Endocrine Abstracts

Introduction: Urine free cortisol measurements are routinely performed to evaluate hypercortisolism. Despite their analytical inaccuracy, immunoassay-based methods are frequently used. Advances in liquid chromatography high resolution mass spectrometry (LC-HRMS) facilitate the incorporation of powerful diagnostic tools into clinical laboratories with high analytical specificity that also allow simultaneous quantification of different metabolites and untargeted compound identification, which may be helpful to identify other clinically relevant metabolites or drugs.

Objectives: The aim was to validate a simple routine LC–HRMS method to quantify simultaneously cortisol, cortisone, 6β-hydroxycortisol, and 18-hydroxycortisol in human urine. In addition to the key role of urine cortisol measurements in Cushing’s syndrome, the cortisol/cortisone ratio (11βHSD2 activity) is a sensitive marker for the apparent mineralocorticoid excess syndrome, the 6β-hydroxycortisol/cortisol ratio is a marker for CYP3A activity, and 18-hydroxycortisol levels are useful in the evaluation of primary hyperaldosteronism, especially familial forms.

Methods: Urine extraction was liquid-liquid with dichloromethane, chromatographic separation was performed by micro-LC and detection used accurate masses of steroids and simultaneous high resolution full scan acquisition. Validation included the assessment of linearity, matrix effect, accuracy, imprecision, selectivity, carry-over and comparison with a GC-MS method. We also quantified the 24-h urine excretion of the four steroids by LC-HRMS in 60 samples of patients with clinical suspicion or follow-up for hypercortisolism. The cross-reactivity of the measured steroids with commercial cortisol immunoassays (Liaison and Atellica) was also investigated.

Results: The analytical method was free of matrix effect and presented acceptable analytical performance. Cortisol concentrations in the urine of patients significantly correlated with cortisone and 6β-hydroxycortisol but not with 18-hydroxycortisol. Patients with antisteroidogenic drugs presented reduced CYP3A4 activity as assessed by the 6β-hydroxycortisol/cortisol ratio. The urines of patients without antisteroidogenic drugs (n=41), metyrapone (n=12), ketoconazole (n=4) and osilodrostat (n=3) presented, respectively, 6β-hydroxycortisol/cortisol ratios of 4.1±0.5, 2.2±0.5, 1.5±0.6 and 0.1±0.1 (P<0.05). In contrast, no differences were observed between these groups in the 11β-HSD2 activity calculated as the ratio cortisol/cortisone. Finally, cross-reactivity with 6β-hydroxycortisol and cortisone was demonstrated in cortisol immunoassays, which presented significantly higher urine cortisol results in comparison with LC-HRMS.

Conclusion: A rapid and accurate routine LC-HRMS method was validated, which is useful for routine evaluation of hypercortisolism and other disorders of glucocorticoid and mineralocorticoid metabolism. Cross-reactivity with 6β-hydroxycortisol may impact cortisol immunoassay reliability to monitor hypercortisolism medical therapies.