分享:
分享到微信朋友圈
X
临床研究
全瘤ADC直方图参数联合影像生物标志物预测直肠腺癌神经脉管浸润的价值
王海升 袁隆 朱凯博 席华泽 刘建强 雒攀 高榕 周俊林 刘宏

Cite this article as: WANG H S, YUAN L, ZHU K B, et al. The value of combining whole-tumor ADC histogram parameters with imaging biomarkers in predicting perineural and lymphovascular invasion in rectal adenocarcinoma[J]. Chin J Magn Reson Imaging, 2025, 16(3): 63-69, 95.本文引用格式:王海升, 袁隆, 朱凯博, 等. 全瘤ADC直方图参数联合影像生物标志物预测直肠腺癌神经脉管浸润的价值[J]. 磁共振成像, 2025, 16(3): 63-69, 95 . DOI:10.12015/issn.1674-8034.2025.03.010.


[摘要] 目的 探究基于全肿瘤表观扩散系数(apparent diffusion coefficient, ADC)直方图参数联合影像生物标志物预测直肠腺癌神经脉管浸润的价值。材料与方法 回顾性分析经病理证实为直肠腺癌的102例患者的术前临床及MRI资料,根据病理结果进行分组,神经侵犯(perineural invasion, PNI)和脉管侵犯(lymphovascular invasion, LVI)中任意一项或两项阳性为PNI/LVI阳性组,两项均阴性为PNI/LVI阴性组。采用FireVoxel软件勾画感兴趣区(region of interest, ROI)后获得原发肿瘤的ADC直方图参数:ADC平均值(ADC-mean)、标准差、变异系数、熵、偏度和ADC第1、5、10、25、50、75、90、95、99百分位数(ADC-1%、ADC-5%、ADC-10%、ADC-25%、ADC-50%、ADC-75%、ADC-90%、ADC-95%、ADC-99%)。分析比较PNI/LVI阳性组和阴性组间ADC直方图参数、MRI评估壁外血管侵犯(MRI assessment extramural venous invasion, mrEMVI)状态、肿瘤位置、mrT分期、mrN分期之间的差异,通过单变量分析筛选出组间差异有统计学意义(P<0.05)的ADC直方图参数,并基于这些参数构建多因素logistic回归模型(ADC直方图模型);进一步联合单变量分析中差异有统计学意义(P<0.05)的非直方图参数进行多因素logistic回归,建立联合预测模型。利用受试者工作特征(receiver operating characteristic, ROC)曲线分析ADC直方图模型和联合模型的预测效能。采用DeLong检验比较各模型间曲线下面积(area under the curve, AUC)的差异。结果 ADC-mean、标准差、ADC-1%、ADC-75%、ADC-95%、ADC-99%、mrEMVI在直肠腺癌PNI/LVI阳性组和阴性组之间差异有统计学意义(P<0.05),在连续变量中,ADC-99%效能最高(AUC、敏感度、特异度分别为0.835、77.1%、86.6%)。由ADC-mean、标准差、ADC-1%、ADC-75%、ADC-95%、ADC-99%、mrEMVI构建的联合模型的AUC、敏感度、特异度分别为0.918、89.6%、82.9%,其诊断效能优于直方图模型(AUC为0.898)及各全肿瘤ADC直方图参数(AUC为0.670~0.835)。联合模型与ADC直方图模型的AUC值差异无统计学意义,而两模型与各直方图参数的AUC值差异均有统计学意义(P<0.05)。结论 全肿瘤ADC直方图参数及影像标志物mrEMVI可用于术前直肠腺癌神经脉管状态的预测,尤其当两者联合时,对直肠腺癌神经脉管状态的预测价值更高。
[Abstract] Objective To explore the value of combining whole-tumor apparent diffusion coefficient (ADC) histogram parameters with imaging biomarkers in predicting perineural invasion (PNI) and lymphovascular invasion (LVI) in rectal adenocarcinoma.Materials and Methods A retrospective analysis was conducted on the preoperative clinical and magnetic resonance imaging (MRI) data of 102 patients with pathologically confirmed rectal adenocarcinoma. Based on pathological results, patients were divided into two groups: the PNI/LVI-positive group (with either or both PNI and LVI positive) and the PNI/LVI-negative group (both PNI and LVI negative). Using FireVoxel software, regions of interest (ROIs) were delineated to obtain ADC histogram parameters of the primary tumor, including ADC mean (ADC-mean), standard deviation, coefficient of variation, entropy, skewness, and the 1st, 5th, 10th, 25th, 50th, 75th, 90th, 95th, and 99th percentiles of ADC (ADC-1%, ADC-5%, ADC-10%, ADC-25%, ADC-50%, ADC-75%, ADC-90%, ADC-95%, ADC-99%). Differences in ADC histogram parameters, MRI assessment extramural venous invasion (mrEMVI) status, tumor location, mrT stage, and mrN stage between the PNI/LVI-positive and negative groups were analyzed. Parameters with statistically significant differences (P < 0.05) were selected through univariate analysis and used to construct a multivariate logistic regression model (ADC histogram model). Additionally, non-histogram parameters that were also statistically significant (P < 0.05) in univariate analysis were included in a multivariate logistic regression analysis to establish a combined predictive model. The predictive performance of the ADC histogram model and the combined model was evaluated using receiver operating characteristic (ROC) curve analysis, and the DeLong test was used to compare the differences in the area under the curve (AUC) between the models.Results Significant differences were observed between the PNI/LVI-positive and negative groups in ADC-mean, standard deviation, ADC-1%, ADC-75%, ADC-95%, ADC-99%, and mrEMVI (P < 0.05). Among these continuous variables, ADC-99% had the highest diagnostic performance (AUC, sensitivity, and specificity were 0.835, 77.1%, and 86.6%, respectively). The combined model, constructed using ADC-mean, standard deviation, ADC-1%, ADC-75%, ADC-95%, ADC-99%, and mrEMVI, had an AUC, sensitivity, and specificity of 0.918, 89.6%, and 82.9%, respectively, outperforming the histogram model (AUC = 0.898) and individual whole-tumor ADC histogram parameters (AUC = 0.670 to 0.835). In addition to the combined model and the histogram model, there were statistically significant differences between the two models and the histogram parameters (P < 0.05).Conclusions Whole-tumor ADC histogram parameters and imaging biomarkers (mrEMVI) can be used to predict the neurovascular status of rectal adenocarcinoma preoperatively. The predictive value is higher when both are combined.
[关键词] 直肠腺癌;神经脉管浸润;磁共振成像;表面扩散系数;直方图;影像生物标志物
[Keywords] rectal adenocarcinoma;perineural and lymphovascular invasion;magnetic resonance imaging;apparent diffusion coefficient;histogram;imaging biomarkers

