Determination of ximelagatran, melagatran and two intermediary metabolites in plasma by mixed-mode solid phase extraction and LC-MS/MS.

An analytical method was developed for the determination of ximelagatran, an oral direct thrombin inhibitor, its active metabolite melagatran, and the two intermediate metabolites, OH-melagatran and ethyl-melagatran in human plasma. Extraction of plasma was carried out on a mixed mode bonded sorbent material (C8/SO(3)(-)). All four analytes, including their isotope-labelled internal standards, were eluted at high ionic strength with a mixture of 50% methanol and 50% buffer (0.25 M ammonium acetate and 0.05 M formic acid, pH 5.3) with an extraction recovery above 80%. The extracts were demonstrated to be clean in terms of a low concentration of albumin and lysoPC. The sample extraction was fully automated and performed in 96-well plates using a Tecan Genesis pipetting robot. Analysis of the extracts were performed with liquid chromatography followed by positive electrospray ionization mass spectrometry. The low organic content and the low pH of the extracts allowed for, after dilution 1:3 with buffer, direct injection onto the LC-column. The four analytes were separated on a C18 analytical LC-column using gradient elution with the acetonitrile concentration varying from 10 to 30% (v/v) and the ammonium acetate and acetic acid concentration kept constant at 10 and 5 mmol/L, respectively, at a flow rate of 0.75 mL/min. Linearity was achieved over the calibrated range 0.010-4.0 micromol/L with accuracy and relative standard deviation in the range 96.9-101.2% and 6.6-17.1%, respectively at LLOQ, and in the range 94.7-102.6% and 2.7-6.8%, respectively at concentrations above 3 x LLOQ. The method replaces a manual method, and displays the advantages of having a fully automated sample clean-up, no evaporation/reconstitution step, high recovery, and complete LC-separation of all four analytes.

Factorial design for the development of automated solid-phase extraction in the 96-well format for determination of tesaglitazar, in plasma, by liquid chromatography-mass spectrometry.

An analytical method was developed for the determination, in blood plasma, of a novel peroxisome proliferator-activated receptor (PPAR) agonist drug, tesaglitazar. The drug and the isotope labelled internal standard were isolated by solid-phase extraction (SPE) on hexylsilica, separated by reversed-phase liquid chromatography and quantified by tandem mass spectrometry. Factorial design and a robotic sample processor were employed in the exploration and optimisation of the SPE procedure in the 96-well format. This allowed rapid development of the method, notably limiting the process to four experiments before validation. The detectability was greatly improved by utilising the formation of sodium adducts in atmospheric pressure positive ionisation mass spectrometry. Absolute recovery was more than 95% with a coefficient of variation of 5% at a level of 8.7 nM. The accuracy and precision of the automated SPE method presented here matched the excellence of the previously used method based on manual liquid-liquid extraction. Furthermore, the method resulted in an increased sample throughput.

Determination of melagatran, a novel, direct thrombin inhibitor, in human plasma and urine by liquid chromatography-mass spectrometry.

Analytical methods for the determination of melagatran (H 319/68) in biological samples by liquid chromatography (LC)-positive electrospray ionization mass spectrometry using multiple reaction monitoring are described. Melagatran in plasma was isolated by solid-phase extraction on octylsilica, either in separate extraction tubes or in 96-well plates. Absolute recovery of melagatran from plasma was >92%. Melagatran and the internal standard, H 319/68 D2 13C2, were separated from other sample components by LC utilizing a C18 stationary phase and a mobile phase comprising 35% acetonitrile and 0.08% formic acid in 0.0013 mol/l ammonium acetate solution. After dilution, urine was injected directly onto the LC column and subjected to gradient LC. The relative standard deviation was 1-5% for concentrations above the limit of quantification, which was estimated for plasma at 10 or 25 nmol/l for sample volumes of 500 or 200 microl, respectively, and 100 nmol/l for urine.