Whole-body MRI for staging and follow-up of primary musculoskeletal tumours: a systematic review.

Abstract

<h4>Objectives</h4>To evaluate the performance and technical parameters of whole body (WB)-MRI for staging and follow-up of primary musculoskeletal tumours.<h4>Methods</h4>A systematic review was done in PubMed and Embase through July 2025. Eligible studies reported WB-MRI for staging or follow-up of bone/soft tissue sarcomas. Extracted data were study design, patient characteristics, MRI protocols, scan duration, and diagnostic performance. Methodological quality was assessed with QualSyst.<h4>Results</h4>A total of 10 studies, published between 2016 and 2024, were included from 432 records. Most were retrospective (90%), with study populations ranging from 9 to 319 patients (total n = 790, age range 2-80 years). Half of the studies focused on myxoid liposarcoma, while others addressed osteosarcoma, Ewing sarcoma, and chondrosarcoma. WB-MRI protocols employed 1.5T and/or 3T scanners. Non-contrast protocols (8/10 studies) mostly included T1 and STIR sequences. Exam durations ranged from 30 to 78 minutes, with outliers up to 250 minutes. Reference standards included CT, PET-CT, and bone scintigraphy. Diagnostic accuracy was investigated in only 2 studies, reporting 100% sensitivity, 96.3% specificity, and 97.3% accuracy for extrapulmonary metastases, 83%-88% sensitivity and 94%-95% specificity for bone metastases. The studies demonstrated high methodological rigour, with scores ranging from 16 to 19 out of 20.<h4>Conclusions</h4>WB-MRI is a feasible and promising modality for staging and follow-up of primary musculoskeletal tumours. Evidence is still limited, based on heterogeneous and mostly retrospective studies. Larger, prospective, and standardized studies are needed to validate its accuracy, optimize imaging protocols and clarify its role in sarcoma imaging.<h4>Advances in knowledge</h4>WB-MRI has been tested scarcely as a comprehensive, non-ionizing alternative for whole-body staging in selected patients with primary musculoskeletal tumours, mostly using unenhanced T1 and STIR sequences, with limited use of contrast media. Current evidence is insufficient to recommend routine use of WB-MRI, standardized protocols and prospective validation are needed.