A New LC-MS/MS Assay for Serum Cortisol, Cortisone, and Dexamethasone: Real-World Utility for the Interpretation of the 1-mg Overnight Dexamethasone Suppression Test

Abstract Objective The 1-mg dexamethasone suppression test (DST) may be affected by dexamethasone (DEX) exposure and assay-related variability. We aimed to define pragmatic post-DST DEX benchmarks and to compare the diagnostic performance of cortisol measured by chemiluminescent immunoassay (CLIA) versus liquid chromatography-tandem mass spectrometry (LC-MS/MS). Design Consecutive real-world observational study. Methods Post-test cortisol was measured by CLIA, while cortisol, cortisone and DEX were quantified by LC-MS/MS on stored samples. Diagnoses were clinically adjudicated. ROC analyses and DeLong tests compared AUCs. Results The study involved 636 participants: 351 controls (55.2%), 135 non-functioning adrenal incidentaloma (21.2%), 14 aldosterone-cortisol cosecretion (2.2%), 116 mild autonomous cortisol secretion (18.2%), and 20 overt Cushing’s syndrome (3.2%). Lower-tail DEX values were uncommon and even below lower-tail thresholds, DST results often remained clinically interpretable. DEX was higher in obesity than BMI <30 kg/m2 (4.87±2.26 vs 4.40±2.44 ng/mL; p=0.030). For predicting overt hypercortisolism, AUCs were 0.976 (CLIA cortisol), 0.974 (LC-MS/MS cortisol), and 0.959 (LC-MS/MS cortisone), without significant pairwise differences. For the diagnosis of all forms of hypercortisolism, AUCs were 0.961 (CLIA cortisol), 0.951 (LC-MS/MS cortisol), and 0.931 (cortisone). LC-MS/MS cortisol optimal cut-offs were close to CLIA practice. Conclusions Very low post-DST DEX concentrations are rare, and DST interpretation often remains informative even when DEX is below conventional thresholds, supporting a selective rather than routine use of DEX measurement. LC-MS/MS cortisol provides excellent discrimination with thresholds close to CLIA practice; cortisone appears best suited as an adjunct marker.