Cancer Health Center

Note: Some citations in the text of this section are followed by a level of evidence. The PDQ editorial boards use a formal ranking system to help the reader judge the strength of evidence linked to the reported results of a therapeutic strategy. (Refer to the PDQ summary on Levels of Evidence for more information.)

Risk Stratification for Medulloblastoma

On the basis of neuroradiographic evaluation for disseminated disease, cerebrospinal fluid (CSF) cytological examination, postoperative neuroimaging evaluation for the amount of residual disease, age of the patient, and impression of the surgeon at the time of surgery, patients with medulloblastoma have been historically stratified into the following two risk groups:

• Average risk: Children older than 3 years with tumors that are totally resected or near-totally resected (<1.5 cm of residual disease) and no metastatic disease.[1]
• High risk: Children aged 3 years and younger or those with metastatic disease and/or subtotal resection (>1.5 cm of residual disease).[1]

The 1.5 cm standard was arbitrarily chosen for evaluation in prospective studies. Metastatic disease includes neuroradiographic evidence of disseminated disease, positive cytology in lumbar or ventricular CSF obtained more than 7 days postsurgery, or extraneural disease.[2]

An intense area of study is the use of molecular and immunohistochemical analysis for disease stratification. In a large, multi-institution, retrospective biologic study, 397 medulloblastoma specimens were analyzed by transcriptional profiling (103 specimens) and immunohistochemical analysis.[3] Four unique clusters of medulloblastomas were identified. The following two subgroups clearly stood out:

• A subset with WNT signaling that occurred in those with monosomy 6 abnormalities and beta-catenin nuclear staining. This subset was primarily older children and adolescents and had a female predominance.
• A subset driven by sonic hedgehog (SHH) signaling, arising predominantly, but not exclusively, in those with desmoplastic tumors and having immunohistochemical staining of SHH pathway elements. This subset consisted of infants, adolescents, and adults.

Patients from both of these subgroups had excellent prognoses; 80% or greater 5-year progression-free survival (PFS) and overall survival (OS). The remainder of medulloblastomas, which comprise the majority of the study cohort, were separable into two other subgroups, which had greater overlap and shared a poorer prognosis, including a 30% to 45% chance of dissemination. The latter two subsets (termed Group C and Group D) differed in the following ways:

• Immunohistochemical staining.
• Histological appearance (Group C was more likely anaplastic).
• MYC amplifications (found only in Group C).
• Likelihood of metastasis (higher in Group C)
• Age at presentation (Group C occurred more commonly in childhood, whereas Group D occurred in childhood, adolescence, and adulthood).
• Overall survival (lower in Group C).

PFS for patients with Group C and D tumors ranged between 32% and 63%.[3] Other studies have confirmed the excellent prognosis of children with WNT tumors and poor prognosis of those with tumors with MYC amplification.[4,5] If this type of molecular and immunohistochemical separation can be confirmed in a homogeneously treated population of patients, preferably studied prospectively, it will dramatically alter disease stratification and likely therapy offered.[6]