Advertisement

European Radiology

, Volume 29, Issue 3, pp 1074–1082 | Cite as

Radiomics nomogram for predicting the malignant potential of gastrointestinal stromal tumours preoperatively

  • Tao ChenEmail author
  • Zhenyuan Ning
  • Lili Xu
  • Xingyu Feng
  • Shuai Han
  • Holger R. Roth
  • Wei Xiong
  • Xixi Zhao
  • Yanfeng Hu
  • Hao Liu
  • Jiang Yu
  • Yu Zhang
  • Yong Li
  • Yikai Xu
  • Kensaku Mori
  • Guoxin LiEmail author
Gastrointestinal
  • 413 Downloads

Abstract

Objective

To develop and evaluate a radiomics nomogram for differentiating the malignant risk of gastrointestinal stromal tumours (GISTs).

Methods

A total of 222 patients (primary cohort: n = 130, our centre; external validation cohort: n = 92, two other centres) with pathologically diagnosed GISTs were enrolled. A Relief algorithm was used to select the feature subset with the best distinguishing characteristics and to establish a radiomics model with a support vector machine (SVM) classifier for malignant risk differentiation. Determinant clinical characteristics and subjective CT features were assessed to separately construct a corresponding model. The models showing statistical significance in a multivariable logistic regression analysis were used to develop a nomogram. The diagnostic performance of these models was evaluated using ROC curves. Further calibration of the nomogram was evaluated by calibration curves.

Results

The generated radiomics model had an AUC value of 0.867 (95% CI 0.803–0.932) in the primary cohort and 0.847 (95% CI 0.765–0.930) in the external cohort. In the entire cohort, the AUCs for the radiomics model, subjective CT findings model, clinical index model and radiomics nomogram were 0.858 (95% CI 0.807–0.908), 0.774 (95% CI 0.713–0.835), 0.759 (95% CI 0.697–0.821) and 0.867 (95% CI 0.818–0.915), respectively. The nomogram showed good calibration.

Conclusions

This radiomics nomogram predicted the malignant potential of GISTs with excellent accuracy and may be used as an effective tool to guide preoperative clinical decision-making.

Key Points

• CT-based radiomics model can differentiate low- and high-malignant-potential GISTs with satisfactory accuracy compared with subjective CT findings and clinical indexes.

• Radiomics nomogram integrated with the radiomics signature, subjective CT findings and clinical indexes can achieve individualised risk prediction with improved diagnostic performance.

• This study might provide significant and valuable background information for further studies such as response evaluation of neoadjuvant imatinib and recurrence risk prediction.

Keywords

Gastrointestinal stromal tumour Classification Radiomics Nomogram Machine learning 

Abbreviations

AFIP

Armed forces institute of pathology

AUC

Area under curve

CI

Confidence interval

EVFDM

Enlarged vessels feeding or draining the mass

GISTs

Gastrointestinal stromal tumours

ICCs

Inter- and intraclass correlation coefficients

NCCN

National comprehensive cancer network

NIH

National institutes of health

OR

Odds ratio

SVM

Support vector machine

Notes

Funding

This study has received funding by the State’s Key Project of Research and Development Plan (2017YFC0108300 and 2017YFC0108303) and JSPS KAKENHI Grant (17H00867 and 26108006).

Compliance with ethical standards

Guarantor

The scientific guarantor of this publication is Tao Chen.

Conflict of interest

The authors of this manuscript declare no relationships with any companies whose products or services may be related to the subject matter of the article.

Statistics and biometry

One of the authors has significant statistical expertise: Hao Liu, Department of General Surgery, Nanfang Hospital, Southern Medical University, Guangdong Provincial Engineering Technology Research Center of Minimally Invasive Surgery, Guangzhou 510515, Guangdong Province, China. Hao Liu has completed postdoctoral research of cancer epidemiological statistics in Heidelberg Cancer Center, Germany. Hao Liu specialises in statistical analysis and provided statistical advice in this study.

Informed consent

Written informed consent or substitute was obtained from all patients in this study.

