The Value of Model Based on Radiomics Features of T2-weighted Imaging and Clinical Feature in Diagnosing the Depth of Stromal Invasion of Cervical Squamous Cell Carcinoma
-
摘要:目的 初步探讨基于T2加权成像(T2-weighted imaging, T2WI)的影像组学特征联合患者临床特征构建的模型对早期宫颈鳞状细胞癌深间质浸润(deep stromal invasion, DSI)的诊断价值。方法 回顾性纳入2017年1月至2021年2月在北京协和医院行根治性子宫切除术的早期宫颈鳞状细胞癌患者，并按8∶2的比例随机分为训练集和验证集。收集训练集患者的术前临床特征和矢状位T2WI图像影像组学特征资料，经筛选、特征降维后，采用Logistic回归分析法建立早期宫颈癌DSI诊断模型，包括临床特征模型、影像组学模型和临床-影像组学模型。基于验证集数据，采用受试者工作特征(receiver operating characteristic, ROC)曲线对上述模型的性能进行验证。结果 共168例符合纳入和排除标准的早期宫颈鳞状细胞癌患者入选本研究。其中训练集135例，验证集33例；经组织病理学证实为浅间质浸润的患者72例，DSI患者96例。共筛选出患者年龄、术前鳞状细胞癌抗原水平、国际妇产科联盟分期3个临床特征和4个影像组学特征用于模型构建。ROC曲线分析显示，临床特征模型、影像组学模型和临床-影像组学模型诊断早期宫颈鳞状细胞癌DSI的曲线下面积分别为0.797(95% CI: 0.623~0.971)、0.793(95% CI: 0.633~0.954)和0.820(95% CI: 0.665~0.974)，且以临床-影像组学模型的诊断效能最高，其灵敏度、特异度和准确度分别为85.7%(95% CI: 49.8%~100%)、73.7%(95% CI: 57.9%~100%)和78.8%(95% CI: 69.7%~93.9%)。结论 基于T2WI图像的影像组学特征联合临床特征构建的临床-影像组学模型可作为一种无创的术前检查手段高效判断早期宫颈鳞状细胞癌间质浸润深度。Abstract:Objective To investigate the prediction value of a clinical-radiomics model based on T2- weighted imaging (T2WI) and clinical features for diagnosing deep stromal invasion (DSI) in patients with early-stage cervical squamous cell carcinoma.Methods Patients with early-stage cervical squamous cell carcinoma that underwent radical hysterectomy in Peking Union Medical College Hospital from January 2017 to February 2021 were retrospectively included and randomly divided into the training set and the validation set with the the ratio of 8∶2. The preoperative clinical features and the radiomics features of sagittal T2WI images were obtained. After selection of key features, a radiomics model, a clinical model, and a clinical-radiomics model for diagnosing DSI in early-stage cervical squamous cell carcinoma were developed by Logistic regression based on the training set. The performance of different models was compared by the receiver operating characteristic (ROC) curve in the validation set.Results A total of 168 patients with early-stage cervical squamous cell carcinoma that met the inclusion and exclusion criteria were included in this study. They were randomly divided into the training set (n=135) and the validation set (n=33), in which 72 cases had histopathologically confirmed superficial stromal invasion and 96 cases had DSI. Four radiomics features and three clinical parameters (age, Federation International of Gynecology and Obstetrics stage, and preoperative squamous cell carcinoma antigen levels) were selected and used to develop models. In the validation set, the clinical-radiomics model showed better diagnostic performance with the area under the curve (AUC) of 0.820 (95% CI: 0.665-0.974) than the clinical model[AUC: 0.797(95% CI: 0.623-0.971)] and the radiomics model[AUC: 0.793(95% CI: 0.633-0.954)].The sensitivity, specificity, and accuracy of the clinical-radiomics model were 85.7%(95% CI: 49.8%-100%), 73.7%(95% CI: 57.9%-100%), and 78.8%(95% CI: 69.7%-93.9%), respectively.Conclusion Radiomics features based on T2WI images combined with clinical features can be used as a noninvasive preoperative method to determine the depth of stromal invasion in early-stage cervical squamous cell carcinoma.
