Low-level copy gain versus amplification of myc oncogenes in medulloblastoma: utility in predicting prognosis and survival

OBJECT: Medulloblastoma (MB) is a malignant embryonal tumor of the cerebellum. Amplification of c-myc or N-myc is infrequently identified and, when present, is often associated with the large cell/anaplastic (LC/A) phenotype. The frequency of low-level copy gain of myc oncogenes and its relationship to prognosis of MB has not been explored. METHODS: Archival cases of MB were histologically reviewed and classified into 3 major subtypes: classic, nodular, and LC/A. Using quantitative real-time polymerase chain reaction (PCR), the authors analyzed 58 cases with a pure histological subtype for the copy number (CN) of myc (c-myc and N-myc) oncogenes. Cases with > 5-fold CN were further analyzed using the fluorescent in situ hybridization (FISH) assay. Kaplan-Meier survival analysis was performed. RESULTS: A > 5-fold myc CN was noted in 5 (20.8%) of 24 LC/A, 1 (5.3%) of 19 classic, and 2 (13.3%) of 15 nodular subtypes. In a significant number of tumors (14 [56%] of 24 LC/A, 13 [68%] of 19 classic, and 10 [67%] of 15 nodular MBs) the CN was > 2-fold but < 5-fold. High-level amplification, defined as > 10-fold CN, was only seen in the LC/A subtype (5 cases), although moderate amplification (> 5-fold but < 10-fold) could be detected in other histological subtypes. Fluorescence in situ hybridization readily detected most cases corresponding to tumors with > 5-fold amplicon CN by quantitative real-time PCR, and could detect all 5 cases with > 10-fold CN by quantitative real-time PCR. The group of patients with > 5-fold myc amplicon CN showed significantly shorter survival than those with < 5-fold CN (p = 0.045), independent of histological subtype. CONCLUSIONS: Since FISH could easily detect most cases in the moderate-to-high myc gene amplification (> 5-fold CN) group, the FISH assay has utility in detecting subsets of MB with poorer prognosis.