Chen ZhiGuo, Li Ming, Yuan YanHua, Wang Qi, Yan Li, Gu Jin
Key Laboratory of Carcinogenesis and Translational Research (Ministry of Education), GI Surgery Department, Peking University School of Oncology, Beijing Cancer Hospital & Institute, Beijing, People's Republic of China.
Dis Colon Rectum. 2010 Jan;53(1):31-8. doi: 10.1007/DCR.0b013e3181bdca3a.
Accumulating evidence suggests that cancer/testis antigens may serve as indicators of tumor malignant phenotype. The purpose of this study is to evaluate cancer/testis antigen genes in predicting metastasis of colorectal cancer to the liver.
The expression levels of 25 cancer/testis antigen genes were determined by reverse-transcription polymerase chain reaction in 288 colorectal cancer tissue samples from the primary tumor or liver metastasis. Pearson chi2 and multiple logistic regression analyses were performed to assess the association between risk factors and probability of liver metastasis of colorectal cancer.
No significant difference was detected between the primary tumor and liver metastasis in expression pattern of cancer/testis antigen genes in colorectal cancer tissue samples. However, 3 cancer/testis antigen genes (PAGE4, SCP-1, and SPANX) and 3 clinicopathologic parameters (lymph node involvement, vessel cancer embolus, and tumor invasion depth) correlated significantly with liver metastasis of colorectal cancer (P < .05). A logistic regression model was constructed for prediction of liver metastasis based on a panel consisting of PAGE4, lymph node involvement, and presence or absence of vessel cancer embolus. The predicted risk of liver metastasis based on the panel was consistent with the actual risk observed. The probability of developing liver metastasis as estimated by the panel was 86.9% when all 3 factors were positive, representing an up to 20% improvement in the prediction level compared with the classic methods of lymph node involvement and vessel cancer embolus.
A new predictive panel including PAGE4 expression may help predict liver metastasis of colorectal cancer.
越来越多的证据表明，癌胚抗原可能是肿瘤恶性表型的指标。本研究旨在评估癌胚抗原基因在预测结直肠癌肝转移中的作用。
采用逆转录聚合酶链反应检测288例原发性肿瘤或肝转移灶的结直肠癌组织样本中25种癌胚抗原基因的表达水平。采用Pearson卡方检验和多因素logistic回归分析评估危险因素与结直肠癌肝转移概率之间的关系。
结直肠癌组织样本中，原发性肿瘤和肝转移灶的癌胚抗原基因表达模式无显著差异。然而，3种癌胚抗原基因（PAGE4、SCP-1和SPANX）和3种临床病理参数（淋巴结转移、血管癌栓和肿瘤浸润深度）与结直肠癌肝转移显著相关（P < 0.05）。基于PAGE4、淋巴结转移和血管癌栓的有无构建了一个logistic回归模型来预测肝转移。该模型预测的肝转移风险与实际观察到的风险一致。当所有3个因素均为阳性时，该模型估计发生肝转移的概率为86.9%，与经典的淋巴结转移和血管癌栓评估方法相比，预测水平提高了20%。
包括PAGE4表达的新预测模型可能有助于预测结直肠癌的肝转移。
