Hefei Cancer Hospital, Chinese Academy of Sciences, Hefei, 230031, P.R. China.
Anhui Province Key Laboratory of Medical Physics and Technology, Institute of Health and Medical Technology, Hefei Institutes of Physical Science, Chinese Academy of Sciences, Hefei, 230031, P.R. China.
BMC Med Imaging. 2023 Oct 2;23(1):145. doi: 10.1186/s12880-023-01089-0.
Radical concurrent chemoradiotherapy (CCRT) is frequently used as the first-line treatment for patients with locally advanced esophageal cancer. Unfortunately, some patients respond poorly. To predict response to radical concurrent chemoradiotherapy in pre-treatment patients with esophageal squamous carcinoma (ESCC), and compare the predicting efficacies of radiomics features of primary tumor with or without regional lymph nodes, we developed a radiomics-clinical model based on the positioning CT images. Finally, SHapley Additive exPlanation (SHAP) was used to explain the models.
This retrospective study enrolled 105 patients with medically inoperable and/or unresectable ESCC who underwent radical concurrent chemoradiotherapy (CCRT) between October 2018 and May 2023. Patients were classified into responder and non-responder groups with RECIST standards. The 11 recently admitted patients were chosen as the validation set, previously admitted patients were randomly split into the training set (n = 70) and the testing set (n = 24). Primary tumor site (GTV), the primary tumor and the uninvolved lymph nodes at risk of microscopic disease (CTV) were identified as Regions of Interests (ROIs). 1762 radiomics features from GTV and CTV were respectively extracted and then filtered by statistical differential analysis and Least Absolute Shrinkage and Selection Operator (LASSO). The filtered radiomics features combined with 13 clinical features were further filtered with Mutual Information (MI) algorithm. Based on the filtered features, we developed five models (Clinical Model, GTV Model, GTV-Clinical Model, CTV Model, and CTV-Clinical Model) using the random forest algorithm and evaluated for their accuracy, precision, recall, F1-Score and AUC. Finally, SHAP algorithm was adopted for model interpretation to achieve transparency and utilizability.
The GTV-Clinical model achieves an AUC of 0.82 with a 95% confidence interval (CI) of 0.76-0.99 on testing set and an AUC of 0.97 with a 95% confidence interval (CI) of 0.84-1.0 on validation set, which are significantly higher than those of other models in predicting ESCC response to CCRT. The SHAP force map provides an integrated view of the impact of each feature on individual patients, while the SHAP summary plots indicate that radiomics features have a greater influence on model prediction than clinical factors in our model.
GTV-Clinical model based on texture features and the maximum diameter of lesion (MDL) may assist clinicians in pre-treatment predicting ESCC response to CCRT.
根治性同期放化疗（CCRT）常被用作局部晚期食管癌患者的一线治疗方法。但不幸的是，部分患者的治疗效果较差。为了预测接受根治性同期放化疗的食管鳞癌（ESCC）患者的反应，并比较原发肿瘤和区域淋巴结的放射组学特征的预测效能，我们基于定位 CT 图像建立了一个放射组学-临床模型。最后，采用 SHapley Additive exPlanation (SHAP) 方法对模型进行解释。
本回顾性研究纳入了 2018 年 10 月至 2023 年 5 月期间接受根治性同期放化疗（CCRT）的 105 例无法手术和/或不可切除的 ESCC 患者。根据 RECIST 标准，患者被分为应答组和非应答组。随机选取最近收治的 11 例患者作为验证集，其余患者分为训练集（n=70）和测试集（n=24）。分别从 GTV 和 CTV 中提取原发肿瘤部位（GTV）、原发肿瘤和可能存在微观疾病的未受累淋巴结（CTV）作为感兴趣区（ROI）。通过统计差异分析和最小绝对值收缩和选择算子（LASSO）对 GTV 和 CTV 的 1762 个放射组学特征进行筛选。结合 13 个临床特征，利用互信息（MI）算法进一步筛选。基于筛选出的特征，采用随机森林算法建立五个模型（临床模型、GTV 模型、GTV-临床模型、CTV 模型和 CTV-临床模型），并对其准确性、精确性、召回率、F1 评分和 AUC 进行评估。最后，采用 SHAP 算法对模型进行解释，以实现透明度和可利用性。
在测试集上，GTV-临床模型的 AUC 为 0.82（95%CI：0.76-0.99），在验证集上的 AUC 为 0.97（95%CI：0.84-1.0），预测 ESCC 对 CCRT 反应的准确性显著高于其他模型。SHAP 力图提供了每个特征对个体患者影响的综合视图，而 SHAP 汇总图则表明，在我们的模型中，放射组学特征对模型预测的影响大于临床因素。
基于纹理特征和病灶最大直径（MDL）的 GTV-临床模型，可能有助于临床医生在治疗前预测 ESCC 对 CCRT 的反应。
