Demirjian Natalie L, Varghese Bino A, Cen Steven Y, Hwang Darryl H, Aron Manju, Siddiqui Imran, Fields Brandon K K, Lei Xiaomeng, Yap Felix Y, Rivas Marielena, Reddy Sharath S, Zahoor Haris, Liu Derek H, Desai Mihir, Rhie Suhn K, Gill Inderbir S, Duddalwar Vinay
College of Medicine - Tucson, University of Arizona, Tucson, AZ, USA.
Department of Radiology, Keck School of Medicine, University of Southern California, Los Angeles, CA, USA.
Eur Radiol. 2022 Apr;32(4):2552-2563. doi: 10.1007/s00330-021-08344-4. Epub 2021 Nov 10.
To evaluate the utility of CT-based radiomics signatures in discriminating low-grade (grades 1-2) clear cell renal cell carcinomas (ccRCC) from high-grade (grades 3-4) and low TNM stage (stages I-II) ccRCC from high TNM stage (stages III-IV).
A total of 587 subjects (mean age 60.2 years ± 12.2; range 22-88.7 years) with ccRCC were included. A total of 255 tumors were high grade and 153 were high stage. For each subject, one dominant tumor was delineated as the region of interest (ROI). Our institutional radiomics pipeline was then used to extract 2824 radiomics features across 12 texture families from the manually segmented volumes of interest. Separate iterations of the machine learning models using all extracted features (full model) as well as only a subset of previously identified robust metrics (robust model) were developed. Variable of importance (VOI) analysis was performed using the out-of-bag Gini index to identify the top 10 radiomics metrics driving each classifier. Model performance was reported using area under the receiver operating curve (AUC).
The highest AUC to distinguish between low- and high-grade ccRCC was 0.70 (95% CI 0.62-0.78) and the highest AUC to distinguish between low- and high-stage ccRCC was 0.80 (95% CI 0.74-0.86). Comparable AUCs of 0.73 (95% CI 0.65-0.8) and 0.77 (95% CI 0.7-0.84) were reported using the robust model for grade and stage classification, respectively. VOI analysis revealed the importance of neighborhood operation-based methods, including GLCM, GLDM, and GLRLM, in driving the performance of the robust models for both grade and stage classification.
Post-validation, CT-based radiomics signatures may prove to be useful tools to assess ccRCC grade and stage and could potentially add to current prognostic models. Multiphase CT-based radiomics signatures have potential to serve as a non-invasive stratification schema for distinguishing between low- and high-grade as well as low- and high-stage ccRCC.
• Radiomics signatures derived from clinical multiphase CT images were able to stratify low- from high-grade ccRCC, with an AUC of 0.70 (95% CI 0.62-0.78). • Radiomics signatures derived from multiphase CT images yielded discriminative power to stratify low from high TNM stage in ccRCC, with an AUC of 0.80 (95% CI 0.74-0.86). • Models created using only robust radiomics features achieved comparable AUCs of 0.73 (95% CI 0.65-0.80) and 0.77 (95% CI 0.70-0.84) to the model with all radiomics features in classifying ccRCC grade and stage, respectively.
评估基于CT的影像组学特征在鉴别低级别（1-2级）透明细胞肾细胞癌（ccRCC）与高级别（3-4级）ccRCC以及低TNM分期（I-II期）ccRCC与高TNM分期（III-IV期）ccRCC中的应用价值。
共纳入587例ccRCC患者（平均年龄60.2岁±12.2；范围22-88.7岁）。其中255个肿瘤为高级别，153个为高分期。对于每位患者，将一个主要肿瘤划定为感兴趣区域（ROI）。然后使用我们机构的影像组学流程，从手动分割的感兴趣体积中，跨12个纹理家族提取2824个影像组学特征。使用所有提取特征（完整模型）以及仅使用先前确定的稳健指标子集（稳健模型）分别开发机器学习模型的不同迭代。使用袋外基尼指数进行变量重要性（VOI）分析，以确定驱动每个分类器的前10个影像组学指标。使用受试者工作特征曲线下面积（AUC）报告模型性能。
区分低级别和高级别ccRCC的最高AUC为0.70（95%CI 0.62-0.78），区分低分期和高分期ccRCC的最高AUC为0.80（95%CI 0.74-0.86）。使用稳健模型进行分级和分期分类时，报告的AUC分别为可比的0.73（95%CI 0.65-0.8）和0.77（95%CI 0.7-0.84）。VOI分析揭示了基于邻域运算的方法（包括灰度共生矩阵（GLCM）、灰度依赖矩阵（GLDM）和灰度游程长度矩阵（GLRLM））在驱动稳健模型进行分级和分期分类性能方面的重要性。
经过验证后，基于CT的影像组学特征可能被证明是评估ccRCC分级和分期的有用工具，并可能为当前的预后模型增添内容。基于多期CT的影像组学特征有潜力作为一种非侵入性分层方案，用于区分低级别和高级别以及低分期和高分期的ccRCC。
• 从临床多期CT图像中得出的影像组学特征能够区分低级别和高级别ccRCC，AUC为0.70（95%CI 0.62-0.78）。• 从多期CT图像中得出的影像组学特征在ccRCC中具有区分低TNM分期和高TNM分期的判别能力，AUC为0.80（95%CI 0.74-0.86）。• 仅使用稳健影像组学特征创建的模型在对ccRCC进行分级和分期分类时，与使用所有影像组学特征的模型相比，分别实现了可比的AUC，即0.73（95%CI 0.65-0.80）和0.77（95%CI 0.70-0.84）。
