Characterization of factors affecting the stability of frozen heparinized plasma

Intact free prostate-specific antigen and free and total human glandular kallikrein 2. Elimination of assay interference by enzymatic digestion of antibodies to F(ab')2 fragments

Various blood constituents can interfere with immunoassays, usually by binding the Fc portion of antibodies. Our previously developed assays for intact free prostate-specific antigen (PSA), free human kallikrein 2 (hK2), and total hK2 frequently yielded falsely high results despite including an excess of scavenger antibodies. We investigated whether this interference could be eliminated by replacing monoclonal capture or tracer antibodies with F(ab')2 or recombinant Fab fragments. Female heparin plasma samples (n = 1092), which should have negligible PSA and hK2, and male samples (n = 957) were analyzed to identify samples manifesting interference, which then were used to optimize protocols for the immunoassays. We compared original assays (monoclonal antibodies) versus optimized assays (F(ab')2 fragments: denatured mouse IgG added as scavenger) using another set of EDTA plasma (n = 113), heparin plasma (n = 160), and serum samples (n = 171). With the original assays, the frequency of falsely elevated hK2 and intact free PSA was 15 and 13%, respectively. The optimized assays eliminated 70-85% of these falsely elevated results and importantly reduced the magnitude in the remainder. F(ab')2 fragmentation was the most important factor in reducing interference. The optimized intact free PSA, free hK2, and total hK2 assays manifested high accuracy close to the lower limit of detection.

Higher throughput bioanalysis by automation of a protein precipitation assay using a 96-well format with detection by LC-MS/MS

Generic methodology for the automated preparation and analysis of drug levels in plasma samples within a drug discovery environment was achieved through the redesign of a protein precipitation assay to a microtiter (96-well) plate format and the application of robotic liquid handling for performance of all transfer and pipetting steps. Validation studies revealed that the application of robotics to sample preparation, in general, maintained the analytical accuracy and precision compared with preparing samples manually. The use of rapid gradient LC-MS/MS for analysis coupled with flow diversion of the solvent front allowed the introduction of protein-precipitated samples into the mass spectrometer without the necessity for source cleaning. The problem inherent in automatically pipetting plasma, caused by fibrinogen clots, was overcome by storing samples at -80 degrees C and thus precluding clot formation. The resulting methodology allowed sample preparation for a 96-well plate designed to accommodate 54 unknowns, duplicate 12-point calibration curves, and 6 sets of quality controls at three levels in approximately 2 h. This approach allowed an increase in throughput of sample preparation and analysis to >400 samples per day per LC-MS/MS instrument with minimal manual intervention. Overall, substantial time savings were realized, demonstrating that automation is an increasingly essential tool in a drug discovery bioanalytical environment.

Influence of time and temperature on coagulation analytes in stored plasma

There is no comprehensive study on the stability of coagulation analytes in plasma. We therefore determined the influence of storage time and temperature on prothrombin time, activated partial thromboplastin time, thrombin time, fibrinogen, factors V and VIII, antithrombin III, protein C and S in plasma from 20 healthy subjects and 20 patients receiving heparin therapy. The stability in plasma, defined as the period during which there was a change of less than 10% from the initial value, was 8 hours for activated partial thromboplastin time, 24 hours for prothrombin time, 48 hours for factor V and 7 days for thrombin time, fibrinogen, protein C and antithrombin III in healthy subjects at 6 degrees C. Factor VIII and protein S showed 19 and 12 % reduction in activity, respectively, after 8 hours. In volunteers not treated with heparin therapy, activated partial thromboplastin time was stable for 8 hours; prothrombin time for 48 hours; and thrombin time, fibrinogen and antithrombin III for 7 days with sample storage at room temperature. Factor VIII showed a decrease of 18 % after 8 hours. For patients receiving heparin therapy, the stability of the analytes in plasma stored at 6 degrees C was 8 hours for thrombin time, 24 hours for prothrombin time and activated partial thromboplastin time and 7 days for fibrinogen and antithrombin III. Factors V and VIII showed a decrease of 13 % and 20 % respectively after 8 hours. When the plasma of these patients was stored at room temperature, factor V was stable for 8 hours, and prothrombin time for 24 hours, whereas fibrinogen and antithrombin III remained unchanged for 7 days. Activated partial thromboplastin time showed an increase of 13 %, thrombin time a fall of 16 %, and factor VIII a decrease of 18 % after 8 hours.