Atypical Lipomatous Tumours vs. Lipomas: A Multimodal Diagnostic Approach.

Area of Science:

• Oncology
• Radiology
• Pathology

Background:

• Differentiating atypical lipomatous tumors (ALTs) from lipomas is crucial for accurate diagnosis and treatment.
• Current diagnostic methods can be expensive, involving molecular pathology testing.
• There is a need for a more accessible and reliable diagnostic tool.

Purpose of the Study:

• To develop and validate a scoring system combining clinical and radiological parameters for distinguishing ALTs from lipomas.
• To improve diagnostic accuracy and reduce the need for molecular pathology testing.
• To create a practical risk calculator for clinical use.

Main Methods:

• Retrospective analysis of 188 patients with lipomatous tumors.
• Review of clinical data, pathology reports, and MRI imaging.
• Development and statistical evaluation of four predictive models using logistic regression and ROC analysis.

Main Results:

• The optimal model (Model 1) with seven parameters achieved an AUC of 0.952, demonstrating high predictive accuracy.
• Model 1 showed a sensitivity of 96.4% and a negative predictive value (NPV) of 97.2%.
• Contrast enhancement on MRI was identified as a significant parameter; a risk calculator was developed.

Conclusions:

• The developed scoring system effectively distinguishes ALTs from lipomas using clinical and imaging data.
• This multi-parameter approach offers a reliable and cost-effective alternative to molecular pathology testing.
• The system improves diagnostic reliability, minimizes misclassification, and potentially lowers healthcare costs.