-
图 1 宫颈鳞癌患者矢状位T2WI图像及ROI勾画示意图
A.患者43岁，FIGO分期Ⅰ B2期，癌灶浸润最大深度为16 mm，此处宫颈间质厚度为25 mm，诊断为DSI；B.与图A为同一患者，黄色区域为沿肿瘤边缘勾画的ROI示意图；C.患者32岁，FIGO分期Ⅰ B2期，癌灶最大浸润深度为10 mm，此处宫颈间质厚度为23 mm，诊断为浅间质浸润；D.与图C为同一患者，黄色区域为沿肿瘤边缘勾画的ROI示意图
T2WI：同表 1；ROI：感兴趣区；FIGO：国际妇产科联盟；DSI: 深间质浸润
图 2 基于LASSO的Logistic回归模型特征选择图
A.通过5折交叉验证法筛选LASSO模型中的最优参数Lambda值，LASSO模型的复杂程度由Lambda(下横坐标)控制，Lambda越大对变量较多的线性模型的惩罚力度越大，从而最终获得1个变量较少的模型。纵坐标表示衡量模型的指标函数，由于因变量为二分类变量，故衡量模型的指标函数选择常用的“Deviance”，即-2×Log-likelihood。红点表示每个Lambda对应的目标参量，左侧竖线表示目标参量最小的Lambda值，右侧竖线表示在目标参量最小值的1个方差范围内得到最简单模型的Lambda值。上横坐标表示随Lambda增大，尚未被剔除变量的数目。B.LASSO模型的系数变化图。每条曲线代表每个特征系数的变化轨迹，在图A选择的最优参数Lambda值的位置(下横坐标)向上划1条竖线(图中未标出)，与之相交的变量即为模型最终所纳入的变量，变量所对应的纵坐标即为该变量的回归系数(可理解为该变量的贡献度)，上横坐标表示此时模型中非零系数的数目，下横坐标同图A
LASSO：最小绝对收缩和选择算子
图 4 早期宫颈鳞癌患者发生DSI的列线图
4个影像组学特征与其对应系数之积的线性和被定义为列线图上的影像组学得分(radiomics signature)；3个临床特征分别根据其赋分值在列线图上展示为“age”“SCC-Ag”和“FIGO”
DSI：同图 1表 1 MRI设备信息及矢状位T2WI图像参数
设备与参数 Signa Excite，GE Optima MR 360，GE Discovery MR 750W，GE Discovery MR 750，GE MagnetomⓇ Skyra，Siemens Ingenia CX，Philips 场强与图像采集 1.5T、FRFSE 1.5T、FRFSE 3T、FRFSE 3T、FRFSE 3T、TSE 3T、TSE 重复时间/回波时间(ms) 3400/88 4653/130 4273/79 3607/111 4010/115 3500/100 视野(mm2) 270×270 260×260 280×280 220×220 300×300 260×260 矩阵(频率×相位) 288×192 288×192 288×192 288×192 320×240 512×512 层厚(mm) 5.5 5 4.5 5 4 3 层间距(mm) 1 1 1 1 1.2 0.3 层数 16 19 24 16 24 29 激励次数 1 1 1.4 2 2 2 回波链 22 21 28 21 默认值 32 呼吸补偿 自由呼吸 自由呼吸 自由呼吸 自由呼吸 自由呼吸 自由呼吸 T2WI：T2加权成像 
下载: 导出CSV
表 2 训练集和验证集患者一般临床资料比较
指标 训练集 P值 验证集 P值 浅间质浸润(n=53) DSI(n=82) 浅间质浸润(n=19) DSI(n=14) 年龄(x±s，岁) 44.17±9.64 43.71±10.69 0.799 40.11±6.72 46.14±9.80 0.044 术前SCC-Ag(x±s，μg/L) 1.45±1.27 3.44±3.50 ＜0.001 1.76±0.98 3.29±3.10 0.097 绝经状态[n(%)] 0.328 0.628 未绝经 41(77.36) 56(68.29) 17(89.47) 11(78.57) 绝经 12(22.64) 26(31.71) 2(10.53) 3(21.43) FIGO分期[n(%)] ＜0.001 0.161 ⅠB1 29(54.72) 13(15.85) 10(52.63) 4(28.57) ⅠB2 21(39.62) 58(70.73) 9(47.37) 8(57.14) ⅠB3 0 3(3.66) 0 2(14.29) ⅡA1 3(5.66) 8(9.76) 0 0 DSI、FIGO：同图 1；SCC-Ag：鳞状细胞癌相关抗原
下载: 导出CSV
-
[1] Small W, JR, Bacon MA, Bajaj A, et al. Cervical cancer: A global health crisis[J]. Cancer, 2017, 123: 2404-2412. doi: 10.1002/cncr.30667 [2] Lu X, Jiang L, Zhang L, et al. Immune Signature-Based Subtypes of Cervical Squamous Cell Carcinoma Tightly Associated with Human Papillomavirus Type 16 Expression, Molecular Features, and Clinical Outcome[J]. Neoplasia, 2019, 21: 591-601. doi: 10.1016/j.neo.2019.04.003 [3] Sedlis A, Bundy BN, Rotman MZ, et al. A randomized trial of pelvic radiation therapy versus no further therapy in selected patients with stage