目的 旨在评估磁共振功能成像分析肿瘤异质性在直肠癌放化疗(CRT)疗效评估中的作用。 方法 2020 年 1 月至 2021 年 11 月期间,选择在惠州市中心人民医院接受 CRT 前磁共振功能成像检查以评估 60 例局部晚期直肠癌(LARC)患者作 为研究对象。 从整体肿瘤的直方图中获得表观扩散系数(ADC)、体素内不相干运动(Dslow、Dfast 和 f)和拉伸指数模型[分布 扩散系数(DDC)和 α]。 比较病理完全缓解(pCR)和非 pCR 患者 CRT 前的直方图指标和预后因素。 构建受试者操作特征曲 线(ROC 曲线)以评估这些参数的诊断性能。 结果 CRT 治疗后,13 例(21. 7%)患者归类为 pCR 组,而其他 47 例患者被归类 为非 pCR 组(78. 3%)。 与 pCR 组相比,非 pCR 组在肿瘤体积、TRG 和 pT 分期方面存在显著差异(P<0. 05)。 pCR 组的 ADC 均值、标准差以及第 10、25、50、75、90 和 95 个百分点显著低于非 pCR 组(P<0. 05)。 pCR 组 DDC 的平均值、偏度、峰度和第 25、50、75、90 和 95 百分位数低于非 pCR 组(P< 0. 05)。 在 ADC 中,其平均值(AUC = 0. 890,P < 0. 001,95%CI = 0. 806 ~ 0. 974)的诊断性能最高。 在 DDC 中,其平均值(AUC = 0. 866,P<0. 001,95%CI = 0. 771 ~ 0. 960)的 AUC 值最高。 ADC 平均 值与肿瘤体积结合预测 pCR 具有最高的诊断性能,其 AUC 为 0. 932,敏感度为 100%,特异度为 81%。 结论 使用整体肿瘤的 ADC 平均值与肿瘤体积相结合能有效预测 LARC 对 CRT 的反应。

Objective Overall tumor assessment such as histogram analysis was a common means to obtain tumor heterogeneity information that may affect the treatment response. The purpose of this study was to evaluate the role of functional magnetic resonance imaging MRI in analyzing tumor heterogeneity in evaluating the efficacy of radiotherapy and chemotherapy CRT for rectal cancer. Methods From January 2020 to November 2021 60 patients with locally advanced rectal cancer LARC who underwent MRI examination before CRT in our hospital were selected as the research subjects. The apparent diffusion coefficient ADC intra-voxel incoherent motion Dslow Dfast and F and stretch index model DDC and α were obtained from the histogram of the whole tumor. Histogram indexes and prognostic factors before CRT were compared between patients with complete pathological remission pCR and those without pCR. The receiver operating characteristic curve ROC curve was constructed to evaluate the diagnostic performance of these parameters. Results After CRT treatment 13 patients 21. 7% were classified as pCR group while the other 47 patients were classified as non-pCR group 78. 3% . Compared with the pCR group the non-pCR group had significant differences in tumor volume TRG and pT stages P < 0. 05 . The mean standard deviation and the 10th 25th 50th 75th 90th and 95th percentile of ADC in pCR group were significantly lower than those in non-pCR group P<0. 05 . The mean skewness kurtosis and 25th 50th 75th 90th and 95th percentile of DDC in pCR group were lower than those in non-pCR group P<0. 05 . In ADC its average value AUC = 0. 890 P<0. 001 95%CI = 0. 806-0. 974 had the highest diagnostic performance. In DDC its average AUC = 0. 866 P<0. 001 95%CI = 0. 771-0. 960 has the highest AUC value. Average ADC combined with tumor volume had the highest diagnostic performance in predicting pCR with AUC of 0. 932 sensitivity of 100% and specificity of 81%. Conclusion The whole-tumor ADC mean combined with the tumor volume can effectively predict the response of LARC to CRT.