Endocrine Abstracts (2015) 38 P398 | DOI: 10.1530/endoabs.38.P398

Discrimination of adrenocortical carcinoma from other adrenal lesions: use of a new 13 steroid serum panel based on LC-MS/MS

David Taylor1, Lea Ghataore1, Royce Vincent1, Roy Sherwood1, Ben Whitelaw2, Dorota Dworakowska2, Klaus-Martin Schulte3, Salvadore Diaz-Cano4, Dylan Lewis5, Simon Aylwin2 & Norman Taylor1

1Department of Clinical Biochemistry (Viapath Analytics), King’s College Hospital, London, UK; 2Department of Endocrinology, King’s College Hospital, London, UK; 3Department of Surgery, King’s College Hospital, London, UK; 4Department of Histopathology, King’s College Hospital, London, UK; 5Department of Radiology, King’s College Hospital, London, UK.

Adrenocortical carcinoma (ACC) is a rare malignancy, but accounts for up to 11% of adrenal masses investigated in referral centres. Diagnosis remains a challenge. Up to two thirds are biochemically inactive, resulting from de facto enzyme deficiencies in the steroid hormone biosynthetic pathways, as shown by urine steroid profiling by gas chromatography-mass spectrometry. Increased metabolites of pathway intermediates in ACC discriminate it from benign adrenal lesions and provide markers for follow up. Serum assays for most intermediates (e.g. 17-hydroxypregnenolone) are unavailable, due to low demand or lack of immunoassay specificity. Serum steroid analysis by liquid chromatography–tandem mass spectrometry (LC–MS/MS) is increasingly replacing immunoassay, especially for those most subject to cross-reaction. We have developed an LC–MS/MS method for measurement of testosterone, progesterone, androstenedione, DHEAS, pregnenolone, 11-deoxycorticosterone, corticosterone, 17-hydroxypregnenolone, 17-hydroxyprogesterone, 11-deoxycortisol, 21-deoxycortisol, cortisol, and cortisone in serum. We investigated its utility in discriminating ACC (six cases) from 24 non-ACC adrenal lesions (11 phaeochromocytoma, three cortisol-producing and three aldosterone-producing adenomas, and seven lesions demonstrating no hormonal excess) and 61 healthy controls. Samples containing internal standards were prepared for LC–MS/MS by sequential protein precipitation and liquid–liquid extraction. Steroids were resolved on a reverse-phase C18 column with the MS operated in positive APCI ionisation mode. In the ACC cases, between four and 10 steroids were increased (mean=6), whilst in the non-ACC group up to two steroids were increased. 11-Deoxycortisol was markedly increased in all ACC cases, whilst increases were also seen for androstenedione (five cases), 17-hydroxyprogesterone (four cases) and pregnenolone, 17-hydroxypregnenolone, 11-deoxycorticosterone, and DHEAS (three cases each). The cortisol:11-deoxycortisol ratio best discriminated between ACC (mean=14.9), non-ACC (335.9), and healthy controls (324.9, P=0.003). In conclusion, serum steroid panelling by LC–MS offers a promising diagnostic test for ACC by combining the measurement of steroid hormones and their precursors in a single analysis.