PURPOSE To explore whether volumetric measurements of 3D-CUBE sequences based on baseline and early treatment time can predict neoadjuvent chemotherapy (NCT) efficacy of locally advanced rectal cancer (LARC). MATERIAL AND METHOD 73 patients with LARC were enrolled from February 2014 to January 2018. All patients underwent MRIs during the baseline period before NCT (BL-NCT) and the first month of NCT (FM-NCT), and tumor volume (TV) was measured using 3D-CUBE, and tumor volume reduction (TVR) and tumor volume reduction rate (TVRR) were calculated. In addition, tumor invasion depth, tumor maximal length, range of tumor involvement in the circumference of intestinal lumen and distance from inferior part of tumor to the anal verge were measured using baseline high-spatial-resolution T2-weighted MRIs. All patients were categorized into sensitive and insensitive groups based on post-surgical pathology after completion of the full courses of NCT. The receiver operating characteristic (ROC) curve was used to analyze the value of different MRI parameters in predicting efficacy of NCT. RESULTS Statistically significant differences in TV of BL-NCT, TVR and TVRR from BL-NCT to FM-NCT were detected between sensitive and insensitive groups (P < 0.05, respectively). The areas under the curves (AUC) of ROC of TVR and TVRR in predicting efficacy of NCT (0.890 [95% CI, 0.795∼0.951], 0.839 [95% CI, 0.735∼0.915]) were significantly better than that of TV (0.660 [95% CI, 0.540∼0.767]) (P < 0.05, respectively). CONCLUSION Reconstruction of 3D-CUBE volume in the first month of NCT is necessary, and both TVR and TVRR can be used as early predictors of NCT efficacy.

中文翻译：

---

局部晚期直肠癌的基线和早期3D-CUBE体积重建可预测新辅助化疗后的肿瘤反应。

目的探讨基于基线和早期治疗时间的3D-CUBE序列的体积测量是否可以预测局部晚期直肠癌（LARC）的新辅助化疗（NCT）疗效。材料与方法自2014年2月至2018年1月，纳入73例LARC患者。所有患者均在NCT之前的基线期（BL-NCT）和NCT的第一个月（FM-NCT）进行了MRI，而肿瘤体积（TV）为使用3D-CUBE测量肿瘤体积缩小率（TVR）和肿瘤体积缩小率（TVRR）。此外，使用基线高空间分辨率T2加权MRI测量了肿瘤浸润深度，肿瘤最大长度，肿瘤在肠腔周围的受累范围以及从肿瘤下半部到肛门边缘的距离。完成全部NCT课程后，根据术后病理将所有患者分为敏感和不敏感人群。接收器工作特征曲线（ROC）用于分析不同MRI参数在预测NCT疗效中的价值。结果在敏感组和不敏感组之间，从BL-NCT到FM-NCT，BL-NCT，TVR和TVRR的电视差异具有统计学意义（分别为P <0.05）。TVR和TVRR的ROC曲线下面积（AUC）在预测NCT疗效方面（0.890 [95％CI，0.795〜0.951]，0.839 [95％CI，0.735〜0.915]）明显优于TV（ 0.660 [95％CI，0.540-0.767]）（分别为P <0.05）。结论必须在NCT的第一个月重建3D-CUBE体积，