文献速递胸部肿瘤

基于原发肿瘤与淋巴结影像组学预测非小细胞肺癌病理缓解

Radiomic-Based Pathological Response Prediction from Primary Tumors and Lymph Nodes in NSCLC.
PubMed 文献 · J Thorac Oncol · 2016-11-27

本研究纳入2003至2013年85例可切除局部晚期(II-III期)非小细胞肺癌患者,对85个原发肿瘤和178个淋巴结进行影像组学分析,以区分病理完全缓解(pCR)与大体残留病灶(GRD)。三个影像组学特征可显著预测pCR(AUC均为0.67,p<0.05),两个淋巴结均匀性特征预测GRD的AUC为0.72-0.75,优于原发肿瘤特征(AUC=0.62)。结论认为淋巴结表型信息对病理缓解有显著预测价值,分类性能高于原发肿瘤影像组学特征。

要点

  • 淋巴结影像组学特征预测GRD的AUC为0.72-0.75,优于原发肿瘤特征的AUC=0.62。
  • 三个影像组学特征预测pCR的AUC均为0.67(p<0.05)。
  • GRD分类中影像组学联合临床数据的AUC为0.73,优于其他特征集。

核心结论

  • 淋巴结表型信息对病理缓解有显著预测价值,分类性能高于原发肿瘤影像组学特征。

研究设计与样本

  • 经机构审查委员会批准的研究,纳入2003至2013年85例可切除局部晚期(II-III期)NSCLC患者(中位年龄60.3岁,65%为女性)。
  • 对85个原发肿瘤和178个淋巴结进行影像组学分析,评估20个非冗余且稳定的特征(每个部位10个)。
  • 采用随机森林和嵌套交叉验证评估各特征集的分类性能。

主要发现

  • 三个影像组学特征(描述原发肿瘤球形度和淋巴结均匀性)可显著预测pCR,AUC均为0.67(p<0.05)。
  • 两个特征(量化淋巴结均匀性)可预测GRD,AUC范围为0.72-0.75(p<0.05),显著优于原发肿瘤特征(AUC=0.62)。
  • 多变量分析显示,对于pCR,仅影像组学特征集分类性能最佳(中位AUC=0.68)。
  • 对于GRD分类,影像组学与临床数据联合显著优于所有其他特征集(中位AUC=0.73)。

局限性

  • 原文未报告局限性。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

INTRODUCTION: Noninvasive biomarkers that capture the total tumor burden could provide important complementary information for precision medicine to aid clinical decision making. We investigated the value of radiomic data extracted from pretreatment computed tomography images of the primary tumor and lymph nodes in predicting pathological response after neoadjuvant chemoradiation before surgery. METHODS: A total of 85 patients with resectable locally advanced (stage II-III) NSCLC (median age 60.3 years, 65% female) treated from 2003 to 2013 were included in this institutional review board-approved study. Radiomics analysis was performed on 85 primary tumors and 178 lymph nodes to discriminate between pathological complete response (pCR) and gross residual disease (GRD). Twenty nonredundant and stable features (10 from each site) were evaluated by using the area under the curve (AUC) (all p values were corrected for multiple hypothesis testing). Classification performance of each feature set was evaluated by random forest and nested cross validation. RESULTS: Three radiomic features (describing primary tumor sphericity and lymph node homogeneity) were significantly predictive of pCR with similar performances (all AUC = 0.67, p < 0.05). Two features (quantifying lymph node homogeneity) were predictive of GRD (AUC range 0.72-0.75, p < 0.05) and performed significantly better than the primary features (AUC = 0.62). Multivariate analysis showed that for pCR, the radiomic features set alone had the best-performing classification (median AUC = 0.68). Furthermore, for GRD classification, the combination of radiomic and clinical data significantly outperformed all other feature sets (median AUC = 0.73). CONCLUSION: Lymph node phenotypic information was significantly predictive for pathological response and showed higher classification performance than radiomic features obtained from the primary tumor.

来源:J Thorac Oncol,查看原文。版权归原作者所有。

引用

  1. Coroller TP, Agrawal V, Huynh E, et al. Radiomic-Based Pathological Response Prediction from Primary Tumors and Lymph Nodes in NSCLC. J Thorac Oncol. 2016. PMID: 27903462. doi:10.1016/j.jtho.2016.11.2226

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。