王海升    袁隆    朱凯博    席华泽    刘建强    雒攀    高榕    周俊林    刘宏 *  

兰州大学第二医院放射科,兰州 730030

通信作者:刘宏,E-mail: liu20190410@163.com

作者贡献声明:刘宏设计本研究的方案,对稿件的重要内容进行了修改,获得了国家自然科学基金和兰州大学第二医院“萃英科技创新计划”应用基础研究项目-青年项目资助;王海升起草和撰写稿件,获取、分析或解释本研究的数据;袁隆、朱凯博、席华泽、刘建强、雒攀、高榕、周俊林获取、分析或解释本研究的数据,对稿件的重要内容进行了修改;全体作者都同意发表最后的修改稿,同意对本研究的所有方面负责,确保本研究的准确性和诚信。


基金项目: 国家自然科学基金项目 82260361 兰州大学第二医院“萃英科技创新计划”应用基础研究项目-青年项目 CY2022-QN-A10
收稿日期:2024-11-27
接受日期:2025-03-07
中图分类号:R445.2  R735.37 
文献标识码:A
DOI: 10.12015/issn.1674-8034.2025.03.010
本文引用格式:王海升, 袁隆, 朱凯博, 等. 全瘤ADC直方图参数联合影像生物标志物预测直肠腺癌神经脉管浸润的价值[J]. 磁共振成像, 2025, 16(3): 63-69, 95 . DOI:10.12015/issn.1674-8034.2025.03.010.

0 引言

       结直肠癌是世界范围内第三常见的恶性肿瘤,死亡率较高[1],其中直肠癌约占三分之一[2]。研究表明,神经浸润(perineural invasion, PNI)和脉管浸润(lymphovascular invasion, LVI)阳性是直肠癌复发、转移及预后的独立危险因素[3, 4, 5]。脉管浸润(lymphovascular invasion, LVI)指肿瘤细胞侵入脉管内皮管腔或破坏淋巴管、血管壁[6],包括壁外血管侵犯(extramural vascular invasion, EMVI)、壁内血管侵犯(intramural vascular invasion, IVI)和淋巴管浸润。神经浸润(perineural invasion, PNI)指肿瘤细胞紧邻神经生长,包绕至少33%的神经周径或穿透神经膜的任一层[7]。2023版中国结直肠癌诊疗规范指出,PNI/LVI状态影响术前评估、放化疗方案的选择及手术策略的制订[8]。因此,术前精准预测直肠癌患者的PNI/LVI状态至关重要,对患者个体化治疗设计和预后评估均具有重要意义。

       MRI评估壁外血管侵犯(MRI assessment extramural venous invasion, mrEMVI)是指在MRI上观察到的肿瘤信号侵及直肠壁外的血管结构。有研究表明mrEMVI与LVI状态存在相关性[9, 10, 11]。表观扩散系数(apparent diffusion coefficient, ADC)是一种量化组织内水分子随机扩散的参数,是扩散加权成像(diffusion-weighted imaging, DWI)的定量分析结果。吕茜婷等[12, 13]证实ADC值与直肠癌的PNI/LVI状态相关。然而,传统的最大层面平均ADC值测量受限于评估维度,难以全面揭示肿瘤内部的复杂异质性。相比之下,全肿瘤ADC直方图分析凭借其多维度特性,深入解析了感兴趣容积(volume of interest, VOI)内病灶的体素分布。这不仅避免了传统感兴趣区(region of interest, ROI)勾画的主观偏差,还从肿瘤微环境角度提供了更丰富的信息,从而更精准、全面地反映肿瘤结构的异质性[14, 15, 16]。既往关于直肠癌PNI/LVI状态的直方图研究主要集中在动态对比增强(dynamic contrast-enhanced, DCE)[17]、扩散峰度成像(diffusion kurtosis imaging, DKI)[18]和体素内不相干运动DWI(intravoxel incoherent motion-DWI, IVIM-DWI)[15]等技术,基于全瘤的ADC直方图在术前直肠癌组织分级、肿瘤沉积及新辅助放化疗疗效评估等方面展现出良好的临床应用前景[19, 20, 21]。然而,目前尚无基于全肿瘤ADC直方图分析预测直肠癌术前PNI/LVI状态的研究。因此,本研究旨在探讨基于全肿瘤ADC直方图参数联合影像生物标志物mrEMVI预测直肠腺癌术前PNI/LVI状态的临床价值。

1 材料与方法

1.1 研究对象

       回顾性分析2020年1月至2023年12月在兰州大学第二医院接受手术治疗的直肠癌患者的临床、病理及MRI资料。本研究遵循《赫尔辛基宣言》并获得兰州大学第二医院医学伦理委员会批准,免除受试者知情同意,伦理批准文号:2025A-066。纳入标准:(1)经病理证实为直肠腺癌;(2)术前未接受放、化疗等;(3)具有完整的MRI资料,包括T1WI、T2WI、DWI及ADC等图像。排除标准:(1)MRI图像质量差或临床资料不全;(2)合并其他盆腔肿瘤;(3)合并远处转移;(4)既往有直肠手术史者;(5)合并严重的全身性疾病(如心脏病、肝肾功能不全等),以减少对影像学评估的干扰。

