OBJECTIVE: To determine the prognostic relevance of neuroendocrine differentiation in colorectal cancer. METHODS: The survival of 116 patients with colorectal cancer of stages III (n = 59) and IV (n = 57) was correlated with the extent of neuroendocrine differentiation. Chromogranin A and synaptophysin were used as neuroendocrine markers. Based on the degree of immunoreactivity for these markers, tumours were classified as 0 (no expression of neuroendocrine markers), 1 (< 2% cells staining positive for neuroendocrine markers) and 2 (> 2% cells staining positive for neuroendocrine markers). Patients were followed up for more than 5 years or until death. RESULTS: Seven of 59 (11.8%) stage III cancers and 13/57 (22.8%) stage IV cancers belonged to group 2. The 96 patients of groups 0 and 1 lived for 48.9 months, whereas the 20 patients of group 2 survived for only 18.6 months (Kaplan-Meier survival curves, P < 0.001). The difference was most striking in stage III disease with 79.4 months' survival for combined groups 0 and 1, and 38.9 months' survival for group 2 (P < 0.01). Using the multivariate Cox regression model, the presence of more than 2% of cells with neuroendocrine differentiation was found to be an independent prognostic parameter for stage III and IV disease. No correlation was observed between neuroendocrine differentiation and tumour location, grade, depth of invasion or stage. CONCLUSION: Neuroendocrine differentiation is often seen in colorectal cancer. It is an independent prognostic factor in stage III-IV colorectal cancer.
OBJECTIVE: To determine the prognostic relevance of neuroendocrine differentiation in colorectal cancer. METHODS: The survival of 116 patients with colorectal cancer of stages III (n = 59) and IV (n = 57) was correlated with the extent of neuroendocrine differentiation. Chromogranin A and synaptophysin were used as neuroendocrine markers. Based on the degree of immunoreactivity for these markers, tumours were classified as 0 (no expression of neuroendocrine markers), 1 (< 2% cells staining positive for neuroendocrine markers) and 2 (> 2% cells staining positive for neuroendocrine markers). Patients were followed up for more than 5 years or until death. RESULTS: Seven of 59 (11.8%) stage III cancers and 13/57 (22.8%) stage IV cancers belonged to group 2. The 96 patients of groups 0 and 1 lived for 48.9 months, whereas the 20 patients of group 2 survived for only 18.6 months (Kaplan-Meier survival curves, P < 0.001). The difference was most striking in stage III disease with 79.4 months' survival for combined groups 0 and 1, and 38.9 months' survival for group 2 (P < 0.01). Using the multivariate Cox regression model, the presence of more than 2% of cells with neuroendocrine differentiation was found to be an independent prognostic parameter for stage III and IV disease. No correlation was observed between neuroendocrine differentiation and tumour location, grade, depth of invasion or stage. CONCLUSION:Neuroendocrine differentiation is often seen in colorectal cancer. It is an independent prognostic factor in stage III-IV colorectal cancer.