Background
Papillary thyroid carcinoma (PTC) is characterized by frequent metastasis to cervical lymph nodes (CLNs), and the presence of lymph node metastasis at diagnosis has a significant impact on the surgical approach. Therefore, we established a radiomic signature to predict the CLN status of PTC patients using preoperative thyroid ultrasound and investigated the association between the radiomic features and underlying molecular characteristics of PTC tumours.
Methods
A radiogenomic map linking radiomics features to gene modules was constructed, and immunohistochemistry was performed to validate key associations. In all, 180 patients were enrolled in this prospective study, and 47 radiomic features, including tumour size, shape, position, margin, echo pattern and calcification, were extracted. Total protein extracted from 49 tumour samples was analysed with LC/MS and iTRAQ technology. Gene modules acquired by clustering were chosen for their diagnostic significance. A radiogenomic map linking radiomic features to gene modules was constructed with the Spearman correlation matrix. Immunohistochemistry was performed to validate key associations between radiomic features and gene modules.
Results
The diagnostic performance of radiomics signature was better than that of the ultrasound-based method in predicting CLN status. Weighted gene co-expression network analysis generated 16 gene modules, and a radiogenomic map with nine significant correlations between radiomics features and gene modules was created. For example, the feature ‘minimum calcification area’ was significantly associated with module MEblue, which represents cell-cell adhesion and glycolysis. Immunohistochemistry showed that LAMC1 and THBS1 expression was associated with several radiomics features.
Conclusions
The radiomic signature proposed here has the potential to noninvasively predict the CLN status in PTC patients. Merging imaging phenotypes with genomic data could allow the noninvasive identification of the molecular properties of PTC tumours, which would support clinical decision making with personalized management.
Trail registration:
This prospective study was approved by the ethics committee of the Fudan University Shanghai Cancer Centre (1809191-1-NSFC) in July 2018.