Objective: Children with cancer-predisposing conditions are at increased risk to develop and die of cancer. Limited data exist on the utility of whole-body MRI as a cancer screening tool in children. In this study, we examined the diagnostic performance of whole-body MRI as a mechanism of tumor surveillance for children at increased genetic risk for cancer.
Materials and methods: Twenty-four children (six boys and 18 girls) with a mean age of 11.2 years (range, 2.1-18.2 years) underwent 50 unenhanced whole-body MRI examinations over a 5-year period. Scans were retrospectively reviewed and assessed for image quality; sequences performed; and the presence of osseous, soft-tissue, or solid organ abnormalities. Findings suggestive of a malignancy were stratified by risk as low (< 20% chance for cancer), moderate (20-80%), or high (> 80%). MRI findings were correlated with medical records, biopsy results, or additional follow-up imaging; biopsy and follow-up were considered as the reference standards.
Results: Forty-eight of 50 (96%) examinations were of very good quality. Nine findings suspicious for malignancy were identified, including two high-risk, two moderate-risk, and five low-risk lesions. One high-risk lesion was proven by biopsy to be a papillary thyroid carcinoma, with the remaining lesions deemed nonmalignant. The sensitivity of whole-body MRI was 100%; specificity, 94%; positive predictive value, 25%; and negative predictive value (NPV), 100%.
Conclusion: Unenhanced whole-body MRI is safe and produces excellent images. The high sensitivity, specificity, and NPV make whole-body MRI a valuable cancer screening tool in children with a genetic predisposition for cancer.
Keywords: Li-Fraumeni syndrome; cancer predisposition; children; hereditary paraganglioma-pheochromocytoma syndrome; whole-body MRI.