Supplemental Materials
Cocaine/Cocaine Metabolite Assay Method
The LC-MSMS method used was adapted from our previously published method for cocaine and BE (Lin et al., 2001). All reference material was purchased from Cerilliant (Round Rock, TX). In brief, d3-deuterated cocaine, BE, EME, and norcocaine were added to 1.0 mL aliquots of plasma as the internal standards. The pH of the plasma was made acidic (≈ 4.0) by the addition of acetate buffer, and the mixture was extracted using solid phase extraction (SPE). The eluant was evaporated and reconstituted with methanol/0.1% formic acid in water mixture (10:90) and analyzed by LC-APCI-MSMS. The mass spectrometer was operated in the selected reaction-monitoring mode. Quadrupole Q1 was set to pass only the MH+ ions, which were caused to undergo collision-induced dissociation in quadrupole Q2 to abundant product ions as follow: cocaine and cocaine-d3 at m/z 304 to 182, and 307 to 185, respectively; BE and BE-d3 at m/z 290 to 168 and 293 to 171, respectively; EME and EME-d3 at m/z 200 to 182 and 203 to 185, respectively; and norcocaine and norcocaine-d3 at m/z 290 to 168 and 293 to 171, respectively. The product ions were then monitored selectively by quadrupole Q3. The concentrations of cocaine, BE, EME and norcocaine were determined from the peak area ratios of the analyte divided by the peak area ratios of its internal standard and comparison of the ratio with the calibration curve (2.5 to 750 ng/mL) that was generated from the analysis of human plasma fortified with known concentrations of the analytes and their internal standards.
Validation of the Cocaine/ Cocaine Metabolite Assay
Figure 1a shows a chromatogram of plasma fortified only with internal standards; Figure 1b shows plasma fortified with analytes at the lower limit of quantitation (LLOQ) of 2.5 ng/mL and internal standards. There was minimal background noise at the retention time of the analytes (Figure 1a) and sufficient signal-to-noise at the LLOQ (Figure 1b). Norcocaine and benzyolecgonine had the same transitions but eluted with sufficient chromatographic separation.
The results of the method validation experiments are summarized in Table 1. Specificity was determined from comparison of blank plasma fortified with internal standard (6 different sources, N=3 per source) to aqueous samples fortified with internal standard and analytes at the LLOQ (N=3). Peak areas at the transition and retention time of the analytes were quantitated (often by forced integration) and the peak area ratios compared to the mean peak area ratios of the LLOQ samples. As percent of LLOQ peak area ratios, those for cocaine, EME and norcocaine were less than 5%; those for BE were 13% (Table 1), which is still below the 20% maximum considered optimal. Excellent recovery was achieved with the SPE method. When peak areas of extracted samples (N=5 per concentration) at 5, 50 and 650 ng/mL were compared to unextracted samples (N=5 per concentration) at the same concentrations, mean recovery for all three concentrations ranged from 80% for BE to 99% for EME. Similar recovery was achieved for the internal standard that was tested at their concentration of use, 25 ng/mL (Table 1). Precision and accuracy were determined at the LLOQ and the concentration of the three quality control sample (QCs) concentrations, 5.0, 50 and 650 ng/mL. For intra-run precision and accuracy, samples were run at an N=5; accuracy was within 12% of target and precision measured as % coefficient of variance (CV) was within 7.4% (Table 1). Inter-run precision and accuracy was determined from the mean results from five analytical bathes which included the intra-run batch (N=5) and four others at N=3. Accuracy was within 10% and precision within 9.5% (Table 1). To determine the stability of processed samples, low and high QCs that had been used as part of the precision and accuracy experiment were stored after their initial analysis either on the autosampler (5 days, N=5 per concentration) or at -20˚C (7 days, N=3 per concentration) and then reanalyzed along with freshly extracted calibrators and QCs. The mean result of all analytes was within 13% of target (Table 1), demonstrating that processed samples were stable for these periods of storage.
Figure 1. Chromatogram of a) blank plasma fortified with internal standards and b) blank plasma fortified with analytes at the LLOQ and internal standards.
Table 1. Operating characteristics for LC-APCI-MSMS method for determination of cocaine, BE, EME, and norcocaine.
Validation Analyte
Experiment Cocaine BE EME Norcocaine
Specificity (mean peak area ratio in 6 blank plasmas fortified with internal standard as %LLOQ)
2.59 13.0 4.99 3.37
Recovery (%, mean of 3 concentrations, deuterated internal standard at 25 ng/mL in parentheses)
91.1 (93.1) 80.4 (81.6) 99.0 (98.2) 85.6 (89.3)
Intra-run precision and accuracy (% Target ± %CV, N = 5)
2.5 ng/mL 97.6 ± 2.9 94.0 ± 6.0 92.0 ± 4.8 111.2 ± 4.3
5.0 ng/mL 89.2 ± 2.0 100.0 ± 7.4 90.6 ± 2.6 88.4 ± 3.6
50 ng/mL 89.2 ± 1.6 91.2 ± 2.9 90.2 ± 2.4 90.0 ± 3.1
650 ng/mL 93.3 ± 4.1 93.1 ± 4.6 93.3 ± 4.4 91.9 ± 3.0
Intra-run precision and accuracy (% Target ± %CV, N = mean of 5 runs with 3-5 replicates/run)
2.5 ng/mL 96.8 ± 6.6 96.8 ± 7.4 95.2 ± 8.4 105.6 ± 9.5
5.0 ng/mL 90.4 ± 2.4 95.6 ± 4.6 92.6 ± 1.7 91.6 ± 2.2
50 ng/mL 93.6 ± 3.2 97.4 ± 4.5 95.4± 4.2 94.2 ± 3.2
650 ng/mL 96.6 ± 3.8 96.5 ± 4.4 97.8 ± 3.0 95.5 ± 2.2
Processed sample stability, 5 days on autosampler (% Target ± %CV, N = 5)
5.0 ng/mL 88.6 ± 6.8 102.0 ± 11.4 93.6 ± 2.8 87.0 ± 14.9
650 ng/mL 98.8 ± 10.0 94.5 ± 2.4 97.1 ± 2.2 92.6 ± 3.2
Processed sample stability, 7 days at -20˚C (% Target ± %CV, N = 3)
5.0 ng/mL 94.1 ± 1.5 97.8 ± 17.0 97.5 ± 2.1 106.1 ± 2.5
650 ng/mL 107.3 ± 2.2 103.2 ± 4.9 105.2 ± 2.4 106.9 ± 4.0