Ethical approval

Institutional review board approval was obtained.

Methodology

• retrospective

• diagnostic or prognostic study

• multicentre study

Supplementary material

330_2018_5629_MOESM1_ESM.docx (220 kb)
ESM 1 (DOCX 219 kb)

References

  1. 1.
    Nishida T, Blay JY, Hirota S, Kitagawa Y, Kang YK (2016) The standard diagnosis, treatment, and follow-up of gastrointestinal stromal tumors based on guidelines. Gastric Cancer 19:3–14CrossRefGoogle Scholar
  2. 2.
    Joensuu H, Hohenberger P, Corless CL (2013) Gastrointestinal stromal tumour. Lancet 382:973–983CrossRefGoogle Scholar
  3. 3.
    Shah R, Jonnalagadda S (2005) The GIST of a stromal tumor. Gastroenterology 128:2170–21714CrossRefGoogle Scholar
  4. 4.
    Nishida T, Kawai N, Yamaguchi S, Nishida Y (2013) Submucosal tumors: comprehensive guide for the diagnosis and therapy of gastrointestinal submucosal tumors. Dig Endosc 25:479–489CrossRefGoogle Scholar
  5. 5.
    Joensuu H (2008) Risk stratification of patients diagnosed with gastrointestinal stromal tumor. Hum Pathol 39:1411–1419CrossRefGoogle Scholar
  6. 6.
    Miettinen M, Lasota J (2006) Gastrointestinal stromal tumors: pathology and prognosis at different sites. Semin Diagn Pathol 23:70–83CrossRefGoogle Scholar
  7. 7.
    Zhou C, Duan X, Zhang X, Hu H, Wang D, Shen J (2016) Predictive features of CT for risk stratifications in patients with primary gastrointestinal stromal tumour. Eur Radiol 26:3086–3093CrossRefGoogle Scholar
  8. 8.
    Burkill GJ, Badran M, Al-Muderis O et al (2003) Malignant gastrointestinal stromal tumor: distribution, imaging features, and pattern of metastatic spread. Radiology 226:527–532CrossRefGoogle Scholar
  9. 9.
    Liu Q, Li Y, Dong M, Kong F, Dong Q (2017) Gastrointestinal bleeding is an independent risk factor for poor prognosis in GIST patients. Biomed Res Int.  https://doi.org/10.1155/2017/7152406
  10. 10.
    Verma V, Simone CB 2nd, Krishnan S, Lin SH, Yang JZ, Hahn SM (2017) The rise of radiomics and implications for oncologic management. J Natl Cancer Inst.  https://doi.org/10.1093/jnci/djx055
  11. 11.
    Gillies RJ, Kinahan PE, Hricak H (2016) Radiomics: images are more than pictures, they are data. Radiology 278:563–577CrossRefGoogle Scholar
  12. 12.
    Wu S, Zheng J, Li Y et al (2017) A radiomics nomogram for the preoperative prediction of lymph node metastasis in bladder cancer. Clin Cancer Res 23:6904–6911CrossRefGoogle Scholar
  13. 13.
    Huang YQ, Liang CH, He L et al (2016) Development and validation of a radiomics nomogram for preoperative prediction of lymph node metastasis in colorectal cancer. J Clin Oncol 34:2157–2164CrossRefGoogle Scholar
  14. 14.
    Kononenko I (1994) Estimating attributes: analysis and extensions of RELIEF. In: Bergadano F, De Raedt L (eds) Machine Learning: ECML-94. ECML 1994. Lecture Notes in Computer Science (Lecture Notes in Artificial Intelligence), vol 784. Springer, Berlin, pp 171–182Google Scholar
  15. 15.
    Noble WS (2006) What is a support vector machine? Nat Biotechnol 24:1565–1567CrossRefGoogle Scholar
  16. 16.
    Brancatelli G, Federle MP, Grazioli L, Blachar A, Peterson MS, Thaete L (2001) Focal nodular hyperplasia: CT findings with emphasis on multiphasic helical CT in 78 patients. Radiology 219:61–68CrossRefGoogle Scholar
  17. 17.
