计算机断层扫描纹理分析技术评估索拉非尼治疗肾癌肺转移的疗效

发布时间：2018-05-22 11:19

本文选题：肾细胞癌 + 索拉非尼　；参考：《上海医学》2017年09期

【摘要】：目的应用CT平扫图像纹理特征评估抗肿瘤血管生成药物索拉非尼治疗肾癌肺转移的疗效。方法选择87例接受索拉非尼治疗的肾癌肺转移患者,所有患者均于治疗前和2个治疗周期后行胸部CT平扫检查,并采用实体肿瘤疗效评价标准(RECIST)评估临床效果。按入组时间将患者分为分析样本组(2008年6月—2009年1月入组,60例)和验证组(2009年2月—2011年1月入组,27例),再根据无进展生存期(PFS)将两组患者分别分为PFS≥1年和PFS1年亚组。在分析样本组中,通过ROC曲线获得统计模型中的界断值,采用逐步选择的方法进行判别分析,评价病灶直径、CT值、熵和均匀性的变化,获取分析样本组用于评估疗效的参数界断值和统计模型,并将其用于验证组,应用RECIST、各影像参数和判别分析建立卡普兰-迈耶生存曲线。结果 2个治疗周期后,分析样本组两亚组除病灶直径均较治疗前显著增大(P值分别为0.03和0.04)外,CT值,以及不使用过滤器和采用不同过滤系数时的熵和均匀性与治疗前的差异均无统计学意义(P值均0.05)。验证组两亚组的病灶直径、CT值,以及不使用过滤器和采用不同过滤系数时的熵和均匀性与治疗前的差异均无统计学意义(P值均0.05)。卡普兰-迈耶生存曲线结果显示,分析样本组和验证组采用判别分析法评估肿瘤治疗疗效的效果(P值分别为0.02、0.04)均优于单独使用其他参数;验证组中,RECIST的P值为0.21,CT值变化百分比的P值为0.49,在不使用过滤器和过滤系数为1.0、1.5、2.0、2.5时熵变化值百分比的P值分别为0.47、0.89、0.72、0.73和0.58,在不使用过滤器和过滤系数为1.0、1.5、2.0、2.5时均匀性变化百分比的P值分别为0.53、0.72、0.51、0.39和0.16。结论与单独使用RECIST相比,结合治疗前后影像参数包括病灶直径、CT值、熵、均匀性等变化的判别分析,更有助于评估抗血管新生药物治疗肾癌肺转移的疗效。
[Abstract]:Objective to evaluate the effect of CT plain scan image texture on the treatment of lung metastasis of renal carcinoma with anti tumor angiogenesis drug Sola Fini. Methods 87 cases of renal carcinoma with lung metastases received Sola Fini were selected. All patients underwent chest CT scan before and after the treatment, and the evaluation criteria of solid tumor effect were adopted (RECIST The patients were divided into analysis sample group (June 2008 - 2009 1 month group, 60 cases) and verification group (February 2009 - 2011 1 month group, 27 cases), and then the two groups were divided into PFS > 1 years and PFS1 subgroups according to the progression free survival period (PFS). The statistical model was obtained by the ROC curve in the analysis sample group. By discriminant analysis in the stepwise selection method, the changes of the diameter, CT value, entropy and uniformity of the lesion were evaluated, and the parameter boundary value and the statistical model of the analysis sample group used to evaluate the effect were obtained and used in the validation group. The Kaplan Meyer survival curve was established by using the RECIST, the image parameters and the discriminant analysis. The results were 2. After the treatment period, the two subgroups of the two subgroup were significantly larger than before the treatment (0.03 and 0.04 respectively), and there was no significant difference between the CT value, the entropy and the uniformity of the group without the filter and the use of the different filter coefficients (P value, 0.05). The lesion diameter, the CT value, and the number of the two subgroups of the group two were verified and did not. The difference of entropy and uniformity between the filter and the different filter coefficients was not statistically significant (P value was 0.05). The results of Kaplan Mayer survival curve showed that the results of discriminant analysis in the analysis group and the validation group (P value 0.02,0.04) were better than the other parameters. In the certificate group, the P value of RECIST is 0.21 and the P value of the CT value change percentage is 0.49. The P value of the entropy change percentage is 0.47,0.89,0.72,0.73 and 0.58 respectively when the filter and filter coefficient are 1.0,1.5,2.0,2.5, respectively, and the P value of the average evenness change when the filter and the filter coefficient is 1.0,1.5,2.0,2.5 is 0.53,0.72,0.51, respectively. 0.39 and 0.16. conclusions compared with the use of RECIST alone, the imaging parameters before and after treatment include the discriminant analysis of the changes of lesion diameter, CT value, entropy and uniformity, which is more helpful to evaluate the efficacy of antiangiogenic drugs in the treatment of renal carcinoma and lung metastasis.
【作者单位】：上海交通大学医学院附属仁济医院南院放射科;上海交通大学医学院附属仁济医院放射科;
【基金】：上海交通大学医工交叉研究项目(YG2014MS50) 上海市卫生与计划生育委员会科研项目(201540231) 上海自然科学基金项目(16ZR1420700)资助
【分类号】：R730.44;R735.2

【相似文献】