Quantitative Determination of Estrone and Estradiol at Low pg/mL in Human Plasma by LC-MS-MS
Authors
Christopher J.L. Buggé, Anders Ljungqvist, Michael P. Sullivan
Introduction
Estrone and estradiol are found in normal human plasma as endogenous components, and are critical for sexual functioning and supporting bone growth. A sensitive assay was required to determine estrone and estradiol in human plasma at low pg/mL concentrations to support clinical trials of hormone replacement therapy.
Experimental
Chemicals
Reference estrone and estradiol and their deuterated internal standards were obtained commercially. All other chemicals were AR grade and all solvents were HPLC grade.
Standards/Solutions
Stock solutions of estrone and estradiol were prepared in acetonitrile/water 1:1, as were combined intermediate, calibration spiking and internal standard solutions. Calibration spiking solutions were prepared at 5, 10, 20, 40, 80, 160, 225, and 250 pg/mL. Quality Control (QC) solutions were prepared in human K3EDTA plasma at 15, 100 and 200 pg/mL for both compounds and stored in 1-mL aliquots at –70° C. QC samples were prepared from different weighings of drugs than the calibration spiking standards.
LC-MS-MS
The extract was injected directly onto a Sciex API 4000 LC-MS-MS in positive TurboIonspray mode using a C-18 column ( 4.0 x 50 mm) and a mobile phase of ACN/water/formic acid. Estrone and estradiol eluted at 2.0 and 1.7 minutes, respectively at a flow rate of 0.85 mL/mm, and were monitored in MRM mode with their deuterated internal standards.
Results
The linear quantitative range was established and validated from 5 to 250 pg/ml for both estrone and estradiol over 3 days. Calibration curves and quality control samples were prepared from human plasma that had been cleansed free of endogenous analytes. In this way, a matrix most similar to the authentic study plasma was used, and more accurate quantitative data can be obtained (compared to subtraction of endogenous levels of analytes from calibration and QC samples). Precision and accuratcy were determined to be acceptable after evaluating LLOQ samples and QC samples over three days of validation (see tables).
Extraction recovery was found to be 90%.
Specificity of the assay was determined by injection of six lots of blank plasma, and injection of a solution of over-the-counter drugs at 100 ug/mL. In all cases, no interferences were seen othe than peaks from endogenous estrone and estradiol.
Stability of the analytes was established in cleansed plasma after 5 freeze/thaw cycles, 24h room temperature exposure and 1 month at –70°C.
Matrix effect was evaluated by analyzing 6 different lots of normal (uncleansed) plasma, and measuring the endogenous concentrations; the same 6 lots of plasma were also fortified with estrone and estradiol at 25 pg/mL and analyzed; analytical recovery was calculated by subtracting the individual endogenous concentrations from the spiked samples and expressing the amount remaining as a percentage of 25 pg/mL. In this way the absolute accuracy of the method could be evaluated with authentic plasma. The results were within +/- 15% of the theoretical amount added for all 6 lots, indicating the rugged nature of the assay.
Conclusions
This method was successfully used to determine concentrations of estrone and estradiol in human plasma samples generated from clinical trials. The method proved to be sufficiently rugged even at the extremely low concentrations (5 pg/ml) required.