A Combined Nomogram Model to Preoperatively Predict Histologic Grade in Pancreatic Neuroendocrine Tumors

Abstract

Purpose: The purpose of this study is to develop and validate a nomogram model combing radiomics features and clinical characteristics to preoperatively differentiate grade 1 and grade 2/3 tumors in patients with pancreatic neuroendocrine tumors (pNET). Experimental Design: A total of 137 patients who underwent contrast-enhanced CT from two hospitals were included in this study. The patients from the second hospital (n ¼ 51) were selected as an independent validation set. The arterial phase in contrast-enhanced CT was selected for radiomics feature extraction. The Mann-Whitney U test and least absolute shrinkage and selection operator regression were applied for feature selection and radiomics signature construction. A combined nomogram model was developed by incorporating the radiomics signature with clinical factors. The association between the nomogram model and the Ki-67 index and rate of nuclear mitosis were also investigated respectively. The utility of the proposed model was evaluated using the ROC, area under ROC curve (AUC), calibration curve, and decision curve analysis (DCA). The Kaplan-Meier (KM) analysis was used for survival analysis. Results: An eight-feature-combined radiomics signature was constructed as a tumor grade predictor. The nomogram model combining the radiomics signature with clinical stage showed the best performance (training set: AUC ¼ 0.907; validation set: AUC ¼ 0.891). The calibration curve and DCA demonstrated the clinical usefulness of the proposed nomogram. A significant correlation was observed between the developed nomogram and Ki-67 index and rate of nuclear mitosis, respectively. The KM analysis showed a significant difference between the survival of predicted grade 1 and grade 2/3 groups (P ¼ 0.002). Conclusions: The combined nomogram model developed could be useful in differentiating grade 1 and grade 2/3 tumor in patients with pNETs.

Prognostic nomograms for predicting overall survival of Paget ’s disease

“…In previous study, increasing age and being unmarried (divorced, separated, and never married) were found to be potential factors that increased the risk of Paget's disease [31][32][33]. According to the previous study, for PD alone and patients with intraductal ductal carcinoma and breast cancer, breast conserving surgery combined with radiotherapy had a 5-year local recurrence rate of 5% [34], which hintingthat surgery combined with radiotherapy could effectively affect the prognosis of Paget's disease. Our analysis was consistent with these previous reports.…”

A novel and validated nomogram to predict overall survival for gastric neuroendocrine neoplasms

“…Nevertheless, literatures reporting nomograms in predicting outcome in various type of neuroendocrine tumors includes similar number of patients in development of the predictive model. [41][42][43] Therefore, it should be a reasonable number of patients in the training cohort of the present study. Second, due to the retrospective nature of two cohorts, our nomogram might inevitably have selection bias.…”

A prospectively validated machine learning model for the prediction of survival and tumor subtype in pancreatic ductal adenocarcinoma

“…For example, CT-derived radiomic signatures were shown to enable prediction of local disease control and overall survival in PDAC (11,12) or tumor grading in pancreatic neuroendocrine tumors (13). The large-scale implementation of such tools thus has the potential to become a game changer in medical image interpretation and individualized patient care.…”