A weighted scoring system to differentiate malignant liposarcomas from benign lipomas

The value of a new diagnostic strategy for adipocytic soft tissue tumors in adults: A retrospective study

INTRODUCTION

The distinction between lipoma and well-differentiated liposarcoma (WDLPS), or "atypical lipomatous tumor" (ALT), is crucial as it impacts patient management. A group of European experts led by Benjamin Moulin recently issued a consensus report to define the role of radiology in managing these lesions. It describes an algorithm defining the criteria prompting a diagnostic biopsy of deep lipomatous tumors of the limbs and chest wall. The primary aim of this study was to evaluate the algorithm's diagnostic performance.

MATERIALS AND METHODS

Between 2012 and 2019, all biopsies of deep fatty tumors of the limbs or chest wall with a pre-biopsy MRI assessment were recorded at our institution. The MRI scans were reviewed by two radiologists. Each lesion was classified according to biopsy status by applying the algorithm of the European panel. The algorithm's diagnostic performance was assessed by calculating the sensitivity, specificity, positive predictive value and negative predictive value. Inter-rater agreement was also assessed.

RESULTS

Of the 156 tumors in our study, 148 (94.9%) required a biopsy, and the algorithm's sensitivity for detecting ALT/WDLPS was 100% with specificity of 6.3% and a PPV of 20.3%. Inter-rater agreement was almost perfect with a kappa value of 0.882.

CONCLUSION

The European algorithm demonstrates perfect sensitivity, an important criterion for a screening examination such as MRI in this setting. The algorithm's low specificity, however, emphasizes the need for further studies to redefine the optimum size cut-off value, especially for lesions without atypical criteria or an anatomical location at risk of post-surgical recurrence.

Prognostic value of clinical and MRI features in the screening of lipomatous lesions

BACKGROUND AND OBJECTIVES

Differentiation of lipomatous tumors mostly requires diagnostic biopsy but is essential to decide for the most adequate therapy. This study aims to investigate the prognostic value of available clinical and radiological features with regard to malignancy of the lesion, recurrence and survival.

METHODS

In this retrospective cohort study, 104 patients with a biopsy-proven lipomatous tumor between 2010 and 2015 and a minimum clinical follow-up of two years were enrolled. Next to clinical features (age, gender, location of the lesion, histopathologic diagnosis, stage of disease, time to recurrence and death), MRI parameters were recorded retrospectively and blinded to the histological diagnosis.

RESULTS

Malignant lipomatous tumors were associated with location in the lower extremities and MRI features like thick septation (>2 mm), presence of a non-adipose mass, foci of high T2/STIR signal and contrast agent enhancement. A non-adipose mass was a predictor for recurrence and inferior overall survival, while lesions with high T2/STIR signal showed higher risk of recurrence only. In combination, clinical and radiological features (lower extremities, septation > 2 mm, existence of non-adipose mass, contrast enhancement, and foci of high T2/STIR signal) predicted a malignant lipomatous tumor with an accuracy of 0.941 (95% CI of 0.899-0.983; 87% sensitivity, 86% specificity).

CONCLUSION

Localization and characteristic MR features predict malignancy in most lipomatous lesions. Non-adipose masses are a poor prognostic factor, being associated with tumor recurrence and disease-related death.

Tumor-to-bone distance and radiomic features on MRI distinguish intramuscular lipomas from well-differentiated liposarcomas

Background

To develop a machine learning model based on tumor-to-bone distance and radiomic features derived from preoperative MRI images to distinguish intramuscular (IM) lipomas and atypical lipomatous tumors/well-differentiated liposarcomas (ALTs/WDLSs) and compared with radiologists.

Methods

The study included patients with IM lipomas and ALTs/WDLSs diagnosed between 2010 and 2022, and with MRI scans (sequence/field strength: T1-weighted (T1W) imaging at 1.5 or 3.0 Tesla MRI). Manual segmentation of tumors based on the three-dimensional T1W images was performed by two observers to appraise the intra- and interobserver variability. After radiomic features and tumor-to-bone distance were extracted, it was used to train a machine learning model to distinguish IM lipomas and ALTs/WDLSs. Both feature selection and classification steps were performed using Least Absolute Shrinkage and Selection Operator logistic regression. The performance of the classification model was assessed using a tenfold cross-validation strategy and subsequently evaluated using the receiver operating characteristic curve (ROC) analysis. The classification agreement of two experienced musculoskeletal (MSK) radiologists was assessed using the kappa statistics. The diagnosis accuracy of each radiologist was evaluated using the final pathological results as the gold standard. Additionally, we compared the performance of the model and two radiologists in terms of the area under the receiver operator characteristic curves (AUCs) using the Delong’s test.

