UNLABELLED: We studied the relevance of CEA, CA 19-9, CA 72-4 and the common classical prognostic factors (age, sex, tumor infiltration, N-classification, staging, grading and Lauren classification) in gastric carcinoma. PATIENTS AND METHODS: Survival function estimates were calculated according the method to Kaplan-Meier. The patients were separated into two groups according to preoperative marker levels. Fixing specificity at 100% for healthy people, cut off levels were calculated. Survival curve differences were assessed using the log-rank-test. Multivariate Cox proportional hazards regression analysis was performed. The mantel-Haenszel method was used to assess the 2-year survival rate of patients with gastric carcinoma and high versus low levels of tumor-associated antigens adjusted to tumor stages. The study was performed on the frozen sera (stored at -80 degrees C) of 103 patients with histologically proven gastric carcinoma. RESULTS: The tumor stage (log-rank chi-square = 55.9; P < 0.0001) represents the best prognostic factor besides preoperative values of CA 19-9 (log-rank chi-square = 13.9; P < 0.001) and CEA (log-rank chi-square = 12.2; P < 0.001). CA 72.4 shows a log-rank chi-square of 6.9 (P < 0.01). We found no statistically significant correlation between survival and sex, tumor grade and Lauren classification. The importance of different parameters in providing additional prognostic information was evaluated by multivariate analysis. Only patients after curative surgical intervention (n = 55, R0) were considered. Cox proportional hazards regression analysis yielded an adjusted relative risk of 2.4 in patients with a preoperative CEA concentrations > or = 4 ng/mL vs. < 4 ng/mL, of 2.8 in patients with a preoperative CA 19-9 concentration > or = 60 U/mL vs. < CA 19-9 and of 1.8 for stage III/IV vs. stage I/II (P < 0.05). For CEA the 2-year survival rates in the group of patients with preoperative serum concentrations > or = 4 ng/mL versus < 4 ng/mL and stadium III/IV were 14% versus 29% and in stadium I/II 50% versus 83% (P < 0.05). For CA 19-9 the 2-years survival rates in the group of patients with preoperative serum concentrations > or = 60 U/mL versus < 60 U/mL and stadium III/IV were 14% versus 28% and in stadium I/II 40% versus 83% (P < 0.05). CONCLUSION: The postoperative R-classification and the tumor stage represent the best prognostic information besides the preoperative values of CA 19-9 or CEA, respectively. The predictive information provided by preoperative CEA and CA 19-9 serum levels is additional to that obtained from other factors investigated.
UNLABELLED: We studied the relevance of CEA, CA 19-9, CA 72-4 and the common classical prognostic factors (age, sex, tumor infiltration, N-classification, staging, grading and Lauren classification) in gastric carcinoma. PATIENTS AND METHODS: Survival function estimates were calculated according the method to Kaplan-Meier. The patients were separated into two groups according to preoperative marker levels. Fixing specificity at 100% for healthy people, cut off levels were calculated. Survival curve differences were assessed using the log-rank-test. Multivariate Cox proportional hazards regression analysis was performed. The mantel-Haenszel method was used to assess the 2-year survival rate of patients with gastric carcinoma and high versus low levels of tumor-associated antigens adjusted to tumor stages. The study was performed on the frozen sera (stored at -80 degrees C) of 103 patients with histologically proven gastric carcinoma. RESULTS: The tumor stage (log-rank chi-square = 55.9; P < 0.0001) represents the best prognostic factor besides preoperative values of CA 19-9 (log-rank chi-square = 13.9; P < 0.001) and CEA (log-rank chi-square = 12.2; P < 0.001). CA 72.4 shows a log-rank chi-square of 6.9 (P < 0.01). We found no statistically significant correlation between survival and sex, tumor grade and Lauren classification. The importance of different parameters in providing additional prognostic information was evaluated by multivariate analysis. Only patients after curative surgical intervention (n = 55, R0) were considered. Cox proportional hazards regression analysis yielded an adjusted relative risk of 2.4 in patients with a preoperative CEA concentrations > or = 4 ng/mL vs. < 4 ng/mL, of 2.8 in patients with a preoperative CA 19-9 concentration > or = 60 U/mL vs. < CA 19-9 and of 1.8 for stage III/IV vs. stage I/II (P < 0.05). For CEA the 2-year survival rates in the group of patients with preoperative serum concentrations > or = 4 ng/mL versus < 4 ng/mL and stadium III/IV were 14% versus 29% and in stadium I/II 50% versus 83% (P < 0.05). For CA 19-9 the 2-years survival rates in the group of patients with preoperative serum concentrations > or = 60 U/mL versus < 60 U/mL and stadium III/IV were 14% versus 28% and in stadium I/II 40% versus 83% (P < 0.05). CONCLUSION: The postoperative R-classification and the tumor stage represent the best prognostic information besides the preoperative values of CA 19-9 or CEA, respectively. The predictive information provided by preoperative CEA and CA 19-9 serum levels is additional to that obtained from other factors investigated.
Authors: Eui Soo Han; Han Hong Lee; Jun Suh Lee; Kyo Young Song; Cho Hyun Park; Hae Myung Jeon Journal: J Gastric Cancer Date: 2014-06-30 Impact factor: 3.720