Prospective observational study on diagnostic accuracy of whole-body MRI in solid small round cell tumours.

AIM: To assess the diagnostic accuracy of whole-body magnetic resonance imaging (WB-MRI) for metastatic disease in patients with solid small round cell tumours (SRCT) by comparing it with routine staging procedures (standard of care).

MATERIALS AND METHODS: Eligible cases of neuroblastoma, primitive neuroectodermal tumour, and rhabdomyosarcoma were enrolled in the study after obtaining informed consent. WB-MRI was undertaken using overlapping coronal T1 and short-tau inversion recovery (STIR) sequences. Lesions were classified into skeletal, pulmonary, and soft-tissue types. Conventional staging, which consisted of combined positron-emission tomography & computed tomography (PET-CT), bone scintigraphy & bone marrow biopsy for bone metastases, CT thorax for lung metastases, combined PET-CT, metaiodobenzylguanidine (MIBG) scintigraphy (in neuroblastoma) for soft tissue metastases and clinical evaluation was used as the reference standard. Parameters for diagnostic accuracy were calculated.

RESULTS: Thirty-four out of forty patients enrolled were included in final analysis, half of them having metastatic disease. The sensitivity, specificity, positive and negative predictive value, and the diagnostic accuracy of WB-MRI and PET-CT for skeletal metastases as compared to reference standard were 91.9%, 99.8%, 97.4%, 99.6%, and 95.5% and 99.1%, 99.9%, 99.1%, 99.9%, and 99.9%, respectively. The sensitivity of MRI, only PET and PET-CT with plain CT thorax was 30%, 40%, and 100%, respectively, for lung metastases. The sensitivity of MRI for soft-tissue lesions was 76.9%.

CONCLUSION: WB-MRI is a radiation-free tool with high diagnostic accuracy for the evaluation of metastatic disease to the marrow. The rate of detection of soft-tissue metastases, such as nodal metastases, is less when WB-MRI is compared with conventional staging using coronal STIR images. CT thorax is essential for accurate evaluation of lung metastases.