Magnetic Resonance Imaging-based Radiomics Nomogram for Prediction of the Histopathological Grade of Soft Tissue Sarcomas: A Two-center Study

Abstract

Background: Preoperative prediction of the soft tissue sarcomas (STSs) grade is important for treatment decisions. To preoperatively distinguish low-grade (grades I and II) and high-grade (grade III) STSs, we developed and validated the performance of a magnetic resonance imaging (MRI)-based radiomics nomogram.Methods: Patients with an STS based on the French Federation of Cancer Centers Sarcoma Group grading system at two independent institutions were enrolled (training set, n = 109; external validation set, n = 71). The minimum redundancy maximum relevance method and least absolute shrinkage and selection operator logistic regression were used to process feature selection and radiomics signature development. Three radiomics signature models were constructed based on T1-weighted imaging (RS-T1 model) and fat-suppressed T2-weighted imaging sequences (RS-FST2 model) and their combination (RS-Combined model). Model performance (discrimination capability, calibration curve, and clinical usefulness) was evaluated in the external validation set. Results: The RS-T1 model, RS-FST2 model, and RS-Combined model achieved predictive abilities with area under the receiver operating characteristic curves (AUCs) of 0.645, 0.641, and 0.829, respectively, in the external validation set. The nomogram, incorporating significant clinical factors and the RS-Combined model, showed extremely high predictive ability in the training set and external validation set with AUCs of 0.916 (95% confidence interval, 0.866–0.966) and 0.879 (0.791–0.967), respectively. The nomogram achieved significant patient stratification.Conclusions: The proposed noninvasive MRI-based radiomics nomogram shows superior predictive performance in differentiating low-grade from high-grade STS.

Prognostic Assessment in High-Grade Soft-Tissue Sarcoma Patients: A Comparison of Semantic Image Analysis and Radiomics

“…For instance, tumor characterization in terms of molecular aberrations (“radiogenomics”) or histological properties could be a potential outcome target. Multiple authors demonstrated noninvasive prediction of the important prognostic factor, “tumor grading” [ 58 , 59 ]. In an ongoing work, we could demonstrate promising results differentiating benign lipomas from atypical lipomatous tumors based on the murine double minutes ( MDM2 ) gene amplification status.…”

Musculoskeletal Soft-Tissue Sarcoma: Quality Assessment of Initial MRI Reports Shows Frequent Deviation from ESSR Guidelines

“…MRI facilitates not only detection, but also reliable differentiation between low-grade and high-grade soft-tissue sarcoma [ 20 , 21 ]. However, with magnetic resonance imaging being the primary recommended imaging method for further diagnostics of suspicious soft-tissue masses, executing radiologists face the risk of neglecting sarcomas in their considerations due to their rareness [ 16 , 22 , 23 ].…”

An Update in Imaging Evaluation of Histopathological Grade of Soft Tissue Sarcomas Using Structural and Quantitative Imaging and Radiomics

“…In a study by Peeken et al, 54 the T2FS-based model showed the best reproducibility (AUC: 0.78) in an independent validation set (103 STSs); in addition, combining the T2FS radiomics model with clinical staging into nomograms improved the prognostic ability with a net clinical benefit over clinical staging. Yan et al 55 Although radiomics can facilitate in-depth characterization of spatially heterogeneous characteristics of tumor tissues, and offers unique advantages in predicting the histopathological grade of STSs, there are some limitations of the studies that have evaluated this methodology. First, due to the low prevalence of STSs, the sample size in these studies was quite limited.…”

“…Fourth, the tumor segmentation method is also a key point in radiomics, as the feature data are derived from the segmented volume, although many margins are not clear. [54][55][56][57][58][59][60][61][62][63][64][65][66][67][68][69] Compared to CT or PET/CT, MRI offers better tissue contrast, has multiplanar capabilities, and shows less radiation artifacts and bone sclerosis, which allows for more accurate delineation of tumor margins to be preserved.…”

“…In a study by Peeken et al, 54 the T2FS‐based model showed the best reproducibility (AUC: 0.78) in an independent validation set (103 STSs); in addition, combining the T2FS radiomics model with clinical staging into nomograms improved the prognostic ability with a net clinical benefit over clinical staging. Yan et al 55 constructed nomograms for 180 STSs confirmed by pathological findings (training set, N = 109; external validation set, N = 71) and performed radiomics feature extraction on T1‐weighted imaging and FS‐T2‐weighted imaging using the minimum redundancy maximum relevance (MRMR) method and the LASSO algorithm. Three radiomics signatures were constructed based on T1‐weighted image (RS‐T1 model), FS‐T2‐weighted image (RS‐FST2 model), and their combination (RS‐Combined model).…”