LAUSR

OBJECTIVE With the widespread use of sex-steroid hormones in contraceptives and hormone replacement therapy, there is an increasing need for reliable analytical methods. We report the development of a sensitive and robust UPLC-MS/MS method for quantitation of both endogenous and synthetic sex-steroid hormones in human serum. STUDY DESIGN We developed and validated a UPLC-MS/MS method to quantify progestogens (etonogestrel, levonorgestrel, medroxyprogesterone acetate, norethindrone, progesterone) and estrogens (estradiol and ethinyl estradiol) with good accuracy, high sensitivity, and excellent robustness. We then applied the method to the analysis of sex-steroid hormones in serum from 451 clinical research participants. RESULTS Each UPLC-MS/MS analysis was 6.5 min. The lower limits of quantitation (LLOQs) were 25 pg/ml for the progestogens, and 2.5 and 5.0 pg/ml for estradiol and ethinyl estradiol, respectively. When estradiol was analyzed without assessment of progestogens, the LLOQ was reduced to 1 pg/ml. The calibration curves were linear from 25-50,000, 2.5-2000 (1-2000 for estrogens-only analysis) and 5-2000 pg/ml, respectively. Both the accuracy and precision were below±15% not only for routine validation (intraday and interday), but for long-term (>2 years) assay robustness with external controls, thereby, demonstrating the utility of this method for multi-year clinical trial assessments of progestogens and estrogens. We applied the method to quantify sex-steroid levels in 1804 clinical samples. CONCLUSIONS We successfully developed a UPLC-MS/MS method, and overcame the matrix suppression to allow sensitive quantitation of both synthetic and endogenous sex-steroid hormones in human serum. IMPLICATIONS We developed a sensitive and robust UPLC-MS/MS method to accurately measure the levels of sex-steroid hormones in serum. The method overcame matrix interference barriers and achieved excellent long-term stability and reproducibility (≥96.9% accuracy; ≤13.0% relative variability measured with external controls over 2 years), demonstrating its utility in clinical sample analysis.