IB carcinoma of the cervix after radical hysterectomy and pelvic lymphadenectomy: A Gynecologic Oncology Group Study[J]. Gynecol Oncol, 1999, 73: 177-183. doi: 10.1006/gyno.1999.5387 [4] National Comprehensive Cancer Network. NCCN Clinical Practice Guidelines in Oncology. Cervical Cancer (Version 1.2021)[EB/OL]. (2020-1-14)[2021-8-26]. https://www.nccn.org/guidelines/guidelines-detail?category=1&id=1426. [5] Chino J, Annunziata CM, Beriwal S, et al. Radiation Therapy for Cervical Cancer: Executive Summary of an ASTRO Clinical Practice Guideline[J]. Pract Radiat Oncol, 2020, 10: 220-234. doi: 10.1016/j.prro.2020.04.002 [6] Cibula D, Pötter R, Planchamp F, et al. The European Society of Gynaecological Oncology/European Society for Radiotherapy and Oncology/European Society of Pathology guidelines for the management of patients with cervical cancer[J]. Radiother Oncol, 2018, 127: 404-416. doi: 10.1016/j.radonc.2018.03.003 [7] Wu Q, Wang S, Chen X, et al. Radiomics analysis of magnetic resonance imaging improves diagnostic performance of lymph node metastasis in patients with cervical cancer[J]. Radiother Oncol, 2019, 138: 141-148. doi: 10.1016/j.radonc.2019.04.035 [8] Xiao M, Ma F, Li Y, et al. Multiparametric MRI-Based Radiomics Nomogram for Predicting Lymph Node Metastasis in Early-Stage Cervical Cancer[J]. J Magn Reson Imaging, 2020, 52: 885-896. doi: 10.1002/jmri.27101 [9] Li Z, Li H, Wang S, et al. MR-Based Radiomics Nomogram of Cervical Cancer in Prediction of the Lymph-Vascular Space Invasion preoperatively[J]. J Magn Reson Imaging, 2019, 49: 1420-1426. doi: 10.1002/jmri.26531 [10] Wu Q, Shi D, Dou S, et al. Radiomics Analysis of Multiparametric MRI Evaluates the Pathological Features of Cervical Squamous Cell Carcinoma[J]. J Magn Reson Imaging, 2019, 49: 1141-1148. doi: 10.1002/jmri.26301 [11] Bhatla N, Aoki D, Sharma DN, et al. Cancer of the cervix uteri[J]. Int J Gynecol Obstet, 2018, 143: 22-36. doi: 10.1002/ijgo.12611 [12] Allam M, Feely C, Millan D, et al. Depth of cervical stromal invasion as a prognostic factor after radical surgery for early stage cervical cancer[J]. Gynecol oncol, 2004, 93: 637-641. doi: 10.1016/j.ygyno.2004.02.029 [13] Bhatla N, Berek JS, Cuello Fredes M, et al. Revised FIGO staging for carcinoma of the cervix uteri[J]. Int J Gynaecol Obstet, 2019, 145: 129-135. doi: 10.1002/ijgo.12749 [14] Shu T, Zhao D, Li B, et al. Prognostic evaluation of postoperative adjuvant therapy for operable cervical cancer: 10 years' experience of National