ABSTRACT
A prospective study of 62 patients with colorectal cancer was carried out to evaluate the value of tumour-associated trypsin inhibitor (TATI) in the diagnosis and staging of this disease. The reference group consisted of 97 patients with benign gastrointestinal disease. Increased serum TATI levels were detected in 46 patients with colorectal cancer (sensitivity 74%), and the highest mean values were measured in the patients with stage II cancer. The frequency of elevated values was not affected by the stage of the disease, however. There were altogether 64 false positives (specificity 34%). In comparison to CEA and CA 242, TATI showed higher sensitivity at low (< 40%) specificity levels, but it was less sensitive at high (70-95%) specificity levels. The performance of the CA 50 test was poorest both at high and low specificity levels. The various combinations of TATI and other markers (two tests positive) did not offer any further advantage as compared to CEA alone. A marginally significant positive correlation (p = 0.07) was found between serum TATI and CEA values in colorectal cancer patients. The results suggest that TATI is a very sensitive marker in colorectal cancer, but its utility is limited because of its low specificity in symptomatic patients. Concomitant measurement of TATI and other markers does not seem to give any further benefit, and nor does TATI seem to have any significant value in staging of the disease. Our results also indicate that CEA is probably the best current test in colorectal cancer.
Subject(s)
Antigens, Tumor-Associated, Carbohydrate/blood , Biomarkers, Tumor/blood , Carcinoembryonic Antigen/blood , Colorectal Neoplasms/blood , Trypsin Inhibitor, Kazal Pancreatic/blood , Adult , Aged , Aged, 80 and over , Colorectal Neoplasms/diagnosis , Female , Humans , Male , Middle Aged , Predictive Value of Tests , Sensitivity and SpecificityABSTRACT
The aim of the present prospective study was to evaluate the clinical value ol serum tumour markers CEA, CA 50 and CA 242 in patients with colorectal cancer (n = 138) and patients with benign gastrointestinal disease (n = 104). The cutoff levels (90% specificity) determined for each test were 2.5 ng/ml for CEA, 17 U/ml for CA 50 and 17 U/ml for CA 242. The diagnostic sensitivity of the CEA test was 0.63, that of the CA 50 test was 0.30 and 0.30 for the CA 242 test in detecting colorectal cancer. CEA, CA 50 and CA 242 tests were tested in a multivariate analysis to find the best combination of independent predictors of colorectal cancer. The most important predictor of colorectal cancer was CEA followed by CA 242. In order to calculate the contributions of tumour marker tests, a diagnostic score (DS) was developed. The sensitivity of the DS in detecting colorectal cancer was 0.47 with a specificity of 0.88 and an efficiency of 0.67. On the basis of this study, serum CEA and CA 242 seem to possess diagnostic value in preoperative evaluation of patients with colorectal cancer.
