder community quite well.