Intra- and peritumoral MRI radiomics assisted in predicting radiochemotherapy response in metastatic cervical lymph nodes of nasopharyngeal cancer

Abstract

Background To establish and validate radiomic models combining intratumoral (Intra) and peritumoral (Peri) features obtained from pretreatment MRI for the prediction of treatment response of lymph node metastasis from nasopharyngeal cancer (NPC). Methods One hundred forty-five NPC patients (102 in the training and 43 in the validation set) were retrospectively enrolled. Radiomic features were extracted from Intra and Peri regions on the metastatic cervical lymph node, and selected with the least absolute shrinkage and selection operator (LASSO). Multivariate logistic regression analysis was applied to build radiomic models. Sensitivity, specificity, accuracy, and the area under the curve (AUC) of receiver operating characteristics were employed to evaluate the predictive power of each model. Results The AUCs of the radiomic model of Intra, Peri, Intra + Peri, and Clinical-radiomic were 0.910, 0.887, 0.934, and 0.941, respectively, in the training set and 0.737, 0.794, 0.774, and 0.783, respectively, in the validation set. There were no significant differences in prediction performance among the radiomic models in the training and validation sets (all P > 0.05). The calibration curve of the radiomic model of Peri demonstrated good agreement between prediction and observation in the training and validation sets. Conclusions The pretreatment MRI-based radiomics model may be useful in predicting the treatment response of metastatic lymph nodes of NPC. Besides, the generalization ability of the radiomic model of Peri was better than that of Intra and Intra + Peri.

CT-based peritumoral radiomics nomogram on prediction of response and survival to induction chemotherapy in locoregionally advanced nasopharyngeal carcinoma

“…Recently, a study revealed the heterogeneity of tumor microenvironment (TME) via virtual microdissection of gene expression profiles, which classified patients into 3 subgroups(Chen et al 2021 ), though it did not utilize the medical imaging generated from each patient. Another study demonstrated the potential significance of peritumoral regions of MRI in the treatment response of lymphatic node metastasis(Xu et al 2023 ). Despite encouraging results, the study cohort only recruited 145 patients and only explored 2 mm dilation, reaching an AUC of 0.78 in the clinic combined with radiomics model in the testing set.…”

Radiomics-based lymph nodes prognostic models from three MRI regions in nasopharyngeal carcinoma

“…Some studies reported on models constructed using radiomics features from the lymph nodes and briefly introduced the method for lymph node delineation in patients with NPC [ [21] , [22] , [23] , [24] , [27] , [28] , [29] , [30] , [31] , [32] ]. Xu et al [ 27 ] employed intra- and peritumoral MRI radiomics directly extracted from the largest metastatic lymph node to predict treatment response in patients with NPC. Lam et al [ 28 ] revealed that CT-based radiomics based on all metastatic lymph nodes was capable of predicting radiotherapy-related events in patients with NPC.…”