ESPE Abstracts (2018) 89 RFC1.4

Mass Spectrometry-Based Assessment of Childhood Androgen Excess in 487 Consecutive Patients Over 5 Years

Jan Idkowiaka,b,c, Yasir S Elhassana,c, Pascoe Manniona,c, Karen Smithd, Rachel Websterd, Vrinda Saraffb,c, Timothy G Barrettb,c, Nick J Shawa,b,c, Nils Kroneb,c,e, Renuka P Diasb,c, Melanie Kershawb,c, Jeremy Kirka,b,c, Ruth E Kroneb,c, Michael W O’Reillya,c & Wiebke Arlta,c


aInstitute of Metabolism and Systems Research, University of Birmingham, Birmingham, UK; bDepartment of Paediatric Endocrinology, Birmingham Women’s and Children’s Hospital NHS Foundation Trust, Birmingham, UK; cCentre for Endocrinology, Diabetes and Metabolism, Birmingham Health Partners, Birmingham, UK; dDepartment of Clinical Biochemistry, University Hospitals Birmingham NHS Foundation Trust, Birmingham, UK; eAcademic Unit of Child Health, Department of Oncology & Metabolism, University of Sheffield, Sheffield, UK


Background: Androgen excess in childhood is a common clinical presentation and might signify serious pathology. We have recently explored patterns and severity of androgen excess in a large female adult cohort to differentiate common polycystic ovarian syndrome (PCOS) from non-PCOS pathology, including congenital adrenal hyperplasia (CAH), ovarian hyperthecosis and adrenal and ovarian tumours (Elhassan et al., JCE&M 2018). Herein, we undertake a similar approach for the differential diagnosis of childhood androgen excess.

Objective: To examine the diagnostic utility of simultaneous measurement of serum dehydroepiandrosterone sulfate (DHEAS), androstenedione (A4), and testosterone (T) to delineate the biochemical signatures of conditions underlying childhood androgen excess.

Design: Retrospective review of all children undergoing serum androgen measurement at a large tertiary care referral centre over 5 years (2013–2017). Serum A4 and T were measured by tandem mass spectrometry, DHEAS by immunoassay; results were interpreted using Tanner-stage defined cut-offs for androgen excess. Patients with at least one increased androgen underwent phenotyping by clinical notes review.

Results: 1525 children underwent serum androgen measurements in the 5-year period; in 487 children, DHEAS, A4, and T were measured simultaneously, with at least one increased androgen in 41% (n=199; 141 girls and 58 boys). Premature adrenarche (PA) was the most common diagnosis (42%), followed by PCOS (12.6%) and CAH (7.0%). In 13% of children, the cause of borderline androgen excess could not be established. There was one case of adrenocortical carcinoma (ACC). PA is characterised by raised DHEAS levels in 85% of cases. A4 was raised in 26% of PA children, T in only 9%. CAH is characterised by A4 excess in 86% of patients; T was raised in 35% and DHEAS in only 21% of CAH cases. In adolescent PCOS, the distribution of androgen excess levels was similar for DHEAS, A4 and T (50, 42 and 42%, respectively). In the ACC case, we observed an isolated, severe increase in DHEAS (28-fold above upper limit of normal).

Conclusions: To our knowledge, this is the first systematic analysis of mass spectrometry-based serum androgen profiling in a large sample of childhood androgen excess. PA was the commonest condition and is characterised by DHEAS excess in the majority of cases, whereas CAH most frequently presents with A4 excess and normal DHEAS. In adolescent PCOS, DHEAS, A4 and T excess are evenly distributed. ACC is extremely rare in childhood and severe DHEAS excess should prompt urgent investigations for this condition.