Cancer Center in China[J]. Chin J Cancer Res, 2017, 29: 510-520. doi: 10.21147/j.issn.1000-9604.2017.06.05 [15] Lakhman Y, Akin O, park KJ, et al. Stage IB1 cervical cancer: role of preoperative MR imaging in selection of patients for fertility-sparing radical trachelectomy[J]. Radiology, 2013, 269: 149-158. doi: 10.1148/radiol.13121746 [16] Landoni F, Colombo A, MIlani R, et al. Randomized study between radical surgery and radiotherapy for the treatment of stage IB-ⅡA cervical cancer: 20-year update[J]. J Gynecol Oncol, 2017, 28: e34. doi: 10.3802/jgo.2017.28.e34 [17] Moloney F, Ryan D, Twomey M, et al. Comparison of MRI and high-resolution transvaginal sonography for the local staging of cervical cancer[J]. J Clin Ultrasound, 2016, 44: 78-84. doi: 10.1002/jcu.22288 [18] 朱汇慈, 曹崑, 李晓婷, 等. MRI评估早期宫颈癌间质浸润深度[J]. 中国医学影像技术, 2020, 36: 559-563. https://www.cnki.com.cn/Article/CJFDTOTAL-ZYXX202004020.htmZhu HC, Cao K, Li XT, et al. MRI assessment of eary cervical cancer invasion depth of stroma[J]. Zhongguo Yixue Yingxiang Jishu, 2020, 36: 559-563. https://www.cnki.com.cn/Article/CJFDTOTAL-ZYXX202004020.htm [19] Epstein E, Testa A, Gaurilcikas A, et al. Early-stage cervical cancer: tumor delineation by magnetic resonance imaging and ultrasound - a European multicenter trial[J]. Gynecol Oncol, 2013, 128: 449-453. doi: 10.1016/j.ygyno.2012.09.025 [20] Xu D, Wang D, Wang S, et al. Correlation Between Squamous Cell Carcinoma Antigen Level and the Clinicopathological Features of Early-Stage Cervical Squamous Cell Carcinoma and the Predictive Value of Squamous Cell Carcinoma Antigen Combined With Computed Tomography Scan for Lymph Node Metastasis[J]. Int J Gynecol Cancer, 2017, 27: 1935-1942. doi: 10.1097/IGC.0000000000001112 [21] Liu S, Xia L, Yang Z, et al. The feasibility of (18)F-FDG PET/CT for predicting pathologic risk status in early-stage uterine cervical squamous cancer[J]. Cancer Imaging, 2020, 20: 63. doi: 10.1186/s40644-020-00340-z [22] 朱成功, 姜继勇, 张文清, 等. 术前血清鳞状细胞癌抗原水平对早期宫颈鳞癌淋巴结转移的预测价值[J]. 现代妇产科进展, 2020, 29: 896-900. https://www.cnki.com.cn/Article/CJFDTOTAL-XDFC202012004.htmZhu CG, Jiang JY, Zhang WQ, et al. Predictive value of preoperative serum squamous cell carcinoma antigen level for lymph node metastasis of early-stage cervical squamous cell carcinoma[J]. Xiandai Fuchanke Jinzhan, 2020, 29: 896-900. https://www.cnki.com.cn/Article/CJFDTOTAL-XDFC202012004.htm -
下载:
点击查看大图
图(4) / 表(3)
计量
- 文章访问数: 215
- HTML全文浏览量: 263
- PDF下载量: 0
- 被引次数: 0
京公网安备 11010502036328号 京ICP备17033152号