Abstract
Medulloblastoma (MB) is composed of four molecular subgroups viz. WNT, SHH, groups 3 and 4, identified using various high-throughput methods. Translation of this molecular data into pathologist-friendly techniques that would be applicable in laboratories all over the world is a major challenge. Ninety-two MBs were analyzed using a panel of 10 IHC markers, real-time PCR for mRNA and miRNA expression, and FISH for MYC amplification. β-catenin, GAB1 and YAP1 were the only IHC markers of utility in classification of MBs into three subgroups viz. WNT (9.8%), SHH (45.6%) and non-WNT/SHH (44.6%). mRNA expression could further classify some non-WNT/SHH tumors into groups 3 and 4. This, however, was dependent on integrity of RNA extracted from FFPE tissue. MYC amplification was seen in 20% of non-WNT/SHH cases and was associated with the worst prognosis. For routine diagnostic practice, we recommend classification of MBs into three subgroups: WNT, SHH and non-WNT/SHH, with supplementation by prognostic markers like MYC for non-WNT/SHH tumors. Using this panel, we propose a new three-tier risk stratification system for MBs. Molecular subgrouping with this limited panel is rapid, economical, works well on FFPE tissue and is reliable as it correlates significantly with clinicopathological parameters and patient survival.