1.2 研究方法

1.2.1 MRI设备与参数

       采用Philips Healthcare Ingenia 3.0 T MRI仪(16通道体线圈)进行扫描。患者直肠MRI检查前禁食4~6 h。扫描序列及参数包括:横断面T1WI序列,TR 491 ms,TE 10 ms,层厚5 mm,FOV 300 mm×300 mm,矩阵376×334;横断面T2WI序列,TR 3000 ms,TE 80 ms,层厚3 mm,FOV 180 mm×180 mm,矩阵300×221;横断面DWI序列,TR 3000 ms,TE 60.7 ms,层厚4 mm,FOV 250 mm×250 mm,矩阵80×80,b=0、800 s/mm2,ADC图像由DWI序列重建生成。

1.2.2 临床参数分析

       我们收集了患者的以下临床资料:性别(男/女)、年龄、癌胚抗原(carcinoembryonic antigen, CEA)状态(阳性/阴性)、糖类抗原19-9(CA19-9)状态(阳性/阴性)和糖类抗原72-4(CA72-4)状态(阳性/阴性)。

1.2.3 MRI常规征象分析

       依据《中国结直肠癌诊疗规范(2023版)》[8]和《中国临床肿瘤学会(CSCO)结直肠癌诊疗指南-2024》[22],使用T2WI、DWI等图像,评估直肠癌患者的mrEMVI、mrT、mrN、肿瘤位置等状态。其中血管腔内出现肿瘤信号、血管形态不规则、血管周围肿瘤信号包绕、血管腔狭窄或闭塞、血管壁模糊征象归为mrEMVI阳性,反之归为阴性;肿瘤信号局限在直肠壁黏膜下层、固有肌层、突破固有肌层侵入周围脂肪组织、肿瘤信号与邻近器官或结构相连分别对应mrT1~mrT4;淋巴结短径≥5 mm、形态不规则、边缘模糊、T2WI信号不均匀、DWI高信号为mrN阳性,淋巴结短径<5 mm、形态规则、边缘清晰、T2WI信号均匀、DWI低信号或无异常信号为mrN阴性;根据肿瘤下缘距离肛缘距离,直肠腺癌可划分为低位(<5 cm)、中位(≥5 cm且<10 cm)和高位(≥10 cm且<15 cm)。

1.2.4 ADC直方图参数分析

       从PACS工作站以DICOM格式拷入硬盘,并将其导入Firevoxel软件(https://firevoxel.org)进行直方图参数分析。参考T2WI和DWI序列,由两名分别具有4年和12年腹部影像诊断经验的主治医师和副主任医师在ADC图像上双盲法独立确定病变的位置和范围,并沿病灶边缘逐层勾画ROI,勾画时尽量避开肠内气体、液体、残留物以及肠管外直肠系膜等。直肠癌病灶勾画完成后,软件自动分析并生成相应参数变异系数:偏度、熵、ADC平均值(ADC-mean)、标准差和ADC第1、5、10、25、50、75、90、95、99百分位数(ADC-1%、ADC-5%、ADC-10%、ADC-25%、ADC-50%、ADC-75%、ADC-90%、ADC-95%、ADC-99%)。当两位医师出现分歧时,由另外一名具有20年工作经验的影像科主任医师做出最终决定。

1.2.5 LVI和PNI状态评估

       依据第8版美国癌症联合委员会(American Joint Committee on Cancer, AJCC)结直肠癌临床和病理分类指南[23],由病理科具有15年工作经验的副主任医师完成病理标本LVI和PNI状态的评估。其中观察到癌细胞侵入淋巴管、血管或癌细胞沿神经周围扩散判断为LVI和PNI阳性,反之则为阴性。本研究中,LVI和PNI任意一项或两项阳性时为PNI/LVI阳性组,两项均阴时为PNI/LVI阴性组。

1.3 统计学分析

       使用SPSS 27.0(https://www.ibm.com/)软件完成数据统计分析。使用Shapiro-Wilk检验对计量资料进行正态性检验。对于满足正态分布的数据,使用独立样本t检验进行组间比较;对于非正态分布的数据,使用Wilcoxon秩和检验进行组间比较。使用χ2检验或Fisher精确概率法对计数资料进行组间比较。以组内相关系数(intra-class correlation coefficient, ICC)为参考,完成对两名医生数据分析一致性的评价,ICC≥0.9认为一致性极好,0.75≤ICC<0.9一致性较好,0.5≤ICC<0.75一致性一般。使用二元logistic回归筛选出与PNI/LVI状态相关的独立危险因素,采用向后逐步回归法(backward stepwise method)建立联合预测模型。随后,绘制受试者工作特征(receiver operating characteristic, ROC)曲线,分析各模型对PNI/LVI状态的预测效能。使用DeLong检验比较各模型间AUC差异。P<0.05为差异有统计学意义

2 结果

2.1 患者基线资料

       本研究最终纳入直肠腺癌患者102例,其中:男63例、女39例,PNI/LVI阳性组67例、PNI/LVI阴性组35例。图1图2分别为直肠腺癌PNI/LVI阳性和阴性患者的典型病例。患者基线资料如表1所示。患者性别、年龄、肿瘤位置、mrT、CA19-9、CA72-4在PNI/LVI阳性与阴性组间差异无统计学意义(P>0.05),而mrN、mrEMVI及CEA状态在两组间差异具有统计学意义(P<0.05)。

