November-December 2017

Imaging Biomarkers for Adult Medulloblastomas: Genetic Entities May Be Identified by Their MR Imaging Radiophenotype

Keil Pic

Vera Keil

First manifestations of medulloblastoma at an adult age are extremely rare. With an estimated incidence as low as 0.6 new adult cases per million, many physicians are not even aware that medulloblastomas in adults are not always recurrent pediatric cases. Like the pediatric entity, the integrative 2016 WHO classification of CNS tumors divides adult medulloblastomas into several histologic and genetic subgroups. However, there is evidence that adult histologic and genetic medulloblastoma subgroups have different prognostic profiles than their pediatric counterparts. So how else are they different?

Adult medulloblastoma research is not exactly in the limelight, though patients with rare diseases deserve the best medical treatment just as anyone else. As a neuroradiologist still in training, I am therefore very delighted to have been given the opportunity to evaluate an entire cohort of adults with medulloblastoma for MR imaging features, or so-called MR biomarkers. These may facilitate an MR-based differentiation of medulloblastoma genetic subgroups. The cohort was established within the long-term, prospective NOA-07 trial, the German multicenter study dedicated exclusively to optimized diagnostics and therapeutic treatment of adult medulloblastomas (NCT01614132).  

Our group analyzed pretherapeutic MRI of all patients with a multiparametric imaging feature set. Indeed, despite the relatively small cohort of 28 patients, we identified several criteria, such as the presence of hemorrhage or the relative volume of perifocal edema, which may serve as MR biomarkers to differentiate genetic medulloblastoma subgroups. An interesting point of discussion is the complete divergence between the MR biomarkers identified in this adult cohort and those identified in a pediatric cohort with an identical study rationale published about 3 years ago in AJNR.1 Is it an arbitrary effect of cohort size or indeed another confirmatory indicator for a distinct evolution of adult medulloblastomas?

The important role neuroradiology plays in the concept of integrative, interdisciplinary neuro-oncologic diagnostics and therapy cannot be underestimated. State-of-the-art diagnostics demand genetic tumor profiling, but one should not forget that this (for numerous reasons) is not available everywhere. MR biomarkers can make a change in this regard by allowing noninvasive tumor profiling. Indeed, this is where the radiogenomic approach, as the emerging diagnostic field of big data analysis, connects with individualized medicine.

After the publication of this article and during conference presentations, I received a lot of positive feedback, particularly from those colleagues who cannot rely on genetic sequencing for their work and indeed need substitute methods like MR biomarkers. At this point, I would like to take the opportunity to advocate for large, multicenter studies in neuroradiology for rare diseases; the smaller the incidence, the bigger our joint effort as an interdisciplinary community must be to gain reliable results in this field.

Our next step within the NOA-07 research group will therefore be a further extension of our own cohort, but we are simultaneously running a long-term follow-up imaging study to identify early imaging markers of tumor recurrence.

Read this article at …


  1. Perreault S, Ramaswamy V, Achrol AS, et al. MRI surrogates for molecular subgroups of medulloblastomaAJNR Am J Neuroradiol 2014;35:1263–69, 10.3174/ajnr.A3990