P2.01-65 Assessment of Individual and Combined of Five Serum Tumor Markers for Lung Cancer

Abstract

Tumor markers were often used to help in diagnosis lung cancer, The best known tumor markers in lung cancer were SCC, CEA,CYFRA21-1,NSE,and ProGRP. The roles of these tumor markers in diagnosis remain to be determined. This retrospective study conducted between January 2015 and April 2017 at the First Affiliated Hospital of Wenzhou Medical University in China. A total of 2569 patients sought medical attention for a pulmonary mass presence to either confirm or exclude the disease.SCC,CEA,Cyfra21-1,NSE and ProGRP were examied with suspicious lung cancer. SPSS 20.0 used for all statistical analyses. Finally,1135 patients were excluded,1434 patients were selected.Lung cancer was established in 1095 patients(737 males and 358 females), average age of（65.3±10.8）years and excluded in 339 patients (229 males, 117 females),average age of (55.7±13.4).Abnormal CEA were represented in 65.6% of adenocarcinoma. SCC-Ag and CYFRA21-1 significantly higher in 59.8% and 79.0% of squamous cell carcinoma respectively. Abnormal values of NSE and ProGRP were associated with 93.3% and 80.2% of SCLC.CEA and CYFRA 21-1 recorded sensitivity of 65.6% and 65.3% respectively in adenocarcinoma. In squamous cell carcinoma, SCC-Ag and CYFRA21-1 displayed sensitivity of 59.8% and 79% respectively. In SCLC, NSE and ProGRP showed sensitivities of 93.5% and 80.8%.The combination of different tumor markers yielded sensitivity values (>90%) and specificity values (>57%). The combination of five tumor markers recorded the highest sensitivity of 95.6%. The CYFRA 21-1 combine with CEA recorded the highest specificity 79.9%. ROC curves were plotted to explore the diagnostic role of these tumor markers individually and in combination. CYFRA 21-1 with AUC of 0.84 (95%CI 0.820-0.863) performed better than CEA and the other remaining individual tumor markers. The combination of five tumor markers, with an AUC of 0.904 (5%CI 0.888-0.920) is the highest. shown as figure 1. Combined TMs can potentially improve the specificity and sensitivity values in LC diagnosis