图1  男,52 岁,直肠腺癌神经脉管浸润。1A:横断面T2WI 示肠壁增厚,呈中等信号强度;1B:同层面DWI 呈高信号强度;1C:同层面ADC图像呈低信号强度;1D:全肿瘤ADC直方图,ADC-mean、标准差、ADC-1%、ADC-75%、ADC-95%、ADC-99%的ADC 值分别为:882.77×10-6 mm2/s、152.12×10-6 mm2/s、614.81×10-6 mm2/s、983.08×10-6 mm2/s、1 167.73×10-6 mm2/s、1 290.49×10-6 mm2/s;1E:病理结果(HE ×200)脉管扩张,脉管内见成巢异型细胞、部分成腺样排列的癌栓浸润(箭);1F:病理结果(HE ×200)神经束一端见核大深染的异型、局灶成腺样结构的癌细胞巢浸润(箭)。DWI:扩散加权成像;ADC:表观扩散系数。
Fig. 1  Male, 52-year-old, the patient with rectal adenocarcinoma showing vascular and neural invasion. 1A: Cross-sectional T2-weighted imaging shows thickening of the intestinal wall with intermediate signal intensity. 1B: The same plane on DWI shows high signal intensity. 1C: ADC imaging in the same plane shows low signal intensity. 1D: Whole-tumor ADC histogram, ADC-mean, standard deviation, and ADC values for the 1st, 75th, 95th, and 99th percentiles are 882.77 × 10-6 mm2/s, 152.12 × 10-6 mm2/s, 614.81 × 10-6 mm2/s, 983.08 × 10-6 mm2/s, 1 167.73 × 10-6 mm2/s, 1 290.49 × 10-6 mm2/s, respectively. 1E: Pathological result (HE ×200) dilated vessels with nests of atypical cells and partial glandular arrangement forming a cancer embolus (arrow) showing cancer infiltration. 1F: Pathological result (HE ×200) one end of the nerve bundle shows infiltration by cancer cell nests with large, deeply stained nuclei and focal glandular structures (arrow). DWI: diffusion-weighted imaging; ADC: apparent diffusion coefficient.
图2  女,65岁,直肠腺癌脉管神经未浸润。2A:横断面T2WI示肠壁增厚,呈中高信号强度;2B:同层面DWI呈稍高信号强度;2C:同层面ADC图像呈低信号强度;2D:全肿瘤ADC直方图,ADC-mean、标准差、ADC-1%、ADC-75%、ADC-95%、ADC-99%的ADC值分别为1 029.04×10-6 mm2/s、222.19×10-6 mm2/s、631.04×10-6 mm2/s、1 163.67×10-6 mm2/s、1 448.75×10-6 mm2/s、1 629.34×10-6 mm2/s;2E:病理结果(HE ×200)脉管内见少量红细胞(箭),未见癌浸润;2F:病理结果(HE ×200)梭形细胞排列的神经束(箭),未见癌细胞浸润。DWI:扩散加权成像;ADC:表观扩散系数。
Fig. 2  Female, 65-year-old, the patient with rectal adenocarcinoma
表1  直肠腺癌患者PNI/LVI阳性和阴性组基线资料
Tab. 1  Baseline characteristics of PNI/LVI positive and negative groups in rectal adenocarcinoma patients

2.2 全肿瘤ADC直方图参数的组间分析

       两名医生对以上直方图数据评估的一致性良好(ICC>0.75),其中ADC-mean、标准差、变异系数、熵、偏度、ADC-1%、ADC-5%、ADC-10%、ADC-25%、ADC-50%、ADC-75%、ADC-90%、ADC-95%、ADC-99%的ICC(95% CI)分别为:0.948(0.905~0.969)、0.861(0.801~0.904)、0.852(0.788~0.897)、0.853(0.768~0.905)、0.852(0.781~0.900)、0.857(0.794~0.901)、0.862(0.800~0.906)、0.920(0.878~0.947)、0.952(0.918~0.970)、0.953(0.920~0.971)、0.939(0.898~0.962)、0.932(0.884~0.959)、0.925(0.876~0.953)、0.900(0.850~0.933)。在PNI/LVI阳性组和阴性组中,ADC-mean、标准差、ADC-1%、ADC-5%、ADC-10%、ADC-25%、ADC-50%、ADC-75%、ADC-90%、ADC-95%、ADC-99%参数差异有统计学意义(P<0.05),变异系数、熵、偏度差异没有统计学意义(P>0.05)。PNI/LVI不同状态患者ADC直方图参数如表2所示。

表2  直肠腺癌患者PNI/LVI阳性和阴性组ADC直方图参数比较
Tab. 2  Comparison of ADC histogram parameters between PNI/LVI positive and negative groups in rectal adenocarcinoma patients

2.3 基于直肠腺癌PNI/LVI状态分别构建ADC直方图模型、ADC直方图联合影像生物标志物联合模型

       二元logistic回归分析结果显示(表3),在PNI/LVI状态的预测中,ADC直方图模型的独立危险因素包括ADC-mean、标准差、ADC-1%、ADC-75%、ADC-95%和ADC-99%;联合模型的独立危险因素包括ADC-mean、标准差、ADC-1%、ADC-75%、ADC-95%、ADC-99%、mrEMVI。

表3  全肿瘤ADC直方图模型、联合模型预测直肠腺癌PNI/LVI状态的二元logistic分析结果
Tab. 3  Binary logistic analysis results of whole-tumor ADC histogram model and combined model for predicting PNI/LVI status in rectal adenocarcinoma

2.4 独立危险因素参数及logistic回归模型参数对直肠癌PNI/LVI状态的预测效能

       关于直肠癌PNI/LVI状态各参数的分析中,ADC-99%的预测效能最高,AUC为0.835 [95%置信区间(confidence interval, CI):0.749~0.920],其余参数的AUC范围为0.670~0.835(表4图3),二分类变量mrEMVI基于混淆矩阵分析显示,其预测PNI/LVI状态的敏感度为82.1%,特异度为42.9%,阳性预测值(positive predictive value, PPV)为73.3%,阴性预测值(negative predictive value, NPV)为55.6%,准确率为68.6%。对于P<0.05的参数,使用logistic回归构建了ADC直方图模型和联合模型。联合模型的AUC为0.918(95% CI:0.861~0.975),高于ADC直方图模型和ADC直方图参数(图4)。DeLong检验结果显示,联合模型与ADC直方图模型间差异无统计学意义,两模型与ADC直方图参数间差异均具有统计学意义(表5)。

