Molecular determinants of prognosis in oesophageal cancer patients treated with chemoradiation

Sir,

The paper by Yoshikawa et al (2008) is of considerable interest to our group. The authors assessed the expression of Gli-1 in the residual squamous cell carcinoma of the oesophagus in the resected specimens of the patients treated with preoperative chemoradiation. The expression of Gli-1 correlated with poor prognosis. When employing the strategy of preoperative chemoradiation in oesophageal carcinoma, the prognosis of the patients can be determined by assessing the degree of residual cancer (Chirieac et al, 2005; Wu et al, 2007). In addition, the residual cancer correlates with metastatic potential (Rohatgi et al, 2005). It would be of interest to document that Gli-1 expression has an independent prognostic value. The authors did not report a multivariate analysis. Our group first documented the expression and temporal kinetics of sonic hedgehog ligand and Gli-1 protein expression in a xenograft model using oesophageal cancer cell lines (Sims-Mourtada et al, 2006). Both biomarkers correlated with chemoradiation resistance and dynamics of tumour repopulation after chemoradiation (Sims-Mourtada et al, 2006). Furthermore, our group documented that sonic hedgehog signalling mediates chemotherapy resistance through multiple drug transporter mechanisms and that this resistance can be overcome by inhibitors of the sonic hedgehog pathway (Sims-Mourtada et al, 2007). Unfortunately, Yoshikawa et al (2008) have neglected to discuss these two publications.