A molecular map of mesenchymal tumors

Stephen R Henderson¹ , David Guiliano^*¹^,2 , Nadege Presneau^*¹ , Sean McLean¹ , Richard Frow¹^,3 , Sonja Vujovic¹ , John Anderson⁴ , Neil Sebire⁵ , Jeremy Whelan⁶ , Nick Athanasou⁷ , Adrienne M Flanagan³ and Chris Boshoff¹

¹Cancer Research UK, Viral Oncology Group, Wolfson Institute for Biomedical Research, Gower Street, University College London, London, WC1E 6BT, UK

²Division of Cell and Molecular Biology, Biochemistry Building, Faculty of Life Sciences, Imperial College London, South Kensington Campus, London, SW7 2AZ, UK

³Institute of Orthopaedics and Department of Pathology, Royal National Orthopaedic Hospital, Stanmore, Middlesex, HA7 4LP, UK

⁴Unit of Molecular Haematology and Cancer Biology, Institute of Child Health and Great Ormond Street Hospital, Guildford Street, London, WC1N 1EH, UK

⁵Department of Pathology, Great Ormond Street Hospital for Children, London, WC1N 3JH, UK

⁶London Bone and Soft Tissue Tumour Service, University College London Hospitals, London, UK

⁷Department of Pathology, Nuffield Department of Orthopaedic Surgery, Nuffield Orthopaedic Centre, Headington, Oxford, OX3 7LD, UK

author email corresponding author email* Contributed equally

Genome Biology 2005, 6:R76doi:10.1186/gb-2005-6-9-r76


Published:	26 August 2005

Subject areas: Cancer, Medicine, Genome studies, Bioinformatics

Abstract

Background

Bone and soft tissue tumors represent a diverse group of neoplasms thought to derive from cells of the mesenchyme or neural crest. Histological diagnosis is challenging due to the poor or heterogenous differentiation of many tumors, resulting in uncertainty over prognosis and appropriate therapy.

Results

We have undertaken a broad and comprehensive study of the gene expression profile of 96 tumors with representatives of all mesenchymal tissues, including several problem diagnostic groups. Using machine learning methods adapted to this problem we identify molecular fingerprints for most tumors, which are pathognomonic (decisive) and biologically revealing.

Conclusion

We demonstrate the utility of gene expression profiles and machine learning for a complex clinical problem, and identify putative origins for certain mesenchymal tumors.