Using mass spectrometry to overcome the longstanding inaccuracy of a commercially-available clinical testosterone immunoassay

Abstract

Accurate measurement of testosterone is important for the diagnosis of gonadal disorders in men, women, and children. Testosterone measurement has limited accuracy at low concentrations by most commercially available immunoassays. We aimed to develop an LC-MS/MS assay to address the inaccuracy of the in-house immunoassay observed over the past decade and to replace it with the new assay. Testosterone in serum/plasma was extracted with commercial supported liquid extraction plates. Method validation was performed following the CLSI C62-A guideline. A total of 126 samples were used for method comparison between the Beckman UniCel DxI immunoassay and LC-MS/MS. Results by immunoassay were 20% lower compared with LC-MS/MS and had minimal correlation (R2 = 0.403) with LC-MS/MS below 100 ng/dL. When comparing specimens from the Accuracy-Based Survey from the College of American Pathologists, the newly developed assay agreed well with the CDC reference measurement procedure. In summary, immunoassay measurement of testosterone can be significantly inaccurate, especially at low concentrations. The newly developed LC-MS/MS assay provides accurate results across the entire measurable range.