CT radiomics may predict the grade of pancreatic neuroendocrine tumors: a multicenter study
To develop and validate a radiomics-based nomogram for preoperatively predicting grade 1 and grade 2/3 tumors in patients with pancreatic neuroendocrine tumors (PNETs).
One hundred thirty-eight patients derived from two institutions with pathologically confirmed PNETs (104 in the training cohort and 34 in the validation cohort) were included in this retrospective study. A total of 853 radiomic features were extracted from arterial and portal venous phase CT images respectively. Minimum redundancy maximum relevance and random forest methods were adopted for the significant radiomic feature selection and radiomic signature construction. A fusion radiomic signature was generated by combining both the single-phase signatures. The nomogram based on a comprehensive model incorporating the clinical risk factors and the fusion radiomic signature was established, and decision curve analysis was applied for clinical use.
The fusion radiomic signature has significant association with histologic grade (p < 0.001). The nomogram integrating independent clinical risk factor tumor margin and fusion radiomic signature showed strong discrimination with an area under the curve (AUC) of 0.974 (95% CI 0.950–0.998) in the training cohort and 0.902 (95% CI 0.798–1.000) in the validation cohort with good calibration. Decision curve analysis verified the clinical usefulness of the predictive nomogram.
We proposed a comprehensive nomogram consisting of tumor margin and fusion radiomic signature as a powerful tool to predict grade 1 and grade 2/3 PNET preoperatively and assist the clinical decision-making for PNET patients.
• Radiomic signature has strong discriminatory ability for the histologic grade of PNETs.
• Arterial and portal venous phase CT imaging are complementary for the prediction of PNET grading.
• The comprehensive nomogram outperformed clinical factors in assisting therapy strategy in PNET patients.
KeywordsNeoplasm grading Pancreas Neuroendocrine tumor Radiomics CT
Area under the curve
Carbohydrate antigen 19-9
Dilatation of the main pancreatic duct
Gray level co-occurrence matrix
Gray level dependence matrix
Gray level run length matrix
Gray level size zone matrix
Intra- and inter-class correlation coefficient
Minimum redundancy maximum relevance
Neighboring gray tone difference matrix
Negative predictive value
Picture archiving and communication system
Preoperative blood glucose
Protrusion from the outline of the pancreas
Preoperative liver metastasis
Pancreatic neuroendocrine tumors
Positive predictive value
Receiver operating characteristics
Region of interest
World Health Organization
This study has received funding by the National Natural Science Foundation of China (No. 81227901, 81527805, 61231004, 81771924, 81501616), National Key Research and Development Program of China (2017YFA0205200, 2017YFC1308700), and the Science and Technology Service Network Initiative of the Chinese Academy of Sciences (KFJ-SW-STS-160).
Compliance with ethical standards
The scientific guarantor of this publication is Jie Tian.
Conflict of interest
The authors declare that they have no competing interests.
Statistics and biometry
Dr. Jingwei Wei from the University of Chinese Academy of Sciences, who is one of the authors, has significant statistical expertise.
Written informed consent was waived by the Institutional Review Board of Zhongshan Hospital Affiliated to Shanghai Fudan University and Affiliated Hospital (Laoshan hospital) of Qingdao University.
Institutional Review Board approval was obtained.
• diagnostic or prognostic study
• multicenter study
- 3.Bosman FT, Carneiro F, Hruban RH, Theise ND (2010) WHO classification of tumours of the digestive system, 4th edn. International Agency for Research on Cancer, LyonGoogle Scholar
- 23.Zhao W, Quan Z, Huang X et al (2018) Grading of pancreatic neuroendocrine neoplasms using pharmacokinetic parameters derived from dynamic contrast-enhanced MRI. Oncol Lett 15:8349–8356Google Scholar
- 26.Chen J, Tian J (2009) Real-time multi-modal rigid registration based on a novel symmetric-SIFT descriptor. Progress in Natural Science-Materials International 19:643-651Google Scholar
- 31.Ger RB, Cardenas CE, Anderson BM, Yang J, Mackin DS, Zhang L (2018) Guidelines and experience using imaging biomarker explorer (IBEX) for Radiomics. J Vis Exp:e57132Google Scholar
- 32.Zhang B, Tian J, Dong D et al (2017) Radiomics Features of Multiparametric MRI as Novel Prognostic Factors in Advanced Nasopharyngeal Carcinoma. Clin Cancer Res 23:4259–4269Google Scholar
- 34.Huang X, Cheng Z, Huang Y et al (2018) CT-based radiomics signature to discriminate high-grade from Low-grade colorectal adenocarcinoma. Acad Radiol. https://doi.org/10.1016/j.acra.2018.01.020
- 35.Tian Q, Yan LF, Zhang X et al (2018) Radiomics strategy for glioma grading using texture features from multiparametric MRI. J Magn Reson Imaging. https://doi.org/10.1002/jmri.26010
- 37.Phuong TM, Lin Z, Altman RB (2005) Choosing SNPs using feature selection. Computational Systems Bioinformatics Conference 301–309Google Scholar
- 40.Hu Y, Rao S, Xu X, Tang Y, Zeng M (2018) Grade 2 pancreatic neuroendocrine tumors: overbroad scope of Ki-67 index according to MRI features. Abdom Radiol (NY). https://doi.org/10.1007/s00261-018-1573-5