An untargeted metabolomics approach to evaluate enzymatically deconjugated steroids and intact steroid conjugates in urine as diagnostic biomarkers for adrenal tumors.

OBJECTIVES

Urinary steroid profiling after hydrolysis of conjugates is an emerging tool to differentiate aggressive adrenocortical carcinomas (ACC) from benign adrenocortical adenomas (ACA). However, the shortcomings of deconjugation are the lack of standardized and fully validated hydrolysis protocols and the loss of information about the originally conjugated form of the steroids. This study aimed to evaluate the quality of the deconjugation process and investigate novel diagnostic biomarkers in urine without enzymatic hydrolysis.

METHODS

24 h urine samples from 40 patients with ACC and 40 patients with ACA were analyzed by untargeted metabolomics using liquid chromatography-high-resolution mass spectrometry both unmodified and after hydrolysis with arylsulfatase/glucuronidase from Both approaches were compared regarding the differentiation of ACC vs. ACA via ROC analyses and to evaluate the hydrolyzation efficiency of steroid conjugates.

RESULTS

Steroid glucuronides were fully deconjugated, while some disulfates and all monosulfates were still largely detectable after enzymatic hydrolysis, suggesting incomplete and variable deconjugation. In unhydrolyzed urine, steroid monosulfates showed the best differentiation between ACC and ACA (highest AUC=0.983 for CHOS, followed by its isomer and two isomers with the molecular formula CHOS). Moreover, several disulfates were highly abundant and increased in ACC compared to ACA.

CONCLUSIONS

This work highlights the limitations of hydrolyzing steroid conjugates before analysis and shows a possible superiority of a direct analysis approach compared to a hydrolysis approach from a methodological point of view and regarding diagnostic accuracy. Several steroid conjugates were found as promising diagnostic biomarkers for differentiation between ACC and ACA.