Tumor grading of soft tissue sarcomas using MRI-based radiomics
| Autor: | Matthew J. Nyflot, Matthew B. Spraker, Mohamed A. Shouman, Rüdiger von Eisenhart-Rothe, Jan C. Peeken, Armin Ott, Michal Devecka, Fridtjof Nüsslin, Nina A. Mayr, Stephanie E. Combs, Hendrik Dapper, Carolin Knebel, Daniela Pfeiffer, Ahmed Thamer |
|---|---|
| Rok vydání: | 2019 |
| Předmět: |
Male
0301 basic medicine Research paper MRI Magnetic resonance imaging ROC receiver operator characteristic 0302 clinical medicine Image Processing Computer-Assisted T1FSGd Contrast-enhanced T1-weighted fat saturated MRI sequence Soft tissue sarcoma medicine.diagnostic_test Soft tissue Sarcoma Multimodal therapy FNCLCC French Federation of Cancer Centers Sarcoma Group General Medicine Magnetic Resonance Imaging Tumor grading ddc PPV Positiv predictive value 030220 oncology & carcinogenesis Female Radiology MRI medicine.medical_specialty MCC Maximum correlation coefficient C-index Concordance index STS Soft tissue sarcomas General Biochemistry Genetics and Molecular Biology T2FS Fat-saturated T2-weighted MRI sequence 03 medical and health sciences medicine Humans NPV Negative predictive value Radiometry Grading (tumors) Risk stratification Neoplasm Staging Radiomics Receiver operating characteristic business.industry Magnetic resonance imaging Biomarker Nomogram AUC Area under the curve medicine.disease OR odds ratio Nomograms 030104 developmental biology ROC Curve GLCM Gray level co-occurrence matrix OS Overall survival GLDM Gray level dependence matrix Soft Tissue Sarcoma Tumor Grading Mri Risk Stratification Neoplasm Grading 95%CI 95% confidence interval business |
| Zdroj: | EBioMedicine 48, 332-340 (2019) EBioMedicine |
| ISSN: | 2352-3964 |
| DOI: | 10.1016/j.ebiom.2019.08.059 |
| Popis: | Background Treatment decisions for multimodal therapy in soft tissue sarcoma (STS) patients greatly depend on the differentiation between low-grade and high-grade tumors. We developed MRI-based radiomics grading models for the differentiation between low-grade (G1) and high-grade (G2/G3) STS. Methods The study was registered at ClinicalTrials.gov (number NCT03798795 ). Contrast-enhanced T1-weighted fat saturated (T1FSGd), fat-saturated T2-weighted (T2FS) MRI sequences, and tumor grading following the French Federation of Cancer Centers Sarcoma Group obtained from pre-therapeutic biopsies were gathered from two independent retrospective patient cohorts. Volumes of interest were manually segmented. After preprocessing, 1394 radiomics features were extracted from each sequence. Features unstable in 21 independent multiple-segmentations were excluded. Least absolute shrinkage and selection operator models were developed using nested cross-validation on a training patient cohort (122 patients). The influence of ComBatHarmonization was assessed for correction of batch effects. Findings Three radiomic models based on T2FS, T1FSGd and a combined model achieved predictive performances with an area under the receiver operator characteristic curve (AUC) of 0.78, 0.69, and 0.76 on the independent validation set (103 patients), respectively. The T2FS-based model showed the best reproducibility. The radiomics model involving T1FSGd-based features achieved significant patient stratification. Combining the T2FS radiomic model into a nomogram with clinical staging improved prognostic performance and the clinical net benefit above clinical staging alone. Interpretation MRI-based radiomics tumor grading models effectively classify low-grade and high-grade soft tissue sarcomas. Fund The authors received support by the medical faculty of the Technical University of Munich and the German Cancer Consortium. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|