A new approach to the effective preparation of plasma samples for rapid drug quantitation using on-line solid phase...

A procedure that permits rapid development of an optimized solid phase extraction (SPE) method for the analysis of drugs in plasma by on-line solid phase extraction-mass spectrometry (SPE-MS) has been developed. This procedure employs the concept of manipulating the pH and the percentage of organic solvent in the chromatographic mobile phase to affect the retention behaviors of both the matrix components and the analytes of interest. This resulted in the effective removal of matrix interferences from biological samples during SPE. During a the method development, only generic HPLC gradient approaches were needed, and multiple samples were pooled so that several SPE methods could be investigated at once. The analysis time per sample was 1.3 minutes. Thus, the time involved in the entire method development (analysis of a set of samples) was less than one hour. With the knowledge of the retention behaviors of the analytes with respect to the pH and the percentage of organic, it was then possible to compose an optimized SPE-MS method. This method consisted of a base/organic and then an acid/organic washing step, followed by a rapid gradient elution step. Due to the rigorous washing procedure, most matrix interferences were removed, and analytes eluted off the SPE sorbent suffered from very little matrix interference. Thus, quantitation of drugs in plasma by a single quadrupole mass spectrometer could be accomplished, something that was not possible when only a generic gradient was used for on-line SPE-MS. In addition, both external and internal calibration curves could be obtained for the concentration range from 5 to 500 ng/mL with correlation coefficients of 0.99 (using 1/x as a weighting factor) and relative standard deviations (RSDs) less than 10%. The results achieved were comparable to those obtained by the use of a triple quadrupole mass spectrometer. Moreover, the robustness of the method was tested by continuously injecting plasma samples. During 136 runs, the absolute peak area variation for these three basic drugs was less than 15% without taking the signal variation from the mass spectrometer into account. Significantly, the on-line developed method can be directly transferred to a 96-well format SPE plate.