|Liquid chromatography-tandem mass spectrometry (LC-MS/MS) has become an increasingly common tool in the clinical laboratory for the measurement of steroid hormones. Methods using this platform are capable of achieving measurements of multiple analytes in a small sample volume with high sensitivity and specificity. We report here the development of a multiplexed LC-MS/MS assay for quanititation of ten steroid hormones in 200µL of human serum: 11-deoxycortisol, 17á-hydroxyprogesterone (17-OHP), aldosterone, androstenedione, corticosterone, cortisol, DHEA (dehydroepiandrosterone), DHEAS (dehydroepiandrosterone sulfate), progesterone and testosterone. The linear range for each analyte was determined as well as precision and recovery within this range using off-the-clot (OTC) human serum spiked with steroids and calibrators formulated in double charcoal stripped (DCS) human serum. A comprehensive analysis of assay interferences was also done.|
All studies were performed on a Shimadzu Prominence HPLC and API 5000 MS equipped with an atmospheric pressure chemical ionization (APCI). Data was acquired in multiple reaction monitoring (MRM) mode. The concentrations of steroids were determined using calibration curves derived from DCS human serum calibrators and isotopically labeled internal standards used for signal normalization. The measurable linear concentration ranges for the analytes span the relevant clinical ranges and are as low as 10-11g /mL (e.g., testosterone, aldosterone) and as high as 10-6 g/mL (e.g., DHEAS). The concentration dynamic range is approximately 200 for each analyte. Spiked OTC human serum samples were used to determine the precision and recovery of the assay. Based on 3 injections per run and 20 runs, %CV values ranged from 2.3-8.8% for within run precision, 2.2-15.1% for between run precision and 3.2-15.4% for total imprecision for all steroids within the range of the calibrators. Higher %CV values occurred for low concentration samples. Recoveries varied from 79 to 101% for the 10 steroids spiked in OTC human serum with %CV values for the measurement ranging from 4-14% for the 20 runs.
A thorough investigation of potential interferences for the assay was also performed. The core steroid structure is common to many naturally occurring biological molecules and drugs and often plagues accurate quantitation in steroid assays. A comprehensive investigation of structural interferences revealed 8 molecules that are structural variants of steroids and posed positive assay interference as revealed by dose response studies. Another 21 molecules were estimated to cause potential assay interferences if present in the sample.