Background: Retinoblastoma is the most common intraocular tumor in childhood with a good prognosis in terms of mortality, but detailed information about tumor morphology and disease extent in retinoblastoma is important for treatment decision making.
Purpose: To demonstrate ultrahigh-field MRI tumor morphology and tumor extent in retinoblastoma correlating with in and ex vivo images with histopathology.
Study type: Prospective case series.
Population: Six retinoblastoma patients (median age 5.5 months, range 2-14) were prospectively included in this study. Median time between diagnosis and enucleation was 8 days (range 7-19).
Field strength/sequence: In vivo pre-enucleation at 1.5T MRI with a circular surface coil. Ex vivo imaging (FLASH T1 -weighted and RARE T2 -weighted) was performed at field strengths of 9.4T and 17.6T.
Assessment: After ex vivo imaging, the eyes were histopathologically analyzed and morphologically matched with MRI findings by three authors (two with respectively 14 and 4 years of experience in ocular MRI and one with 16 years of experience in ophthalmopathology).
Results: Small submillimeter morphological aspects of intraocular retinoblastoma were successfully depicted with higher-resolution MRI and matched with histopathology images. With ex vivo MRI a small subretinal tumor seed (300 μm) adjacent to the choroid was morphologically matched with histopathology. Also, a characteristic geographical pattern of vital tumor tissue (400 μm) surrounding a central vessel interspersed with necrotic areas correlated with histopathology images. Tumor invasion into the optic nerve showed a higher signal intensity on T1 -weighted higher-resolution MRI.
Data conclusion: Higher-resolution MRI allows for small morphological aspects of intraocular retinoblastoma and extraocular disease extent not visible on currently used clinical in vivo MRI to be depicted.
Level of evidence: 4 Technical Efficacy: Stage 2 J. Magn. Reson. Imaging 2018;47:1487-1497.
Keywords: 17.6 Tesla; 9.4 Tesla; MR microscopy; diagnostic accuracy; magnetic resonance imaging; morphology; optic nerve; retinoblastoma; tumor seeding.
© 2017 International Society for Magnetic Resonance in Medicine.