Results

There were 68 tumors (38 IM lipomas and 30 ALTs/WDLSs). The AUC of the machine learning model was 0.88 [95% CI 0.72–1] (sensitivity, 91.6%; specificity, 85.7%; and accuracy, 89.0%). For Radiologist 1, the AUC was 0.94 [95% CI 0.87–1] (sensitivity, 97.4%; specificity, 90.9%; and accuracy, 95.0%), and as to Radiologist 2, the AUC was 0.91 [95% CI 0.83–0.99] (sensitivity, 100%; specificity, 81.8%; and accuracy, 93.3%). The classification agreement of the radiologists was 0.89 of kappa value (95% CI 0.76–1). Although the AUC of the model was lower than of two experienced MSK radiologists, there was no statistically significant difference between the model and two radiologists (all P > 0.05).

Conclusions

The novel machine learning model based on tumor-to-bone distance and radiomic features is a noninvasive procedure that has the potential for distinguishing IM lipomas from ALTs/WDLSs. The predictive features that suggested malignancy were size, shape, depth, texture, histogram, and tumor-to-bone distance.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13018-023-03718-4.

Surgical resection of a giant intramuscular lipoma of the biceps brachii: a case report and review of the literature

Lipomas are frequent benign soft-tissue tumors mainly located in the subcutis. Occasionally, subfascial or inter- or intramuscular lipomas are encountered. This case report describes the surgical management of a very rare giant intramuscular lipoma of the right biceps brachii muscle in a 71-year-old male patient. Preoperative magnetic resonance imaging as well as intraoperative photographs depict the complexity of this case and aid in the discussion about indications for surgery, as management guidelines of these deep-seated tumors are still inconsistent.

Diagnosis strategy of adipocytic soft-tissue tumors in adults: a consensus from European experts

Fat-containing tumors are very commonly found in daily practice with benign lipoma accounting for the majority of superficial tumors. Overlap in imaging findings between benign and intermediate or malignant fat-containing tumor is common. Guidelines recommend a core needle biopsy (CNB) for all deep tumors, and superficial tumors over 3 cm. However, specific strategy for diagnosis and referral to a sarcoma center should be applied on adipocytic tumors. The aim of this consensus statement is to provide a specific algorithm for adipocytic tumors, to discriminate patients who do require a CNB for preoperative diagnosis from those who can simply undergo active surveillance or a simple excision.

MRI-based radiomics to predict lipomatous soft tissue tumors malignancy: a pilot study

Objectives

To develop and validate a MRI-based radiomic method to predict malignancies in lipomatous soft tissue tumors.

Methods

This retrospective study searched in the database of our pathology department, data from patients with lipomatous soft tissue tumors, with histology and gadolinium-contrast enhanced T₁w MR images, obtained from 56 centers with non-uniform protocols. For each tumor, 87 radiomic features were extracted by two independent observers to evaluate the inter-observer reproducibility. A reduction of learning base dimension was performed from reproducibility and relevancy criteria. A model was subsequently prototyped using a linear support vector machine to predict malignant lesions.

Results

Eighty-one subjects with lipomatous soft tissue tumors including 40 lipomas and 41 atypical lipomatous tumors or well-differentiated liposarcomas with fat-suppressed T₁w contrast enhanced MR images available were retrospectively enrolled. Based on a Pearson’s correlation coefficient threshold at 0.8, 55 out of 87 (63.2%) radiomic features were considered reproducible. Further introduction of relevancy finally selected 35 radiomic features to be integrated in the model. To predict malignant tumors, model diagnostic performances were as follow: AUROC = 0.96; sensitivity = 100%; specificity = 90%; positive predictive value = 90.9%; negative predictive value = 100% and overall accuracy = 95.0%.

Conclusion

This work demonstrates that radiomics allows to predict malignancy in soft tissue lipomatous tumors with routinely used MR acquisition in clinical oncology. These encouraging results need to be further confirmed in an external validation population.

