MBRS-34. MOLECULAR CHARACTERIZATION OF RECURRENT MEDULLOBLASTOMA REVEALS AN UNEXPECTEDLY HIGH INCIDENCE OF SECONDARY MALIGNANCY. Issue 2 (22nd June 2018)

• Record Type: Journal Article
• Authors: Kumar, Rahul; Deng, Maximilian; Rudneva, Vasilisa A; Robinson, Giles W; Dhall, Girish; Klesse, Laure J; Bowers, Daniel C; Chintagumpala, Murali; Leary, Sarah; Orr, Brent A; Pfister, Stefan M; Gajjar, Amar; Jones, David T W; Northcott, Paul A
• Is Part Of: Neuro-oncology. Volume 20:Issue 2(2018)supplement 2
• Page Start: i135
• Page End: i135
• Publication Date: 2018-06-22
• Subjects: Brain Neoplasms -- Periodicals; Brain -- Tumors -- Periodicals; Brain -- Cancer -- Periodicals; Nervous system -- Cancer -- Periodicals; 616.99481
• DOI: 10.1093/neuonc/noy059.479
• Languages: English
• ISSNs: 1522-8517

Abstract

Abstract: INTRODUCTION: Despite an overall 5-year progression free survival of approximately 80% for primary medulloblastomas (MBs), recurrent disease confers an abysmal prognosis with less than 10% 5-year survival. Subgroup affiliation at recurrence has been reported as stable though divergent clonal selection may explain treatment failure upon relapse. The confirmation of histopathologic diagnosis at recurrence thus represents a key task to facilitate selection of efficacious therapy and to gain deeper understanding of the biology driving recurrent disease. METHODS: DNA methylation-based classification of pediatric brain tumors has emerged as a robust methodology for refining histopathologic diagnosis and defining disease-specific subgroups. We leveraged Illumina Infinium BeadChip arrays to characterize a series of >50 patient-matched, histopathologically diagnosed, primary/relapse MB pairs. Entity and subgroups were assigned by comparison with a database consisting of >2800 reference methylation profiles (https://www.molecularneuropathology.org/mnp ). Genome-wide copy-number aberrations were inferred from the BeadChip arrays, and mutational analysis (whole exome sequencing) was performed on the majority of samples. RESULTS/CONCLUSION: Molecular classification was performed on n=59 primary/relapse MB pairs, inferring the following subgroup representation amongst our cohort: WNT (n=1, 2%), SHH (n=17, 30%), Group 3 (n=15, 25%), Group 4 (n=25, 42%). MB subgroup status was highly conserved across the series (90%), consistent with prior reports. Notably, we determined that 6/59 (10%) of histopathologically diagnosed MB 'relapses' molecularly classified as non-MB, including multiple glioblastomas (3/6). Pairwise copy-number and mutational analyses are ongoing to determine the degree of conservation and/or divergence amongst somatic alterations between primary/relapse MB pairs and to identify molecular patterns associated with secondary malignancy.