Novel urine test for adrenal cancer

Researchers from the Johann Bernoulli Institute, University of Groningen, and from the Centre for Endocrinology, Diabetes and Metabolism at the University of Birmingham/UK, have developed a novel urine test for the diagnosis of adrenal cancer. This breakthrough is reported in the December issue of the Journal of Clinical Endocrinology and Metabolism, 96: 3775 - 3784 (2011).

Adrenal tumors affect about 2% of the general population and become more common with age.A small but significant fraction of these tumors turns out to be malignant. Adrenal cancer is an aggressive disease with poor survival rates, and its detection constitutes a major diagnostic challenge. The novelurine test bears the promise to provide a faster and more reliable diagnostic tool than current methods, including costly imaging techniques

In collaboration with the European Network for the Study of Adrenal Tumors, the Birmingham University team led by Prof. Dr. Wiebke Arlt measured the excretion of steroid hormones in urine samples from about 150 patients with adrenal tumors. Computer scientists from the Johann Bernoulli Institute applied a new method called Matrix Relevance Learning in order to analyse the steroid excretion profiles. This way, the “hormone fingerprint” of malignant tumors could be determined. Moreover, the computational analysis identifies the most informative steroid markers. Forthcoming studies will aim the validation of the developed test in clinical practice.

Matrix Relevance Learning was developed within the Intelligent Systems Group at the Johann Bernoulli Institute, Groningen, and its application in the context of adrenal cancer was put forward by:

Prof.Dr. Michael Biehl, Professor in Computer Science

Han Stiekema, former MScstudent at the Johann Bernoulli Institute

Dr. Petra Schneider, former PhD student at the Johann Bernoulli Institute, now Postdoctoral Researcher at the University of Birmingham

Further information:
www.cs.rug.nl/~biehl
www.ensat.org
http://www.birmingham.ac.uk/staff/profiles/cem/EDM/Arlt-Wiebke.aspx