[The diagnostic value of serum carcinoma markers, fecal K-ras and p53 gene mutation in pancreatic cancers].

Objective: To evaluate the diagnostic value of serum carcinoma markers CA19-9, CA242, CA50 and carcinoembryonic antigens (CEA), fecal K-ras and p53 gene mutation in pancreatic cancer.

Methods: We collected 136 new cases of pancreatic cancer and 240 patients with benign digestive diseases including 49 patients with benign pancreatic diseases diagnosed in Peking Union Medical College Hospital, Chinese Academy of Medical Sciences Tumor Hospital and Shenyang PLA General Hospital from February 2002 to March 2004. Blood samples were collected and serum carcinoma markers CA19-9, CA242, CA50 and CEA were measured. Fecal K-ras and p53 gene mutation were also measured. We decided the optimal cut-off points with receiver operating characteristic curves and calculated the areas under the curve (AUC).

Results: The AUC of serum CA19-9 and CA242 were 0.855 +/- 0.031 (95% CI 0.794 - 0.916) and 0.859 +/- 0.031 (95% CI 0.799 - 0.920) respectively. The optimal cut-off point for serum CA19-9 was 68 U/ml, with the sensitivity of 84.4% (98/116) and the specificity of 84.3% (145/172) for the diagnosis of pancreatic cancer. The optimal cut-off point for serum CA242 was 25 U/ml, with the sensitivity of 88.4% (84/95) and the specificity of 79.1% (144/182). The sensitivity of fecal K-ras mutation was 77.8%, and the specificity was 82.2%. The sensitivity and specificity of fecal p53 gene mutation were 27.8% and 95.2% respectively. The diagnostic scale of pancreatic cancer was calculated by four variables: serum CA19-9, CA242, fecal K-ras and p53, of which each variable deserved one point if measured positive. The optimal cut-off point for the scale was 2, and the AUC of the diagnostic scale was 0.946 +/- 0.017 (95% CI 0.912 - 0.980).

Conclusions: Serum CA19-9 and CA242 are valuable diagnostic tools for pancreatic cancer. The diagnostic value will be further improved when they are combined with the measurement of fecal K-ras and p53 gene mutation.