Role of RHAMM within the hierarchy of well-established prognostic factors in colorectal cancer
- 1Institute of Pathology, University Hospital of Basel, Basel, Switzerland
- 2Department of Radiation Oncology, McGill University Health Centre, Montreal, Canada
- 3Department of Cellular Pathology, St. Mark’s Hospital, Middlesex, UK and Imperial College, London, UK
- Dr I Zlobec, Institute of Pathology, University Hospital of Basel, Schönbeinstrasse 40, Basel, CH-4031, Switzerland;
- Revised 12 March 2008
- Accepted 9 April 2008
- Published Online First 24 April 2008
Objective: To compare the independent prognostic effect of a panel of immunohistochemical protein markers in colorectal cancer (CRC) and determine their ranking among the established prognostic factors T stage, N stage, vascular invasion, tumour budding and tumour grade.
Design: A tissue microarray of 1420 CRCs was immunostained for 23 markers and mismatch repair (MMR) proteins. Immunoreactivity was assessed semi-quantitatively. Receiver operating characteristic (ROC) curves were used to determine cut-off scores for tumour marker positivity. Survival time was investigated for each marker in multivariable analysis with T stage, N stage, vascular invasion, tumour budding and tumour grade. The hazard ratio (HR) was used to compare the prognostic effect of each marker on 5 year survival.
Results: To the standard prognostic features, only six markers added independent prognostic information including receptor for hyaluronic acid mediated motility (RHAMM) (HR = 2.39 (1.88 to 3.05)), epidermal growth factor receptor (HR = 1.65 (1.31 to 2.09)), tumour infiltrating lymphocytes (HR = 0.7 (0.54 to 0.92)), urokinase plasminogen activator (HR = 1.38 (1.09 to 1.75)), Raf-1 kinase inhibitor protein (HR = 0.75 (0.58 to 0.96)) and mammalian sterile 20-like kinase 1 (MST1) (HR = 0.75 (0.58 to 0.95). Diffuse (>90% staining) expression of RHAMM ranked above T stage, vascular invasion, tumour budding and tumour grade in terms of adverse prognostic significance and was associated with distant metastasis (p = 0.012) and with worse outcome in patients with metastatic disease (p = 0.031).
Conclusions: The strong adverse effect of RHAMM on outcome in addition to its position within the hierarchy of well-established prognostic factors suggest that RHAMM should be considered a more important prognosticator than tumour grade, tumour budding and vascular invasion in patients with CRC.
Funding: UG and LTe were supported by the Swiss Cancer League (OSC – 01265-08-2002) and IZ was supported by the Novartis Foundation, formerly the Ciba–Geigy Jubilee Foundation.
Competing interests: None.
Ethics approval: The use of tissue for this study was approved by the local Ethics Committee of the University Hospital of Basel on 3 November 2003.