    Iasonos A, Schrag D, Raj GV, Panageas KS (2008) How to build and interpret a nomogram for cancer prognosis. J Clin Oncol 26:1364–1370CrossRefGoogle Scholar
  18. 18.
    Han DS, Suh YS, Kong SH et al (2012) Nomogram predicting long-term survival after d2 gastrectomy for gastric cancer. J Clin Oncol 30:3834–3840Google Scholar
  19. 19.
    von Mehren M, Randall RL, Benjamin RS et al (2016) Soft tissue sarcoma, Version 2.2016, NCCN Clinical Practice Guidelines in Oncology. J Natl Compr Cancer Netw 14:758–786CrossRefGoogle Scholar
  20. 20.
    Nishida T, Goto O, Raut CP, Yahagi N (2016) Diagnostic and treatment strategy for small gastrointestinal stromal tumors. Cancer 122:3110–3118CrossRefGoogle Scholar
  21. 21.
    Tanaka J, Oshima T, Hori K et al (2010) Small gastrointestinal stromal tumor of the stomach showing rapid growth and early metastasis to the liver. Dig Endosc 22:354–356CrossRefGoogle Scholar
  22. 22.
    Lv A, Li Z, Tian X et al (2013) SKP2 high expression, KIT exon 11 deletions, and gastrointestinal bleeding as predictors of poor prognosis in primary gastrointestinal stromal tumors. PLoS One.  https://doi.org/10.1371/journal.pone.0062951
  23. 23.
    Ng F, Kozarski R, Ganeshan B, Goh V (2013) Assessment of tumor heterogeneity by CT texture analysis: can the largest cross-sectional area be used as an alternative to whole tumor analysis? Eur J Radiol 82:342–348CrossRefGoogle Scholar
  24. 24.
    Kickingereder P, Gotz M, Muschelli J et al (2016) Large-scale radiomic profiling of recurrent glioblastoma identifies an imaging predictor for stratifying anti-angiogenic treatment response. Clin Cancer Res 22:5765–5771CrossRefGoogle Scholar
  25. 25.
    Chapiro J, Duran R, Lin M et al (2015) Identifying staging markers for hepatocellular carcinoma before transarterial chemoembolization: comparison of three-dimensional quantitative versus non-three-dimensional imaging markers. Radiology 275:438–447CrossRefGoogle Scholar
  26. 26.
    Benjamin RS, Choi H, Macapinlac HA et al (2007) We should desist using RECIST, at least in GIST. J Clin Oncol 25:1760–1764CrossRefGoogle Scholar
  27. 27.
    Dudeck O, Zeile M, Reichardt P, Pink D (2011) Comparison of RECIST and Choi criteria for computed tomographic response evaluation in patients with advanced gastrointestinal stromal tumor treated with sunitinib. Ann Oncol 22:1828–1833CrossRefGoogle Scholar
  28. 28.
    Joensuu H, Wardelmann E, Sihto H et al (2017) Effect of KIT and PDGFRA mutations on survival in patients with gastrointestinal stromal tumors treated with adjuvant imatinib an exploratory analysis of a randomized clinical trial. JAMA Oncol 3:602–609CrossRefGoogle Scholar

Copyright information

© European Society of Radiology 2018

Authors and Affiliations

  1. 1.Department of General SurgeryNanfang Hospital, Southern Medical University, Guangdong Provincial Engineering Technology Research Center of Minimally Invasive SurgeryGuangzhouChina
  2. 2.Graduate School of Information ScienceNagoya UniversityNagoyaJapan
  3. 3.School of Biomedical EngineeringSouthern Medical UniversityGuangzhouChina
  4. 4.Medical Image Center, Nanfang HospitalSouthern Medical UniversityGuangzhouChina
  5. 5.Department of General SurgeryGuangdong General Hospital, Guangdong Academy of Medical ScienceGuangzhouChina
  6. 6.Department of General Surgery, Zhujiang HospitalSouthern Medical UniversityGuangzhouChina

Personalised recommendations