图3  ADC直方图各参数诊断直肠腺癌PNI/LVI状态的ROC曲线。图4 直方图模型和联合模型诊断直肠腺癌PNI/LVI状态的ROC曲线。ADC:表观扩散系数;PNI:神经侵犯;LVI:脉管侵犯;ROC:受试者工作特征;AUC:曲线下面积。
Fig. 3  ROC curves of ADC histogram parameters for diagnosing PNI/LVI status in rectal adenocarcinoma. Fig. 4 ROC curves of histogram model and combined model for diagnosing PNI/LVI status in rectal adenocarcinoma. ROC: receiver operating characteristic; ADC: apparent diffusion coefficient; PNI: perineural invasion; LVI: lymphovascular invasion; AUC: area under the curve.
表4  独立危险因素参数、直方图模型及联合模型对直肠腺癌PNI/LVI状态的预测效能
Tab. 4  Predictive performance of independent risk factors, histogram model, and combined model for PNI/LVI status in rectal adenocarcinoma
表5  PNI/LVI状态各参数与联合模型和直方图模型间AUC的DeLong检验结果(P值)
Tab. 5  DeLong test results of AUC between PNI/LVI status single parameter and combined model and histogram model (P value)

3 讨论

       本研究通过全肿瘤ADC直方图参数及影像生物标志物mrEMVI,构建了术前预测直肠腺癌PNI/LVI状态的ADC直方图模型和联合模型。结果显示,由ADC-mean、标准差、ADC-1%、ADC-75%、ADC-95%、ADC-99%、mrEMVI组成的联合模型在预测直肠腺癌术前PNI/LVI状态中的诊断效能最高(AUC为0.918)。这是国内外首次提出将ADC直方图参数与mrEMVI联合用于直肠腺癌PNI/LVI的术前无创评估,为PNI/LVI状态的术前精准分层提供了新思路。

3.1 全肿瘤ADC直方图参数术前预测直肠腺癌PNI/LVI状态的价值

       Firevoxel是一款医学图像定量分析软件,目前在神经[24]、消化[25]及泌尿[26]系统疾病的术前病理状态评估均有应用,具有体素级别分析、自动化处理和评估肿瘤异质性的优点,本研究使用该软件分析发现,无论通过单因素还是多因素分析筛选出的直方图参数,直肠腺癌PNI/LVI阳性组患者的ADC值均低于阴性组,这与既往研究结果一致[13, 27]。导致这一现象的可能原因包括:(1)PNI/LVI阳性组肿瘤侵袭性更强,细胞密度更高,细胞外间隙更小[28];(2)阳性组肿瘤微环境更为复杂[29],周围组织结构破坏更为严重,导致水分子扩散受限程度更高,最终使得直方图参数中的ADC值显著降低。既往关于子宫内膜癌、乳腺癌LVI状态[30, 31]以及胃腺癌、眼眶鳞状细胞癌PNI状态[32, 33]的研究也支持这一观点。

       在本研究中,直方图参数ADC-99%对PNI/LVI状态的术前预测表现出最高的诊断效能,这一现象可能与其反映的肿瘤生物学特性密切相关。首先,高百分位数(ADC-99%)ADC值更可能捕捉到肿瘤内部的异质性,尤其是低细胞密度区域或坏死区域,相较于ADC-mean能够更好地体现肿瘤的异质性[34]。其次,肿瘤细胞的高代谢活动(如糖酵解增强)导致局部乳酸堆积和细胞外酸化,从而促进肿瘤局部侵袭性生长和转移[35],这进一步证实了肿瘤内部的异质性。此外,直方图模型中ADC直方图的百分位数整体偏高,这与既往关于前列腺癌包膜外侵犯[26]、ⅠB~ⅡA期宫颈癌脉管间隙浸润[36]以及直肠癌组织分级[19]的研究结果一致。具体而言,低百分位数(ADC-1%)和高百分位数(ADC-75%、ADC-95%、ADC-99%)的ADC值均与直肠癌的PNI/LVI显著相关。低百分位数的ADC值较低,提示肿瘤中存在高细胞密度区域,反映了肿瘤的高侵袭性,进而增加了PNI/LVI的风险[37]。相反,高百分位数的ADC值较高,可能反映肿瘤的坏死或低细胞密度区域,提示肿瘤的异质性和快速生长特性[38],并与PNI/LVI密切相关。

3.2 mrEMVI术前预测直肠腺癌LVI状态的价值

       mrEMVI是指通过MRI检测到肠壁外血管侵犯[39]。mrEMVI阳性通常与肿瘤的侵袭性和转移风险增加相关,因此它可能是预测LVI状态的一个重要影像学指标。在本研究中,基于混淆矩阵分析显示mrEMVI单独预测PNI/LVI状态的特异度较低,但其较高的敏感度表明,mrEMVI阳性状态可以作为PNI/LVI阳性状态的一个重要提示指标。与黄伟康等[40]的研究相比,本研究准确率略低,但敏感度更高,可能与观察者经验差异(黄伟康等[40]的研究两名医生工作时间更长、经验更加丰富,准确率更高[41])、病例数略少、肿瘤分期分布(晚期肿瘤通常伴随着较高的血管侵犯和神经侵犯,这使得mrEMVI在这些患者中的敏感度较高。因此,本研究中的mrT分期偏高导致相对较高的PNI/LVI阳性患者比例,从而影响mrEMVI的敏感度[42])、检查设备的成像质量情况有关。

3.3 联合模型的临床应用价值

       本研究构建的基于ADC直方图模型在术前预测直肠癌PNI/LVI状态方面表现出较高的诊断效能(AUC为0.898)。当与影像生物标志物组建联合模型时,对PNI/LVI状态预测的AUC、敏感度和特异度分别为0.918、89.6%和82.9%。这一结果表明,联合模型能够全面反映肿瘤的侵袭性和内部结构异质性,相较于徐林生等[43]的由血清双调蛋白(AREG)水平及CT定量参数组成的联合模型对直肠癌脉管侵犯相关性及预测价值的研究,本研究联合模型的影像生物标志物mrEMVI可在结构化报告中直接获得,模型预测准确性更高,无辐射及避免增强CT对比剂所带来的过敏风险,有望成为术前无创评估直肠癌PNI/LVI状态的有效工具。

