Validation of methods for testing drugs of abuse

Abstract

Several methods of drug testing, for cases of seized drugs, have been developed over the years. All methods need to be validated. This study validates methods for the identification and quantification (percentage purity) of illegal drugs using Gas Chromatography-Mass Spectrometry (GC-MS). The methods were developed, calibrated and validated, according to standards set by the United Nations Office on Drugs and Crime (UNODC). In the first phase, methods were developed specifically for cocaine, diacetylmorphine (DAM) and 3,4-methylenedioxymetamphetamine (3,4 MDMA) and in the second phase, a method was developed for cannabis. Diphenylamine was used as an internal standard. Validation results in phase 1 (testing for cocaine, DAM, and 3,4-MDMA) showed that: 1) the method was linear from the limit of detection (LOD) to the highest calibration point, with R2 values >0.99 [0.9982, 0.9961, 0.997] 2) the LOD was 1% 3) the limit of quantification (LOQ) was 1% for cocaine and 3,4-MDMA, and 12.5% for DAM 4) the method was successfully validated for accuracy and precision [% CV: -5.361, 2.445, 1.346 and % E: 9.257, 14.448, 13.550] 5) samples of cocaine and 3,4-MDMA were stable at room temperature for 36 hours and for 6 hours for DAM 6) no carry-over was observed between samples. The validation results in phase 2 (testing for levels of cannabidiol, cannabinol and delta 9-tetrahydrocannabinol [Δ9-THC] in cannabis) showed that: 1) the method was linear from the LOD to highest calibration point, with R2 values >0.99 [0.999,0.999,0.999] 2) the LOD was 0.01% and the LOQ was 0.1% 3) the method was validated for accuracy and precision at 2%, 10%, and 20% concentrations, as all the results fell under the recommended value of 15 4) samples were stable at room temperature for 24 hours 5) no carry-over was observed between samples. All validated parameters were within the limits set by the UNODC. The study was carried out at the forensic laboratory BioDNA at Life Sciences Park. The systems used were a Varian 450 GC-220 MS in phase 1 and an Agilent 6890 GC-5973 MSD in phase 2. Results were generated through Varian Inc. MS Workstation1 and Hewlett-Packard ChemStation2 respectively.