Can we use MRI to detect clinically silent recurrent soft-tissue sarcoma?

Area of Science:

• Radiology
• Oncology
• Medical Imaging

Background:

• Early detection of local recurrence (LR) in high-grade soft-tissue sarcomas (STS) using MRI remains challenging.
• The diagnostic accuracy of MRI for identifying clinically silent LR needs further substantiation.

Purpose of the Study:

• To evaluate the contribution of MRI criteria, including dynamic contrast-enhanced (DCE) MRI, in detecting LR of STS.
• To assess how knowledge of surgical margins impacts the accuracy of MRI in identifying LR.
• To identify reasons for misdiagnosing LR in STS.

Main Methods:

• Retrospective analysis of MRI scans from 23 patients with STS LR and 22 matched controls without LR.
• Comparison of preoperative conventional and DCE-MRI characteristics with post-treatment MRIs before proven LR.
• Assessment of recurrence likelihood using a 5-point Likert scale, with and without knowledge of surgical margins, analyzed via ROC analysis.

Main Results:

• Combining conventional MRI, DCE-MRI, and surgical margin data yielded the highest accuracy (AUC 0.779) in differentiating LR from post-therapeutic changes.
• DCE-MRI alone achieved an AUC of 0.706, while conventional MRI had an AUC of 0.648.
• Suboptimal MRI technique and misinterpretation of microscopic positive margins were key causes of false results.

Conclusions:

• MRI, particularly DCE-MRI, enhances the detection of recurrent, clinically silent STS when integrated with surgical margin status.
• Inadequate MRI protocols can lead to missed LR diagnoses.
• Comprehensive MRI protocols including multiple orientations and comparison with initial tumor characteristics improve diagnostic value.