3.4 本研究的局限性

       (1)本研究为单中心回顾性研究,不可避免地存在选择偏倚风险;(2)样本量相对有限,在一定程度上影响结论的普适性;(3)本研究分析了b值为800 s/mm2时DWI生成的ADC图像,研究范围有所局限。后续仍需进一步扩大样本量、开展多中心研究,并拓展对不同b值的研究,以增强研究结论的可靠性与稳健性,从而为临床实践提供更为有力的支撑。

4 结论

       综上所述,基于全肿瘤的ADC直方图参数模型(ADC-mean、标准差、ADC-1%、ADC-75%、ADC-95%、ADC-99%)在术前预测直肠癌PNI/LVI状态中具有一定的应用价值。通过ADC直方图参数联合影像生物标志物构建的联合模型,其预测效能提升明显,有望成为术前无创准确评估直肠腺癌PNI/LVI状态的有效工具。

[1]
SUNG H, FERLAY J, SIEGEL R L, et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries[J]. CA Cancer J Clin, 2021, 71(3): 209-249. DOI: 10.3322/caac.21660.
[2]
SIEGEL R L, WAGLE N S, CERCEK A, et al. Colorectal cancer statistics, 2023[J]. CA A Cancer J Clin, 2023, 73(3): 233-254. DOI: 10.3322/caac.21772.
[3]
KIM S, HUH J W, LEE W Y, et al. Risk factors and treatment strategies for local recurrence of locally advanced rectal cancer treated with neoadjuvant chemoradiotherapy followed by total mesorectal excision[J/OL]. Eur J Surg Oncol, 2024, 50(11): 108641 [2024-11-10]. https://www.ncbi.nlm.nih.gov/pubmed/39213693/. DOI: 10.1016/j.ejso.2024.108641.
[4]
ZENG D X, YANG Z, TAN L, et al. Risk factors for lateral pelvic lymph node metastasis in patients with lower rectal cancer: a systematic review and meta-analysis[J/OL]. Front Oncol, 2023, 13: 1219608 [2024-11-10]. https://www.ncbi.nlm.nih.gov/pubmed/37746256/. DOI: 10.3389/fonc.2023.1219608.
[5]
GAO Z, CAO H, XU X, et al. Prognostic value of lymphovascular invasion in stage Ⅱ colorectal cancer patients with an inadequate examination of lymph nodes[J/OL]. World J Surg Oncol, 2021, 19(1): 125 [2024-11-10]. https://www.ncbi.nlm.nih.gov/pubmed/33866973/. DOI: 10.1186/s12957-021-02224-3.
[6]
LUCHINI C, BIBEAU F, LIGTENBERG M L, et al. ESMO recommendations on microsatellite instability testing for immunotherapy in cancer, and its relationship with PD-1/PD-L1 expression and tumour mutational burden: a systematic review-based approach[J]. Ann Oncol, 2019, 30(8): 1232-1243. DOI: 10.1093/annonc/mdz116.
[7]
LIEBIG C, AYALA G, WILKS J A, et al. Perineural invasion in cancer: a review of the literature[J]. Cancer, 2009, 115(15): 3379-3391. DOI: 10.1002/cncr.24396.
[8]
中华人民共和国国家卫生健康委员会. 中国结直肠癌诊疗规范(2023版)[J]. 中华消化外科杂志, 2023, 22(6): 667-698. DOI: 10.3760/cma.j.cn115610-20230526-00236.
National Health Commission of the People′s Republic of China. Chinese protocol of diagnosis and treatment of colorectal cancer(2023 edition)[J]. Chin J Dig Surg, 2023, 22(6): 667-698. DOI: 10.3760/cma.j.cn115610-20230526-00236.
[9]
LI J, MEI S W, ZHOU S C, et al. Perineural invasion is a prognostic factor in stage Ⅱ colorectal cancer but not a treatment indicator for traditional chemotherapy: a retrospective cohort study[J]. J Gastrointest Oncol, 2022, 13(2): 710-721. DOI: 10.21037/jgo-22-277.
[10]
杨彦松, 丁勇生, 张益飞, 等. 基于高分辨率T2WI的影像组学列线图预测直肠癌神经周围侵犯的研究[J]. 肿瘤影像学, 2022, 31(2): 130-138. DOI: 10.19732/j.cnki.2096-6210.2022.02.005.
YANG Y S, DING Y S, ZHANG Y F, et al. High resolution T2WI image-based radiomics nomogram for prediction of perineural invasion in rectal cancer[J]. Oncoradiology, 2022, 31(2): 130-138. DOI: 10.19732/j.cnki.2096-6210.2022.02.005.
[11]
CHEN S L, LI N, TANG Y, et al. The prognostic value of MRI-detected extramural vascular invasion (mrEMVI) for rectal cancer patients treated with neoadjuvant therapy: a meta-analysis[J]. Eur Radiol, 2021, 31(12): 8827-8837. DOI: 10.1007/s00330-021-07981-z.
[12]
吕茜婷, 陈勇, 李珊玫, 等. 基于磁共振扩散加权成像直肠癌ADC值与其分化程度及神经脉管侵犯相关性研究[J]. 磁共振成像, 2016, 7(12): 915-920. DOI: 10.12015/issn.1674-8034.2016.12.005.
LV Q T, CHEN Y, LI S M, et al. Correlation research of the ADC value of MR diffusion weighted imaging in differential differentiation of rectal tumor, perineural invasion and peritumor-intravascular cancer emboli[J]. Chin J Magn Reson Imag, 2016, 7(12): 915-920. DOI: 10.12015/issn.1674-8034.2016.12.005.
[13]
LIU J K, LI Q Y, TANG L L, et al. Correlations of mean and mimimum apparent diffusion coefficient values with the clinicopathological features in rectal cancer[J]. Acad Radiol, 2021, 28(Suppl 1): S105-S111. DOI: 10.1016/j.acra.2020.10.018.
[14]
MD N M, SHINYA FUJII MD P, CHIE INOUE MD P, et al. Apparent diffusion coefficient (ADC) measurement in ovarian tumor: Effect of region-of-interest methods on ADC values and diagnostic ability[J]. J Magn Reson Imag, 2016, 43(3): 720-725. DOI: 10.1002/jmri.25011.
