| Aim a rapid and sensitive method utilizing dried blood spots (DBS) to determine valproic acid in human whole blood by ultra-performance liquid chromatography (UPLC) coupled with tandem mass spectrometry(MS/MS)with electrospray ionization source in negative ion modes was established and validated.Method30μL blood containing valproic acid(VPA) was spotted on the filter paper and was dried at room temperature at least3h, then a6mm diameter of disk was punched out from the center of dried blood spots into2mL polypropylene centrifuge tube, followed by addition of150μL contained with internal standard (IS). After10min in ultrasonic bath, sample was performed by protein precipitation with350μl acetonitrile. The mixture was vortex-mixed for5min and followed by centrifugation at15000rpm for10min, the supernatant was dilute60times before it was injected into the system, and a2-ul aliquot of the diluted supernatant was injected into the UPLC-MS/MS system. Chromatographic separation was performed on a ACQUITY UPLC BEH C18column (1.7μm,2.1×100mm), with an isocratic mobile phase consisting of acetonitrile-5mM ammonium formate buffer (PH=7.8)(80:20, v/v) at flow rate of0.5mL/min. The column temperature was maintained at50℃. Mass spectrometric detection was performed on an triple quadrupole tandem spectrometer API5500instrument equipped in the negative mode, the ion transitions were monitored on m/z143.0'143.0for VPA and120.9-+76.8for benzoic acid under multiple reaction monitoring (MRM) mode. Method validation was conducted in terms of linearity, specificity, sensitivity, accuracy, precision, matrix effect, extraction recovery, stability, chromatographic effect of filter paper, effect of volume and hematocrit (HCT) of whole blood. Established method (DBS-UPLC-MS/MS) was applied to detect50clinical specimens from50patients, which concentration data was compared with those obtained from Plasma-ultra performance liquid chromatography coupled with tandem mass spectrometry(Plasma-UPLC-MS/MS) and Fluoresence Polarization Immunoassay (FPIA), statistical software was used for analyzing and plotting. Results with the chromatographic and mass spectrometric condition described above, valproic acid and IS exhibited good chromatography with baseline resolution at retention time of0.57min and0.48min. Calibration range was5-160μg/mL for valproic acid with correlation coefficient(r) of0.9978. Intra-assay, inter-assay imprecision and biases were all less than15%for DBS. The recovery of valproic acid was82.0-88.4%and matrix effect was not detected, valproic acid on DBS was stable at least7days at room temperature, at least10days at4℃, at least24h at50℃, and at least30days at-20℃. Chromatographic effect of filter paper was not detected. Volume of blood (30μL,50μL) and hematocrit of blood (25%、35%.45%、55%) did not show significant influence on detection of valproic acid concentration by DBS-UPLC-MS/MS. Valproic acid concentrations measured by DBS-UPLC-MS/MS, Plasma-UPLC-MS/MS and FPIA were compared. Passing-Bablok regression analysis showed that DBS-UPLC-MS/MS has good identity with Plasma-UPLC-MS/MS, while identity between DBS-UPLC-MS/MS and FPIA or between Plasma-UPLC-MS/MS and FPIA was poor. Bland-Altman plot showed that valproic acid concentrations detected by DBS-UPLC-MS/MS or Plasma-UPLC-MS/MS were lower than FPIA, while DBS-UPLC-MS/MS were similar to Plasma-UPLC-MS/MS. Conclusions A sensitive and specific UPLC-MS/MS method for quantitative analysis of valproic acid from DBS samples was successfully developed, with good performance in strictly validation including chromatographic effect of filter paper, effects of volume and HCT of whole blood applied on filter paper, it would facilitate clinical therapeutic drug monitoring of valproic acid. For the difference exists between established method and FPIA, which has been applying generally in clinics, application of DBS-LC-MS/MS assay for determination of valproic acid still needs further clinical validation. |
PDF Full Text Request