A monolithic-phase based on-line extraction approach for determination of pharmaceutical components in human plasma by HPLC-MS/MS and a comparison with liquid-liquid extraction.

An automated procedure using monolithic-phase based on-line extraction is described for pharmaceutical component analysis in plasma by LC-MS/MS. In this approach, a short monolithic C(18) 4.6 mm x 10 mm cartridge is used for high flow extraction at 4 mL/min. Plasma samples were subjected to protein precipitation first with acetonitrile, and the supernatant was diluted and loaded onto a monolithic cartridge. Sample elution was accomplished with narrow-bore LC-MS/MS system. A method for determination of Amprenavir (APV) and Atazanavir (AZV) in human plasma was developed with this approach. After 0.1 mL of plasma was transferred into each well of a 96-well plate by a liquid handler, the rest of sample preparation time typically only takes about 20 min. A Phenomenex Luna C18(2) 2.0 mm x 150 mm analytical column was used for the separation at a flow rate of 0.3 mL/min. The run time for each sample was 4 min. The standard curve range was 2.77-1520 ng/mL for Atazanavir, and 4.50-2560 ng/mL for Amprenavir. The accuracy (%bias) at the lower limit of quantitation (LLOQ) for Atazanavir was 2.7% and the precision (%CV) at the LLOQ was 7.9%, while the accuracy at LLOQ for Amprenavir was -1.3% and the precision at LLOQ was 7.8%. The inter-day %bias and %CV of the quality control samples of Atazanavir were < or = 4.5% and < or = 6.5%, respectively. The inter-day %bias and %CV of the quality control samples of Amprenavir were < or = 1.1% and < or = 7.2%, respectively. Coefficients of determination, a measure of linearity, ranged from 0.993 to 0.999. Very low carry-over (0.006%) even after high standard sample was demonstrated in the monolithic-phase based method. Other characteristics of such method include high recovery and good tolerance to matrix effect, which was demonstrated by 12 lots of plasma. The back pressure of the monolithic extraction cartridge remained the same after 450 samples injected. The performance of the monolithic-phased on-line extraction method was compared with that done by an automated 96-well liquid-liquid extraction procedure, which was carried out using hexane:ethyl acetate as the extraction solvent. The results showed that similar precision and accuracy were achieved by both methods.