AIMS: Chromogranin-A (CG), a cytoplasmic glycoprotein, is one of the markers most frequently used to identify the presence of neuroendocrine cells in the human gastrointestinal tract. Several authors have identified a subgroup of colorectal cancer patients with a sever prognosis whose tumors contained neuroendocrine CG-positive cells. In the present study, CG expression in 100 patients with colorectal adenocarcinoma treated from January 1983 to December 1988 with potentially curative surgery was analyzed and correlated with other prognostic factors and 5-year survival rate. METHODS: Samples tested immunohistochemically for CG were divided into three groups: I) negative; II) less than 1 CG-positive cell/mm2; III) more than 1 CG-positive cell/mm2. RESULTS: Of 100 patients with primary colorectal adenocarcinoma, 79% had tumors comprised of CG-negative cells, 17% had rare CG-positive cells, and 4% of cases could be classified in group III. No significant relation between CG expression and location of primary tumor, bowel wall infiltration, stage of disease or tumor grade according to Broders and Jass was observed. The 5-year survival was 53% and 52% for CG-positive and CG-negative lesions, respectively. Survival of patients with Dukes-Kirklin stage C and D was comparable in patients with CG-positive (33.3%) and CG-negative (30%) tumors. CONCLUSIONS: CG expression cannot, at present, be recommended as a marker to identify prognostic subgroups in colorectal cancer patients.
AIMS: Chromogranin-A (CG), a cytoplasmic glycoprotein, is one of the markers most frequently used to identify the presence of neuroendocrine cells in the human gastrointestinal tract. Several authors have identified a subgroup of colorectal cancer patients with a sever prognosis whose tumors contained neuroendocrine CG-positive cells. In the present study, CG expression in 100 patients with colorectal adenocarcinoma treated from January 1983 to December 1988 with potentially curative surgery was analyzed and correlated with other prognostic factors and 5-year survival rate. METHODS: Samples tested immunohistochemically for CG were divided into three groups: I) negative; II) less than 1 CG-positive cell/mm2; III) more than 1 CG-positive cell/mm2. RESULTS: Of 100 patients with primary colorectal adenocarcinoma, 79% had tumors comprised of CG-negative cells, 17% had rare CG-positive cells, and 4% of cases could be classified in group III. No significant relation between CG expression and location of primary tumor, bowel wall infiltration, stage of disease or tumor grade according to Broders and Jass was observed. The 5-year survival was 53% and 52% for CG-positive and CG-negative lesions, respectively. Survival of patients with Dukes-Kirklin stage C and D was comparable in patients with CG-positive (33.3%) and CG-negative (30%) tumors. CONCLUSIONS: CG expression cannot, at present, be recommended as a marker to identify prognostic subgroups in colorectal cancer patients.