Ability of FDG PET and CT radiomics features to differentiate between primary and metastatic lung lesions
To evaluate the ability of CT and PET radiomics features to classify lung lesions as primary or metastatic, and secondly to differentiate histological subtypes of primary lung cancers.
A cohort of 534 patients with lung lesions were retrospectively studied. Radiomics texture features were extracted using the LIFEx package from semiautomatically segmented PET and CT images. Histology data were recorded in all patients. The patient cohort was divided into a training and a validation group and linear discriminant analysis (LDA) was performed to classify the lesions using both direct and backward stepwise methods. The robustness of the procedure was tested by repeating the entire process 100 times with different assignments to the training and validation groups. Scoring metrics included analysis of the receiver operating characteristic curves in terms of area under the curve (AUC), sensitivity, specificity and accuracy.
Radiomics features extracted from CT and PET datasets were able to differentiate primary tumours from metastases in both the training and the validation group (AUCs 0.79 ± 0.03 and 0.70 ± 0.04, respectively, from the CT dataset; AUCs 0.92 ± 0.01 and 0.91 ± 0.03, respectively, from the PET dataset). The AUC cut-off thresholds identified by LDA using direct and backward elimination strategies were −0.79 ± 0.06 and −0.81 ± 0.08, respectively (CT dataset) and −0.69 ± 0.05 and −0.68 ± 0.04, respectively (PET dataset). For differentiation between primary subgroups based on CT features, the AUCs in the training and validation groups were 0.81 ± 0.02 and 0.69 ± 0.04 for adenocarcinoma (Adc) vs. squamous cell carcinoma (Sqc) or “Other”, 0.85 ± 0.02 and 0.70 ± 0.05 for Sqc vs. Adc or Other, and 0.77 ± 0.03 and 0.57 ± 0.05 for Other vs. Adc or Sqc. The same analyses for the PET data revealed AUCs of 0.90 ± 0.10 and 0.80 ± 0.04, 0.80 ± 0.02 and 0.61 ± 0.06, and 0.97 ± 0.01 and 0.88 ± 0.04, respectively.
PET radiomics features were able to differentiate between primary and metastatic lung lesions and showed the potential to identify primary lung cancer subtypes.
KeywordsRadiomics Texture analysis, lung cancer Lung metastases Solitary lung nodule CT PET/CT Lung adenocarcinoma, squamous cell carcinoma
We thank Elena Vanni for support in patient selection; Paola Bossi and Dahoud Rahal for collaboration in pathological analyses; Marco Alloisio, Giulia Veronesi and the Thoracic Surgery Unit for close collaboration in patient selection and follow-up; Lorenzo Leonardi for image processing; and Riccardo Muglia, Nicolò Gennaro and Orazio Giuseppe Santonocito for their help in patient selection. Tommaso Cozzi is acknowledged for support in the processing of radiomics features. M.K. is supported by an AIRC (Italian Association for Cancer Research) scholarship funded by a grant won by A.C. (IG-2016-18585).
M.S., M.K. and A.C. conceived the idea of the study; L.C. and A.F. performed the statistical analysis; E.V. collected the data and selected the patients; M.S., M.K. and L.A. reviewed and segmented the images; L.C. performed the radiomics image analysis; M.S., M.K. and L.C. wrote the manuscript; A.R. edited and reviewed the manuscript.
All the authors discussed the results and commented on the manuscript.
Compliance with ethical standards
Conflicts of interest
A. Chiti received speaker honoraria from General Electric and Sirtex Medical System, acted as scientific advisor to Blue Earth Diagnostics and Advanced Accelerator Applications, and benefited from an unconditional grant from Sanofi to Humanitas University. All honoraria and grants are outside the scope of the submitted work.
L. Cozzi acts as Scientific Advisor to Varian Medical Systems. All honoraria are outside the scope of the submitted work.
M. Kirienko is supported by an AIRC (Italian Association for Cancer Research) scholarship funded by a grant won by A.C. (IG-2016-18,585).
All other authors have no conflicts of interest.
The study was approved by the institutional Ethics Committee. All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the principles of the 1964 Declaration of Helsinki and its later amendments or comparable ethical standards.
For this type of study (i.e. retrospective), formal consent was not required.
- 2.van Gómez López O, García Vicente AM, Honguero Martínez AF, Jiménez Londoño GA, Vega Caicedo CH, León Atance P, et al. (18)F-FDG-PET/CT in the assessment of pulmonary solitary nodules: comparison of different analysis methods and risk variables in the prediction of malignancy. Transl Lung Cancer Res. 2015;4:228–35.PubMedPubMedCentralGoogle Scholar
- 4.Dabrowska M, Krenke R, Korczynski P, Maskey-Warzechowska M, Zukowska M, Kunikowska J, et al. Diagnostic accuracy of contrast-enhanced computed tomography and positron emission tomography with 18-FDG in identifying malignant solitary pulmonary nodules. Medicine (Baltimore). 2015;94:e666.CrossRefGoogle Scholar
- 5.Kikano GE, Fabien A, Schilz R. Evaluation of the solitary pulmonary nodule. Am Fam Phys. 2015;92:1084–91.Google Scholar
- 20.Zwanenburg A, Leger S, Vallières M, Löck S; Image Biomarker Standardisation Initiative. Image biomarker standardisation initiative (arXiv:1612.07003). Cornell University Library: arXiv. http://arxiv.org/abs/1612.07003
- 22.Nicole C, Orlhac F, Boughdad S, Reuze S, Soussan M, Robert C, et al. A freeware for tumor heterogeneity characterization in PET, SPECT, CT, MRI and US to accelerate advances in radiomics [abstract]. J Nucl Med. 2017;58:1316.Google Scholar
- 30.Budiawan H, Cheon GJ, Im HJ, Lee SJ, Paeng JC, Kang KW, et al. Heterogeneity analysis of 18F-FDG uptake in differentiating between metastatic and inflammatory lymph nodes in adenocarcinoma of the lung: comparison with other parameters and its application in a clinical setting. Nucl Med Mol Imaging. 2013;47:232–41.CrossRefPubMedPubMedCentralGoogle Scholar
- 35.Gao X, Chu C, Li Y, Lu P, Wang W, Liu W, et al. The method and efficacy of support vector machine classifiers based on texture features and multi-resolution histogram from 18F-FDG PET-CT images for the evaluation of mediastinal lymph nodes in patients with lung cancer. Eur J Radiol. 2015;84:312–7.CrossRefPubMedGoogle Scholar
- 36.Kirienko M, Gallivanone F, Sollini M, Veronesi G, Voulaz E, Antunovic L, et al. Texture analysis of FDG-PET/CT images to characterize lung cancer lesions and to predict outcome in patients undergoing primary surgery: preliminary results. Eur J Nucl Med Mol Imaging. 2016;43:306.Google Scholar