Wen Xin, Liu Meng-Wen, Qiu Bin, Wang Yan-Mei, Jiang Jiu-Ming, Zhang Xue, Jiang Xu, Li Lin, Li Meng, Zhang Li
Department of Diagnostic Radiology, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China.
Department of Thoracic Surgery, National Cancer Center/National Clinical Research Center for Cancer/Cancer Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China.
Transl Lung Cancer Res. 2024 Aug 31;13(8):1794-1806. doi: 10.21037/tlcr-24-283. Epub 2024 Aug 12.
BACKGROUND: Research has demonstrated that radiomics models are capable of forecasting the characteristics of lung cancer. Nevertheless, due to radiomics' poor interpretability, its applicability in clinical settings remains restricted. This investigation sought to verify the correlation between radiomics features (RFs) and the biological behavior of clinical stage IA adenocarcinomas. METHODS: A retrospective analysis was conducted on patients diagnosed with clinical stage IA lung adenocarcinoma who underwent resection between May 2005 and December 2018. Detailed radiomics examination of the primary tumor was carried out utilizing preoperative computed tomography (CT) images. Subsequently, patients were grouped based on their RFs using consensus clustering, enabling comparison of tumor biological characteristics among the clusters. Survival disparities among the clusters were evaluated through Kaplan-Meier and Cox analyses. RESULTS: A consensus cluster analysis was performed on 669 patients [median age, 58 years; interquartile range (IQR), 50-64 years, 257 males, 412 females], and three distinct clusters were identified. Cluster 2 was associated with radiological solid adenocarcinoma [119 of 324 (36.7%), P<0.001], larger tumors with median tumor size of 2.1 cm with IQR of 1.7 to 2.5 cm (P<0.001), central tumor [91 of 324 (28.1%), P=0.002], pleural invasion [87 of 324 (26.9%), P<0.001], occult lymph node metastasis (ONM) [106 of 324 (32.7%), P<0.001], and a higher frequency of metastasis or recurrence [62 of 324 (19.1%), P<0.001]. The frequency of histological grade 3 was the highest in Cluster 3 [8 of 34 (23.5%), P<0.001]. Cluster 1 was associated with pure ground glass nodules (pGGNs) [184 of 310 (59.4%), P<0.001], smaller tumors with median tumor size of 1.1 cm with IQR of 0.8 to 1.4 cm (P<0.001), no pleural invasion [276 of 310 (89.0%), P<0.001], histological grade 1 [114 of 248 (46.0%), P<0.001], ONM negative [292 of 310 (94.2%), P<0.001], and a lower rate of metastasis or recurrence [298 of 310 (96.1%), P<0.001]. CONCLUSIONS: Differences in tumor biological behavior were detected among consensus clusters based on the RFs of clinical stage IA adenocarcinoma.
背景：研究表明，放射组学模型能够预测肺癌的特征。然而，由于放射组学的可解释性较差，其在临床环境中的适用性仍然受限。本研究旨在验证放射组学特征（RFs）与临床ⅠA期腺癌生物学行为之间的相关性。 方法：对2005年5月至2018年12月期间接受手术切除的临床ⅠA期肺腺癌患者进行回顾性分析。利用术前计算机断层扫描（CT）图像对原发性肿瘤进行详细的放射组学检查。随后，使用一致性聚类根据患者的RFs进行分组，以便比较各聚类之间的肿瘤生物学特征。通过Kaplan-Meier和Cox分析评估各聚类之间的生存差异。 结果：对669例患者[中位年龄58岁；四分位间距（IQR），50 - 64岁，男性257例，女性412例]进行了一致性聚类分析，确定了三个不同的聚类。聚类2与放射学实性腺癌相关[324例中的119例（36.7%），P<0.001]，肿瘤较大，中位肿瘤大小为2.1 cm，IQR为1.7至2.5 cm（P<0.001），中央型肿瘤[324例中的91例（28.1%），P = 0.002]，胸膜侵犯[324例中的87例（26.9%），P<0.001]，隐匿性淋巴结转移（ONM）[324例中的106例（32.7%），P<0.001]，以及转移或复发频率较高[324例中的62例（19.1%），P<0.001]。组织学3级的频率在聚类3中最高[34例中的8例（23.5%），P<0.001]。聚类1与纯磨玻璃结节（pGGNs）相关[310例中的184例（59.4%），P<0.001]，肿瘤较小，中位肿瘤大小为1.1 cm，IQR为0.8至1.4 cm（P<0.001），无胸膜侵犯[310例中的276例（89.0%），P<0.001]，组织学1级[248例中的114例（46.0%），P<0.001]，ONM阴性[310例中的292例（94.2%），P<0.001]，以及转移或复发率较低[310例中的298例（96.1%），P<0.001]。 结论：基于临床ⅠA期腺癌的RFs，在一致性聚类之间检测到肿瘤生物学行为的差异。
Zhonghua Zhong Liu Za Zhi. 2019-6-23
Cancer Biol Med. 2022-12-5
J Natl Compr Canc Netw. 2022-5
Zotero 插件
