多种癌基因联合表达判断非小细胞肺癌预后的研究

目的　探讨多种癌基因蛋白表达与非小细胞肺癌 (NSCLC)预后的关系。方法　 1981～ 1994年 ,15 8例符合条件患者进入本研究。免疫组化方法检测Pan -ras,cmyc ,c -erbB2 ,EGFR和p 5 3等基因蛋白表达。Kaplan -Meier法计算生存率、局控率及远处转移率 ,Log-Rank法检验差异性 ,多因素分析为Cox模型。用记分方法探讨多种癌基因表达与预后关系。结果　全组 5年生存率、局控和远处转移率分别为 4 4 %、6 3%和 4 0 %。单个癌基因单因素分析 :Pan -ras、p 5 3与NSCLC远处转移有关 (P=0 .0 39,0 .0 6 )。多因素分析 :5个癌基因与NSCLC三项观察指标均无关。采用记分方法综合多种癌基因蛋白表达分析可提高其对NSCLC预后预测意义。结论　 1.多种癌基因综合分析较单一癌基因对NSCLC预后判断更具价值。 2 .癌基因对预后判断主要是对远处转移预测

The study of prognosis in nonsmall cell lung cancer by the expression of multiple oncogenes

Objective To predict the prognosis of non-small cell lung cancer by measuring multiple oncoprotein expression. Methods From 1981 to 1994, 158 patients with non-small cell lung cancer entered this study. The protein expression of pan-ras, c-myc, c-erb B2, EGFR and p 53 were measured by immunohitochemistry assay. The rates of survival, local control and distant metastasis were estimated by Kaplan-Meier method with Log-rank test to compare the difference among subgroups. Cox model was used for multivariate analyses. Results The overall five year actuarial survival?local control and distant metastasis rates were 44%,63% and 40%,respectively. Univariate analysis showed that the patients with either pan-ras or p 53 protein expression developed more distant metastases than those with negative ( P =0.039,0.06). However, there were no significant differences in survival?local control and distant metastasis by Cox analyses for each oncogene protein expression. The prediction had been improved by using the sum of positive oncoprotein stains. Especially, when all patients were divided into six sub groups according to the protein expression statues , there was a significant trend. As the sum of positive stains was risen distant metastasis rate was increasing in distant metastasis among groups. Conclusion 1. It would be more accurate to predict the prognosis of NSCLC by comprehensive multi oncogenes study than single oncogene study. 2. Oncoprotein expression could predict the potential of the distant metastases, not the local control.