Prognostic significant of overexpressed phosphodiesterase 4B in patients with diffuse large B-cell lymphoma
-
摘要: 目的 探讨磷酸二酯酶4B(phosphdiesterase 4B,PDE4B)基因过表达对弥漫大B细胞淋巴瘤(diffuse large B-cell lymphoma,DLBCL)患者预后的影响,并分析其预后价值。方法 从GEO(gene expression omnibus)数据库获得320 例DLBCL患者的基因芯片数据,比较PDE4B基因过表达与正常表达患者生存率的差异。构建单变量国际预后指数(international prognostic index,IPI)、PDE4B和多变量(IPI+PDE4B)Cox比例风险模型,并用时点/动态受试者工作曲线(receiver operating characteristic curves,ROC)法分析模型的预测能力随时间的变化趋势,一致性统计量C用于评价和比较各模型的预测准确度。结果 PDE4B基因过表达的患者死亡的风险是正常表达的1.431倍(HR=1.431,95% CI:1.013~2.002,P=0.042)。多变量模型的C统计量与单变量模型相比,差异有统计学意义(IPI vs. IPI+PDE4B:Z=15.0,P<0.001;PDE4B vs. IPI+PDE4B:Z=53.7,P<0.001)。结论 PDE4B基因过表达是DLBCL患者预后的危险因素,结合IPI和PDE4B基因能提高DLBCL的预后准确性。
-
关键词:
- 环核苷酸磷酸二酯酶类,4型 /
- 淋巴瘤,B细胞 /
- 基因表达 /
- 预后
Abstract: Objective This study used to investigate the association between overexpressed phosphdiesterase 4B (PDE4B) and the prognosis of diffuse large B-cell lymphoma (DLBCL) and evaluate its predictive values. Methods Micro-array data of 320 patients with DLBCL was downloaded from gene expression omnibus (GEO) data set. The statistically significant test of overall survival (OS) between high and normal PDE4B expression level was made by Log-rank test. Univariate (international prognostic index, IPI or PDE4B) and multivariate (IPI+PDE4B) Cox proportional hazards model were constructed and time-dependent receiver operating characteristic curves method was used to analysis the tendency of model's predictive power over time. The overall prediction accuracy of each model was evaluated and compared with the C index. Results The risk of death in DLBCL patients with overexpressed PDE4B was 1.431 times (HR=1.431,95% CI:1.013-2.002, P=0.042) that of normal expression. C index was statistically significant difference (IPI vs. IPI+PDE4B:Z=15.0, P<0.001;PDE4B vs. IPI+PDE4B:Z=53.7, P<0.001) between multivariate and univariate Cox regression model. Conclusions Overexpressed PDE4B is a hazard for DLBCL. Prognostic accuracy of DLBCL may be improved by combining IPI and PDE4B.
点击查看大图
计量
- 文章访问数: 472
- HTML全文浏览量: 63
- PDF下载量: 16
- 被引次数: 0