[15]
HE R, SONG G S, FU J Y, et al. Histogram analysis based on intravoxel incoherent motion diffusion-weighted imaging for determining the perineural invasion status of rectal cancer[J]. Quant Imaging Med Surg, 2024, 14(8): 5358-5372. DOI: 10.21037/qims-23-1614.
[16]
WAN L J, PENG W J, ZOU S M, et al. Predicting perineural invasion using histogram analysis of zoomed EPI diffusion-weighted imaging in rectal cancer[J]. Abdom Radiol, 2022, 47(10): 3353-3363. DOI: 10.1007/s00261-022-03579-1.
[17]
于晓军, 陈永升, 王春艳, 等. MRI表现及ADC直方图对预测乳腺浸润性导管癌脉管浸润的价值[J]. 临床放射学杂志, 2020, 39(2): 294-298. DOI: 10.13437/j.cnki.jcr.2020.02.016.
YU X J, CHEN Y S, WANG C Y, et al. MRI performance and ADC histogram predicts lymphatic invasion in invasive ductal breast carcinoma[J]. J Clin Radiol, 2020, 39(2): 294-298. DOI: 10.13437/j.cnki.jcr.2020.02.016.
[18]
CUI Y F, YANG X T, DU X S, et al. Whole-tumour diffusion kurtosis MR imaging histogram analysis of rectal adenocarcinoma: Correlation with clinical pathologic prognostic factors[J]. Eur Radiol, 2018, 28(4): 1485-1494. DOI: 10.1007/s00330-017-5094-3.
[19]
董立杰, 张林, 杲霄源, 等. 全病灶ADC直方图分析在直肠癌组织学分级的临床应用研究[J]. 磁共振成像, 2022, 13(7): 48-54. DOI: 10.12015/issn.1674-8034.2022.07.009.
DONG L J, ZHANG L, GAO X Y, et al. Clinical application of whole-volume apparent diffusion coefficient histogram parameters of histological grading rectal adenocarcinoma[J]. Chin J Magn Reson Imag, 2022, 13(7): 48-54. DOI: 10.12015/issn.1674-8034.2022.07.009.
[20]
冯飞文, 刘原庆, 胡粟, 等. 全容积ADC直方图分析联合ADC值术前预测直肠癌肿瘤沉积的价值[J]. 磁共振成像, 2024, 15(4): 88-92. DOI: 10.12015/issn.1674-8034.2024.04.014
FENG F W, LIU Y Q, HU S, et al. The value of whole-volume ADC histogram analysis combined with ADC value in preoperatively prediction of tumor deposits in rectal cancer[J]. Chin J Magn Reson Imag, 2024, 15(4): 88-92. DOI: 10.12015/issn.1674-8034.2024.04.014
[21]
ENKHBAATAR N E, INOUE S, YAMAMURO H, et al. MR imaging with apparent diffusion coefficient histogram analysis: evaluation of locally advanced rectal cancer after chemotherapy and radiation therapy[J]. Radiology, 2018, 288(1): 129-137. DOI: 10.1148/radiol.2018171804.
[22]
中国临床肿瘤学会指南工作委员会组织. 中国临床肿瘤学会(CSCO)结直肠癌诊疗指南-2024[M]. 北京: 人民卫生出版社, 2024: 14-27.
Organized by the Guidelines Working Committee of the Chinese Society of Clinical Oncology. Guidelines of Chinese society of clinical oncology (CSCO) colorectal cancer[M]. Beijing: People's Medical Publishing House, 2024: 14-27.
[23]
WEISER M R. AJCC 8th edition: colorectal cancer[J]. Ann Surg Oncol, 2018, 25(6): 1454-1455. DOI: 10.1245/s10434-018-6462-1.
[24]
XUE C Q, ZHOU Q, ZHANG P, et al. MRI histogram analysis of tumor-infiltrating CD8+ T cell levels in patients with glioblastoma[J/OL]. Neuroimage Clin, 2023, 37: 103353 [2024-11-10]. https://pubmed.ncbi.nlm.nih.gov/36812768/. DOI: 10.1016/j.nicl.2023.103353.
[25]
ZHANG Y M, TAN H Q, HUANG B, et al. Application of a combined clinical prediction model based on enhanced T1-weighted image(T1WI) full volume histogram in peripheral nerve invasion (PNI) and lymphatic vessel invasion (LVI) in rectal cancer[J]. Abdom Radiol, 2025, 50(3): 1069-1078. DOI: 10.1007/s00261-024-04556-6.
[26]
徐鸿昊, 刘百川, 丁效蕙, 等. 表观扩散系数直方图对评估前列腺癌包膜外侵犯的附加价值[J]. 中国医学影像学杂志, 2024, 32(9): 938-944. DOI: 10.3969/j.issn.1005-5185.2024.09.014.
XU H H, LIU B C, DING X H, et al. Added value of apparent diffusion coefficient histogram in predicting extraprostatic extension of prostate cancer[J]. Chin J Med Imag, 2024, 32(9): 938-944. DOI: 10.3969/j.issn.1005-5185.2024.09.014.
[27]
YUAN Y, CHEN X L, LI Z L, et al. The application of apparent diffusion coefficients derived from intratumoral and peritumoral zones for assessing pathologic prognostic factors in rectal cancer[J]. Eur Radiol, 2022, 32(8): 5106-5118. DOI: 10.1007/s00330-022-08717-3.
[28]
ZENG Q, HONG Y L, CHENG J, et al. Quantitative study of preoperative staging of gastric cancer using intravoxel incoherent motion diffusion-weighted imaging as a potential clinical index[J/OL]. Eur J Radiol, 2021, 141: 109627 [2024-11-10]. https://pubmed.ncbi.nlm.nih.gov/34126429/. DOI: 10.1016/j.ejrad.2021.109627.
[29]
CHAUDHURI O, COOPER-WHITE J, JANMEY P A, et al. Effects of extracellular matrix viscoelasticity on cellular behaviour[J]. Nature, 2020, 584(7822): 535-546. DOI: 10.1038/s41586-020-2612-2.
