蛋白芯片技术检测肿瘤标志物对乳腺癌的诊断价值

目的探讨应用蛋白芯片技术检测肿瘤标志物对乳腺癌的诊断价值。方法采用多肿瘤标志物蛋白芯片诊断系统，检测90例乳腺癌患者、60例乳腺良性病变者和100名健康体检者12种血清肿瘤标志物，采用SPSS13．0统计软件进行数据分析。结果乳腺癌组CEA、CA125、FER、CA153水平，均显著高于良性病变组和对照组（P〈0．01）：良性捐变组CEA、CA125、FER、CA153水平与对照组比较，差异无显著性意义（P〉0．05）；而CA199、NSE、CA242、HCG、AFP、f-PSA、PSA、HGH在三三组间比较均未见统计学意义（P〉0．05）；乳腺癌组与对照组的logistic回归方程P=1／1＋e^-（-1.29＋2.25×cea＋3.12×CA125＋10.59×CA153）；ROC曲线中，CA125、CA153、CEA、FER联合检测的AUC大于各项肿瘤标志物单项检测的AUC（P=0．000）。结论蛋白芯片技术检测肿瘤标志物对乳腺癌的有较高的临床诊断价值；血清CA125、CA153、CEA、FER是乳腺癌辅助诊断的理想指标，其联合检测有利于提高乳腺癌的诊断率；CA199，NSE，CA242，HCG，AFP，f-PSA，PSA，HGH这些指标对乳腺癌诊断价值较低；作为1种统计手段，logistic回归可改善诊断的灵敏度和特异性。

The Diagnostic Value of Tumor Markers by Protein Chip Technology for Breast Cancer

Objective To evaluate the diagnostic value of detection of tumor markers by protein chip technology for breast cancer. Methods 12 tumor markers were detected by the multi - tumor marker protein biochip diagnostic system in 90 patients with breast cancer,60 patients with benign diseases,and 100 healthy controls. Results The serum levels of CEA, FER, CA125,CAI53 in the patients with breast cancer were significantly higher than that in heahhy controls and the patients with benign breast diseases （P 〈 0.01 ） ;The serum levels of the above markers were not significantly different between the healthy controis and the patients with benign diseases （ P 〉 0.05 ） ;The serum levels of CA199, NSE, CA242, HCG, AFP, f - PSA, PSA and HGH had no significant differences between the patients with breast cancer, the patients with benign breast disease and control groups （ all P 〉 0 . 0 5 ） ; The logistic regression equation in the patients with breast cancer and control groups was P=1/1＋e^-（-1.29＋2.25×cea＋3.12×CA125＋10.59×CA153） ; In cancer - control group, the AUC of combination of CEA, FER, CA125 and CA153 was larger than any AUC of CEA, FER,CAI25 or CA153 alone （ P =0. 000）. Conclusion Combined detection of CEA,FER,CA125 and CA153 by protein bioehip technique might be hclptul in the diagnosis of hreast cancer; Where as CA199, NSE, CA242, HCG, AFP,f- PSA, PSA and HGH have limited value; As an advanced statistical method,logistic regression can improve the diagnostic sensitivity and specificity.