Lipomatous Soft Tissue Masses: Challenging the Paradigm of Routine Preoperative Biopsy

BACKGROUND

Lipomatous masses are the most common soft tissue tumors. While the majority are benign lipomas, it is important to identify those masses that are malignant prior to excision. Current guidelines recommend core needle biopsy (CNB) for all lipomatous masses larger than 3-5 cm. The objective of this study was to determine if routine preoperative CNB based on mass size is necessary, or if radiographic features can guide the need for CNB.

MATERIALS AND METHODS

Patients who underwent excision of extremity or truncal lipomatous masses at a single institution from October 2014 to July 2017 were retrospectively reviewed. By protocol, preoperative imaging was routinely obtained for all masses larger than 5 cm. High-risk radiographic features (intramuscular location, septations, nonfat nodules, heterogeneity, and ill-defined margins) and surgical pathology were evaluated to determine patients most likely to benefit from preoperative CNB.

RESULTS

Of 178 patients, 2 (1.1%) had malignant tumors on surgical pathology. All masses smaller than 5 cm were benign and, if imaging was obtained, had two or fewer high-risk radiographic features. Both of the patients with malignant tumors had masses larger than 5 cm, preoperative imaging that showed at least four high-risk radiographic features, and underwent CNB prior to excision.

CONCLUSIONS

The overall rate of malignancy is very low. The results of this study suggest that lipomatous masses smaller than 5 cm without concerning clinical characteristics do not require preoperative imaging or CNB. Conversely, lipomatous masses larger than 5 cm should undergo routine MRI with subsequent CNB if multiple high-risk radiographic features are present.

Liposarcoma Preoperatively Diagnosed as Lipoma: 10-Year Experience at a Single Institution

BACKGROUND

On rare occasions, a lesion preoperatively diagnosed as a lipoma is ultimately diagnosed as a liposarcoma. It is important to differentiate liposarcomas from lipomas preoperatively.

OBJECTIVE

To examine characteristic features of liposarcomas preoperatively diagnosed as lipomas.

METHODS

Patients (n = 637) who underwent resection of tumors preoperatively diagnosed as lipomas from January 2006 to October 2016 were retrospectively reviewed.

RESULTS

Based on pathological examination, 8 of 637 lesions were diagnosed as liposarcomas postoperatively. All the liposarcomas were well-differentiated liposarcomas. The rate of male patients was higher (87.5% vs 38.9%) and the size of tumors was larger (8.75 vs 4.64 cm) in these cases than in accurately diagnosed lipoma cases. On imaging, nonfatty septa were more frequently observed (71.4% vs 20.0%) and were thicker (2.22 vs 1.33 mm) than in true lipoma cases.

CONCLUSION

If the patient with a lipomatous tumor is male and the tumor is large, we should consider the possibility of it being a liposarcoma. A thick internal septum in the image findings is a good predictor of malignancy.

The Value of MRI in Distinguishing Subtypes of Lipomatous Extremity Tumors Needs Reassessment in the Era of MDM2 and CDK4 Testing

Introduction

Extremity lipomas and well-differentiated liposarcomas (WDLs) are difficult to distinguish on MR imaging. We sought to evaluate the accuracy of MRI interpretation using MDM2 amplification, via fluorescence in-situ hybridization (FISH), as the gold standard for pathologic diagnosis. Furthermore, we aimed to investigate the utility of a diagnostic formula proposed in the literature.

Methods

We retrospectively collected 49 patients with lipomas or WDLs utilizing MDM2 for pathologic diagnosis. Four expert readers interpreted each patient's MRI independently and provided a diagnosis. Additionally, a formula based on imaging characteristics (i.e. tumor depth, diameter, presence of septa, and internal cystic change) was used to predict the pathologic diagnosis. The accuracy and reliability of imaging-based diagnoses were then analyzed in comparison to the MDM2 pathologic diagnoses.

Results

The accuracy of MRI readers was 73.5% (95% CI 61-86%) with substantial interobserver agreement (κ=0.7022). The formula had an accuracy of 71%, which was not significantly different from the readers (p=0.71). The formula and expert observers had similar sensitivity (83% versus 83%) and specificity (64.5% versus 67.7%; p=0.659) for detecting WDLs.

Conclusion

The accuracy of both our readers and the formula suggests that MRI remains unreliable for distinguishing between lipoma and WDLs.