Application of a Microflow LC/MS/MS Source for Quantitative Bioanalysis of Small Molecules Using Reduced Sample Volumes in Discovery DMPK

摘要

Introduction The suitability of a CaptiveSpray ionization source optimized for 0.5 - 5.0 uL/min flowrates and coupled to a Sciex API4000 MS was evaluated for quantitative bioanalysis in our laboratory. Advantages of working in this flow range include the ability to achieve concentration sensitivity similar to conventional LC/MS/MS with greatly reduced sample volumes and decreased solvent consumption. Mouse serial PK profiling was performed using microsampling and dried blood spot (DBS) techniques to facilitate small volume blood collection (<=15uL blood). Samples were then analyzed by both the CaptiveSpray source and a conventional TurboV source each coupled to an API4000 LC/MS/MS system. The comparison included assessment of chromatographic separation, bioanalytical figures of merit and PK results on each system. Method Timolol was dosed orally (50mg/kg) to C57/Blk6 mice. Blood was collected serially at 7 time points post dose via saphenous vein using a capillary tube coated with KEDTA. Fifteen uL aliquots of blood per time point were spotted onto filter paper cards (DBS). Analytical standards and QC were prepared in blank mouse blood and spotted (DBS). Timolol was extracted from 3mm punches of DBS (approx3uL blood) with 60uL of 60percent methanol in water. Samples were analyzed on both systems. The conventional system included a TurboV source and a Halo C18 50X2mm column at 440uL/min flowrate. The CaptiveSpray system used a Halo C18 50X0.2mm column at flowrate of 4uL/min. Injection volumes on the conventional and CaptiveSpray systems were 10uL and 0.25uL, respectively. Preliminary Data Initial results demonstrated that the CaptiveSpray microflow LC/MS/MS system is suitable for quantitative bioanalysis of small molecules using limited sample volumes in a discovery DMPK laboratory. Calibration curves (0.5-2500ng/mL) for timolol from DBS using buspirone as internal standard were fit using a 1/x~(2) weighting with correlation coefficients r>0.993 on both systems. A Signal-to-noise (S/N) ratio of 1.9 was obtained with the conventional system from a 10 uL injection volume of a 0.5ng/mL standard of extracted timolol. Since S/N was <3, the lower limit of quantification (LLOQ) was increased to 1ng/mL (S/N 4.5) on the conventional system. Using the CaptiveSpray source, a S/N of 4.2 was obtained with an injection volume of only 0.25uL of a 0.5ng/mL standard of extracted timolol and LLOQ was 0.5ng/mL. Peak widths at base were 0.09 min and 0.045 min on the conventional and CaptiveSpray systems, respectively. Both the conventional and CaptiveSpray systems demonstrated equivalent figures of analytical merit. Accuracy and precision of standards and QC samples at four concentrations (3, 15, 250 and 750ng/mL) were within 20 percent. Non-compartmental analysis was performed using Watson LIMS software and calculated PK parameters (mean n(velence)6 mice) were within 20percent. Mean calculated AUC(0-7hr) was 961 493ng~(*)hr/mL on the conventional system and 804 417ng~(*)hr/mL on the CaptiveSpray. Mean Cmax was 2480 1510ng/mL and 1994 1288ng/mL on the conventional and CaptiveSpray systems, repectively. Mean Tmax was 0.17 hours on both systems. Advantages of the CaptiveSpray over conventional LC/MS/MS from this study include similar concentration sensitivity with greatly reduced sample volumes (40-fold less sample) and reduced solvent consumption (99percent less). Studies are underway in our laboratory to assess additional benefits of the CaptiveSpray over the conventional system especially with respect to matrix effects.