[30]
漆万银, 程勇, 杨述根. 扩散加权成像评估子宫内膜癌淋巴血管间隙侵犯的价值[J]. 放射学实践, 2021, 36(2): 222-226. DOI: 10.13609/j.cnki.1000-0313.2021.02.014.
QI W Y, CHENG Y, YANG S G. Value of diffusion-weighted imaging in evaluating lymphovascular space invasion of endometrial cancer[J]. Radiol Pract, 2021, 36(2): 222-226. DOI: 10.13609/j.cnki.1000-0313.2021.02.014.
[31]
DURANDO M, GENNARO L, CHO G Y, et al. Quantitative apparent diffusion coefficient measurement obtained by 3.0Tesla MRI as a potential noninvasive marker of tumor aggressiveness in breast cancer[J]. Eur J Radiol, 2016, 85(9): 1651-1658. DOI: 10.1016/j.ejrad.2016.06.019.
[32]
LI J, WANG Y, ZHANG H K, et al. The value of intravoxel incoherent motion diffusion-weighted imaging in predicting perineural invasion for resectable gastric cancer: a prospective study[J/OL]. Clin Radiol, 2024, 79(1): e65-e72 [2024-11-10]. https://www.ncbi.nlm.nih.gov/pubmed/37833144/. DOI: 10.1016/j.crad.2023.09.014.
[33]
ABDELAZIZ T T, ABDEL KHALEK ABDEL RAZEK A. Magnetic resonance imaging of perineural spread of head and neck cancer[J]. Magn Reson Imaging Clin N Am, 2022, 30(1): 95-108. DOI: 10.1016/j.mric.2021.06.017.
[34]
LIU H L, ZONG M, WEI H, et al. Added value of histogram analysis of apparent diffusion coefficient maps for differentiating triple-negative breast cancer from other subtypes of breast cancer on standard MRI[J]. Cancer Manag Res, 2019, 11: 8239-8247. DOI: 10.2147/CMAR.S210583.
[35]
ESTRELLA V, CHEN T A, LLOYD M, et al. Acidity generated by the tumor microenvironment drives local invasion[J]. Cancer Res, 2013, 73(5): 1524-1535. DOI: 10.1158/0008-5472.CAN-12-2796.
[36]
CHENG J M, LUO W X, TAN B G, et al. Whole-tumor histogram analysis of apparent diffusion coefficients for predicting lymphovascular space invasion in stage IB-IIA cervical cancer[J/OL]. Front Oncol, 2023, 13: 1206659 [2024-11-10]. https://www.ncbi.nlm.nih.gov/pubmed/37404753/. DOI: 10.3389/fonc.2023.1206659.
[37]
CURVO-SEMEDO L, LAMBREGTS D M J, MAAS M, et al. Diffusion-weighted MRI in rectal cancer: apparent diffusion coefficient as a potential noninvasive marker of tumor aggressiveness[J]. J Magn Reson Imaging, 2012, 35(6): 1365-1371. DOI: 10.1002/jmri.23589.
[38]
HEO S H, SHIN S S, KIM J W, et al. Pre-treatment diffusion-weighted MR imaging for predicting tumor recurrence in uterine cervical cancer treated with concurrent chemoradiation: value of histogram analysis of apparent diffusion coefficients[J]. Korean J Radiol, 2013, 14(4): 616-625. DOI: 10.3348/kjr.2013.14.4.616.
[39]
FOURNIER F R, ALI K MOTAMEDI M, BROWN C J, et al. Oncologic outcomes associated with MRI-detected extramural venous invasion (mrEMVI) in rectal cancer: a systematic review and meta-analysis[J]. Ann Surg, 2022, 275(2): 303-314. DOI: 10.1097/SLA.0000000000004636.
[40]
黄伟康, 卿勇, 刘岘, 等. 探讨直肠腺癌壁外血管侵犯磁共振评分与ADC诊断EMVI的价值[J]. 临床放射学杂志, 2023, 42(7): 1149-1153. DOI: 10.13437/j.cnki.jcr.2023.07.018.
HUANG W K, QING Y, LIU X, et al. Exploring the value of extramural vascular invasion score of MRI and apparent diffusion coefficient in diagnosing EMVI of rectal adenocarcinoma[J]. J Clin Radiol, 2023, 42(7): 1149-1153. DOI: 10.13437/j.cnki.jcr.2023.07.018.
[41]
JHAVERI K S, HOSSEINI-NIK H, THIPPHAVONG S, et al. MRI detection of extramural venous invasion in rectal cancer: correlation with histopathology using elastin stain[J]. AJR Am J Roentgenol, 2016, 206(4): 747-755. DOI: 10.2214/AJR.15.15568.
[42]
BAE J S, KIM S H, HUR B Y, et al. Prognostic value of MRI in assessing extramural venous invasion in rectal cancer: multi-readers' diagnostic performance[J]. Eur Radiol, 2019, 29(8): 4379-4388. DOI: 10.1007/s00330-018-5926-9.
[43]
徐林生, 徐翔, 杨金凤, 等. 血清AREG水平及CT定量参数与直肠癌脉管侵犯、淋巴结转移的相关性及预测价值研究[J]. 放射学实践, 2024, 39(5): 652-657. DOI: 10.13609/j.cnki.1000-0313.2024.05.015.
XU L S, XU X, YANG J F, et al. Correlation and predictive value of serum AREG levels and CT quantitative parameters with vascular invasion and lymph node metastasis in rectal cancer[J]. Radiol Pract, 2024, 39(5): 652-657. DOI: 10.13609/j.cnki.1000-0313.2024.05.015.

上一篇 联合临床指标和MR征象的列线图预测肝细胞癌肿瘤包绕型血管
下一篇 基于术前多参数MRI的深度迁移学习预测子宫内膜癌淋巴脉管间隙浸润
  
诚聘英才 | 广告合作 | 免责声明 | 版权声明
联系电话:010-67113815
京ICP备19028836号-2