The accurate measurement of androstenedione in human serum and plasma is required for steroid profiling to assure the appropriate diagnosis and differential diagnosis of hyperandrogenism. In this work, we introduce an isotope dilution liquid chromatography–tandem mass spectrometry (LC–MS/MS) candidate reference measurement procedure for the quantification of androstenedione in human serum and plasma. The performance of the procedure enables its use in the evaluation and standardization of routine assays and for the evaluation of patient samples to ensure the traceability of individual patient results. As the primary standard, a certified reference material from NMIA (National Measurement Institute, Australia) was used. Additionally, a quantitative nuclear magnetic resonance (qNMR) method was developed for the value assignment of the primary reference material, which ensures the direct traceability to SI units, as well as the independence from the availability of reference materials. ¹³C₃-labeled androstenedione was used as the internal standard. The introduced method allows the measurement of androstenedione in the range of 0.05–12 ng/mL, and the assay imprecision was found to be <2% between 5 and 12 ng/mL, 3.5% at 1.5 ng/mL, and 5.2% at 0.05 ng/mL, with an accuracy of 95–105% for the serum and 91–103% for the plasma matrix. The transferability to a second laboratory was validated by method comparison based on 112 patient samples. The comparison of the results obtained from the presented method and an LC–MS/MS routine assay, using 150 native patient samples, showed a good correlation with a bias of the routine method of ≤4.0%.

类固醇分析需要准确测量人血清和血浆中的雄烯二酮，以确保对雄激素过多症进行适当的诊断和鉴别诊断。在这项工作中，我们介绍了一种同位素稀释液相色谱-串联质谱 (LC-MS/MS) 候选参考测量程序，用于定量人血清和血浆中的雄烯二酮。该程序的性能使其可用于常规检测的评估和标准化以及患者样本的评估，以确保单个患者结果的可追溯性。作为主要标准，使用来自 NMIA（澳大利亚国家测量研究所）的认证参考材料。此外，¹³ C ₃标记的雄烯二酮用作内标。引入的方法允许测量 0.05-12 ng/mL 范围内的雄烯二酮，并且发现测定不精确度在 5 和 12 ng/mL 之间 <2%，在 1.5 ng/mL 时为 3.5%，在 5.2% 时为0.05 ng/mL，血清准确度为 95-105%，血浆基质准确度为 91-103%。通过基于 112 名患者样本的方法比较验证了向第二个实验室的可转移性。使用 150 个本地患者样本，比较从本方法和 LC-MS/MS 常规分析获得的结果，显示与 ≤ 4.0% 的常规方法偏差的良好相关性。