AOAC CASP Cannabinoids ERP October Method Book

177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207

added to a diode array library. The four negative control test materials were extracted and analyzed by HPLC to confirm the absence of cannabinoids and any other co-eluting interference compounds. Peak identification in chocolate samples was tested using the 4 in-house reference materials. Three replicates prepared by the same technician on the same day were analyzed by HPLC. Cannabinoid peaks were confirmed by matching with the created diode array library, with a minimum acceptable match of 90%. Finally, the absence of co-eluting peaks in the same samples was confirmed by Agilent’s peak purity analysis, where peaks were considered free of co-eluting (b) Linearity – The linearity of the response of the HPLC systemwas examined. All analytical standards from Cerilliant Corp (1 mg/mL) were diluted with a gas-tight syringe and methanol as a diluent to create an 11-point calibration curve (ranging from 0.1% w/v to 0.1 ppm). From the resulting chromatograms, linear regression was performed for all analytes, including analysis of the residual plots for each analyte. Outliers were removed as necessary to improve linearity. Acceptance criteria for all calibration curves included a correlation coefficient(R 2 ) of ≥ 0.99, and a mean of the (c) Limits of Detection (LOD) and Limits of Quantification (LOQ) – Once the calibration curve for each analyte was finalized, the signal-to-noise (S/N) ratio of the lowest calibration point was determined. The LOQ was chosen as the lowest point on the calibration curve, since in all cases the S/N was greater than 10. The LOD was chosen as the LOQ÷3.3. To confirm the LOD and LOQ values, analytes were prepared at the LOD, and again at the LOQ. Replicate injections (n = 6) of each were analyzed to ensure that the LOD peaks were clearly visible, and that LOQ peak areas had a relative standard deviation of ≤ 2.20%. Standard deviation in peak area limits were calculated from the formula provided in the European Pharmacopoeia’s monograph 2.2.46 (1) and the US Pharmacopoeia’s monograph <621> (2). The formula was set for an upper limit of 95% (d) Accuracy - Accuracy will be assessed by spiking chocolate with analytical standards at two distinct concentration and measuring percent recovery. This approach is in line with ICH Q2(R1) guidelines (4), as well as AOAC Annex F of Appendix F (5). Five replicates of each spike will be assessed, satisfying the ICH Q2(R1) requirement of 10 determinations. The percent recovery will be calculated, along with the relative standard deviation. Acceptance criteria was set to a percent peaks if the peak purity value exceeds the threshold value for each peak. relative residuals of ≤± 5%. confidence, in accordance with Eudralex guideline 3AQ11a for drug products (3).

recovery of 90 – 110% for concentrations up to 1% w/w.

Page 7 of 19