Radiomics Nomogram in Assisting Lymphadenectomy Decisions by Predicting Lymph Node Metastasis in Early-Stage
Xue-Fei Liu1, Bi-Cong Yan1,2, Ying Li1
1Department of Radiology, Jinshan Hospital, Fudan University, Shanghai, China.
Background:
Lymph node metastasis (LNM) is an important risk factor affecting treatment strategy and prognosis for endometrial cancer (EC) patients. A radiomics nomogram was established in assisting lymphadenectomy decisions preoperatively by predicting LNM status in early-stage EC patients.
Methods:
A total of 707 retrospective clinical early-stage EC patients were enrolled and randomly divided into a training cohort and a test cohort. Radiomics features were extracted from MR imaging. Three models were built, including a guideline-recommended clinical model (grade 1-2 endometrioid tumors by dilatation and curettage and less than 50% myometrial invasion on MRI without cervical infiltration), a radiomics model (selected radiomics features), and a radiomics nomogram model (combing the selected radiomics features, myometrial invasion on MRI, and cancer antigen 125). The predictive performance of the three models was assessed by the area under the receiver operating characteristic (ROC) curves (AUC). The clinical decision curves, net reclassification index (NRI), and total integrated discrimination index (IDI) based on the total included patients to assess the clinical benefit of the clinical model and the radiomics nomogram were calculated.
Results:
The predictive ability of the clinical model, the radiomics model, and the radiomics nomogram between LNM and non-LNM were 0.66 [95% CI: 0.55-0.77], 0.82 [95% CI: 0.74-0.90], and 0.85 [95% CI: 0.77-0.93] in the training cohort, and 0.67 [95% CI: 0.56-0.78], 0.81 [95% CI: 0.72-0.90], and 0.83 [95% CI: 0.74-0.92] in the test cohort, respectively. The decision curve analysis, NRI (1.06 [95% CI: 0.81-1.32]), and IDI (0.05 [95% CI: 0.03-0.07]) demonstrated the clinical usefulness of the radiomics nomogram.
Conclusions:
The predictive radiomics nomogram could be conveniently used for individualized prediction of LNM and assisting lymphadenectomy decisions in early-stage EC patients.
