A quantitative, broad-panel LC-MS/MS method for the analysis of intact steroid conjugates: A novel approach to steroid profiling for biomarker research in corticoid-dependent diseases

In endocrine oncology changes in hormone synthesis and metabolism play a major role in disease formation and progression. Accordingly, steroid metabolism is a possible pathophysiological factor in adrenal tumours and metabolic conjugates of steroids are promising biomarker candidates for the clinically challenging differential diagnosis of highly prevalent, benign adrenocortical adenomas versus rare, aggressive adrenocortical carcinomas. The analysis of steroidal compounds in human biological samples is most accurately achieved by mass spectrometry-based approaches. Commonly, metabolic end products of steroids, such as sulfate and glucuronide conjugates, are hydrolysed before measurements to simplify the procedure. Therefore, data on actual steroid conjugate concentrations in humans is scarce. In this work we present a compact LC-MS/MS method for the absolute quantification of 22 intact steroid conjugates in urine and plasma with an instrument run time of 11 min and a straightforward sample preparation procedure. The method was successfully validated according to established international guidelines and applied to samples of patients with adrenal tumours to demonstrate the method's suitability regarding sample preparation, analyte selection and quantifiable concentration ranges. Furthermore, the exploratory comparison of steroid abundance in the patient samples support the previously proposed potential of steroid conjugates as differentiating biomarkers between adrenocortical carcinoma and adenoma (e.g., 17-OH-pregnenolone sulfate, median plasma concentration in carcinoma (114 ng/ml) versus adenoma (3.76 ng/ml), p < 0.001). The newly developed method enables absolute quantification of multiple steroid conjugates in humans which provides the basis for profound biomarker research for the early detection of adrenocortical carcinomas and valuable data for related research questions.