Endocrine Abstracts

JOINT1173

Background: Quantification of urinary free cortisol excretion is one of the recommended initial screening tests for diagnosing hypercortisolism. The use of liquid chromatography–tandem mass spectrometry (LC–MS/MS) allows for the simultaneous quantification of both urinary cortisol and cortisone. This study aims to determine the optimal cut-off values for urinary free cortisol (UFCortisol) and urinary free cortisone (UFCortisone) in the diagnosis of neoplastic hypercortisolism and to determine whether either one has superior diagnostic accuracy.

Methods: In this retrospective study, we analyzed data from subjects who were screened for endogenous hypercortisolism at Amsterdam UMC between December 2015 and February 2022. Subjects had been instructed to collect urine for 24 hours on two consecutive days. UFCortisol and UFCortisone concentrations were measured using an in-house developed LC–MS/MS method. The average of both collections was used to determine the diagnostic accuracy. Confirmation of the diagnosis of neoplastic hypercortisolism was based on (postoperative) clinical parameters and histopathology, while the diagnosis was excluded based on a lack of progression of clinical signs during follow-up of at least 12 months. The diagnostic accuracy was evaluated by ROC-curve analysis.

Results: We included 426 subjects without neoplastic hypercortisolism (test indication: 136 clinical signs/symptoms, 274 adrenal incidentaloma, 9 pituitary incidentaloma, 2 genetic predisposition, 5 unclear) and 27 subjects with neoplastic hypercortisolism (test indication: 23 clinical signs/symptoms, 2 adrenal incidentaloma, 2 pituitary incidentaloma) in which UFCortisol was measured. In a subgroup of 304 subjects without and 17 subjects with neoplastic hypercortisolism UFCortisone was measured as well. The median UFCortisol concentration was 369 nmol/24 h (range 47–8987) for subjects with and 63 nmol/24 h (1.3–614) for subjects without neoplastic hypercortisolism (P<0.001). The median UFCortisone concentration was 606 nmol/24 h (131–1174) for subjects with and 190 nmol/24 h (13–863) for subjects without neoplastic hypercortisolism (P<0.001). For UFCortisol, a cut-off value of 120 nmol/24 h provided optimal diagnostic accuracy with a sensitivity of 88.9% (95% CI 0.771–1.000) and specificity of 85.9% (0.826–0.892). The optimal cut-off value for UFCortisone was 300 nmol/24 h with a sensitivity of 94.1% (0.829–1.000) and specificity of 87.8% (0.841–0.915). The sum of UFCortisol and UFCortisone provided a slightly higher diagnostic accuracy at a cut-off of 460 nmol/24 h with a sensitivity of 94.1% (0.829–1.000) and specificity of 93.0% (0.901–0.959).

Conclusion: We established cut-off values for UFCortisol and UFCortisone for the diagnosis of neoplastic hypercortisolism. UFCortisone had similar diagnostic accuracy to UFCortisol, but the sum UFCortisol+UFCortisone seemed to perform slightly better.