3.2 Method validation
This method performed stable baseline in analytical channel (m/z 586.3→513.2) and IS channel (595.4→514.2). The selectivity was acceptable since there were no interference peak in these channels. The total run time was 2.5 min, which was shorter than most methods to analyze TGC in bio-samples [10, 11]. Respectively, the retention time of TGC in serum samples and in CSF samples was 1.464±0.005 min and 1.466±0.005 min with five continuously reduplicate injections. The chromatograms obtained from blank serum and CSF, blank bio-samples spiked with TGC and IS, and the samples from patients were described in Figure 1. The serum and CSF samples spiked with TGC at lower limit of quantification (LLOQ, 5 ng/mL) also depicted unique and prominent peak at retention time.
Calibration curves were Y=0.019X+0.182 for serum samples (5-2000 ng/mL) and Y=0.0322X+0.53 for CSF (5-2000ng/mL) (a weighting of 1/x was used). Linear coefficients (r2) were 0.9999 and 0.9997 respectively, which could absolutely achieve accurate samples analysis.
Precision and accuracy data of TGC analysis in serum and CSF samples were tabulated in Table 2. At three QC concentrations of TGC in serum, the accuracy bias was ranged from -4.69% to 3.00%, with the intra-day and inter-day RSD ranged from 2.85% to 7.86%. For the CSF samples at concentration of 10.0 ng/mL, 400 ng/mL and 800 ng/mL, the accuracy expressed in terms of percentage bias was within -1.85% to 6.40%, and the precision varied from 3.06% to 13.72%. The data of matrix effect was shown in Table 3. The IS normalized matrix factors (MFs) for TGC ranged from 79.04% to 109.64% at all levels, and the RSD ranged from 4.93% to 9.43%. All of the data met the requirement of bio-samples quantification.
The results utilized to evaluate the stability of the method were listed in Table 4. There was no significant change of analyte in varied storage conditions within the prescribed time. The accuracy bias of TGC in bio-samples at low, medium and high concentration was ranged from -9.30% to 3.92%, and the RSD was ranged from 2.18% to 8.33%. Simultaneously, the variations of CSF and serum samples spiked with standard solution in auto-sampler for 90 min were acceptable. And the stability of TGC also not affected by three cycles of freezing and thawing significantly.