Object: Radiation-associated meningiomas (RAMs) arise after treatment with radiation to the cranium and are recognized as clinically separate from sporadic meningiomas. Compared with their sporadic counterparts, RAMs are often aggressive or malignant, likely to be multiple, and have a high recurrence rate. However, limited information exists about the clinical, pathological, and cytogenetic features of RAMs in pediatric patients. The authors report the findings in 9 children with meningiomas following therapeutic radiation to the cranium. In addition, they performed a critical review of the English language literature on pediatric RAMs.

Methods: Medical files were searched for patients who demonstrated meningiomas after a history of radiation to the brain. Only those patients in whom a meningioma occurred before the age of 18 years were included in this study. Clinical and demographic data along with the MIB-1 labeling index and cytogenetic studies were evaluated.

Results: The patients consisted of 5 males and 4 females with a median age of 5 years (range 2-10 years) at radiation therapy. The latency period was a median of 10 years after radiation therapy (range 6-13 years). The MIB-1 labeling index was a median of 6.6% (range 4%-10%). Five patients (55.6%) displayed multiple meningiomas at the first presentation. Histological types included clear cell meningioma in 1 patient, fibroblastic meningioma in 2, chordoid meningioma in 2, meningothelial meningioma in 7 (atypical in 2 cases), xanthomatous meningioma in 1, and chordoid meningioma in 1. Cytogenetic studies showed that the loss of 22q12.2 was the most common abnormality (3 patients), followed by complex cytogenetic abnormalities (2 patients) and rearrangements between chromosomes 1 and 12 (1 patient) and a 1p deletion (1 patient).

Conclusions: In contrast to RAMs occurring in adults, those in pediatric patients show an increased incidence of multiplicity on first presentation and unusual histological variants, some of which are described here for the first time. There was no difference in the MIB-1 labeling index in children with RAMs as compared with that in children with non-RAMs.