基于多模态磁共振成像非转移性肾透明细胞癌预后评估的影像基因组学研究

肾透明细胞癌（clear cell renal cell carcinoma, ccRCC）是肾脏最常见的恶性肿瘤，初诊时未发生转移的ccRCC患者（即非转移性ccRCC）术后约30%发生复发转移。如何准确地对非转移性ccRCC患者的预后做出预测是当前研究的热点。本课题组将在前期肾细胞癌MR研究基础上，利用基于多模态磁共振成像的影像组学新技术，回顾性提取非转移性ccRCC病变术前MR图像的影像组学特征，筛选与肿瘤预后高度相关的影像组学特征，并分析其与预后相关基因（BAP1、PBRM1、SETD2）的潜在联系。针对非转移性ccRCC患者的预后差异问题，课题组将利用遴选的特征构建非转移性ccRCC预后评估模型以及联合临床信息及病理指标的综合评估模型，并对模型进行验证优化，最后前瞻性收集病例对模型进行测试，以期为非转移性ccRCC患者的预后提供新颖、早期、定量的评估方法，从而改善患者预后。

Clear cell renal cell carcinoma (ccRCC) is the most common malignant renal tumor. About 30% newly diagnosed patients with ccRCC that are not metastasized (non-metastatic ccRCC) relapsed and metastasized after surgery. It has been a hot topic that the prognostic prediction of patients with non-metastatic ccRCC was expected to be early and accurate. Using radiomics technology concerning multimodal magnetic resonance imaging(MRI), we respectively extracted radiomics features from preoperative MR images of non-metastatic ccRCC and filtrated features highly correlated with tumor prognosis based on our previous studies about MR imaging of the renal cell carcinoma. Then, we will explore the potential correlation between features and prognosis-oriented genes (BAP1, PBRM1 & SETD2). Regarding the prognostic heterogeneity among patients with non-metastatic ccRCC, the prognostic assessment model for non-metastatic ccRCC will be established based on radiomics features. Moreover, the comprehensive assessment model is also scheduled combined with clinical indexes, imaging parameters and pathological indicators. Both the models will get validated and updated with repeated practice. Eventually, we are going to test the model by prospectively gathering cases of non-metastatic ccRCC. It is hoped that this new method will be able to make early quantitative prognostic evaluation on patients with non-metastatic ccRCC, so as to improve their prognosis.