Biomarkers on the Brain

Medulloblastoma – the most common malignant childhood brain tumor –originates in the cerebellum and comprises various disease entities that differ in their characteristic mutations and biological profiles. The most effective current treatments for medulloblastoma are based on multi-modal surgery, radiotherapy, and chemotherapy. Despite the adaptation of treatment intensity to reflect disease risk, quantification of that risk is imprecise. About 60 percent of all patients – termed the standard risk group – receive essentially the same treatment. This subgroup, which is characterized by the absence of high-risk features, such as metastatic disease and large-cell histology, has a five-year survival rate of 75–85 percent. How could we improve upon those numbers? One possible solution was to search for ways of better personalizing treatment within this standard-risk group. In an attempt to discover a more precise biomarker of disease risk, we performed a large-scale genomics study. The result? A chromosome signature associated with greater patient survival.

Using data from the most recent pan-European medulloblastoma trial, we identified a group of patients whose tumors were characterized by multiple whole-chromosome aberrations associated with an excellent prognosis. Clustering and biostatistical methods enabled us to establish the combination of chromosome defects that best predicted favorable outcomes in the standard risk category, indicating those who would benefit from a reduction in treatment intensity. Ultimately, the “favorable risk” patients who possessed the signature exhibited a 100 percent survival rate.

Conversely, patients without the chromosome signature showed a five-year survival rate of only 60 percent, placing them in a higher risk category. For these patients, it’s possible that treatment intensification is the best course of action. We estimate that stratification using chromosome biomarkers could facilitate the assignment of 150–200 patients per year in Europe to a “favorable risk” group, reducing the number of standard risk patients receiving intense treatment.

We now aim to validate this signature in further clinical trials before taking it to the clinic. To detect the signature, neuropathologists can use various methods – for instance, interphase FISH with specific chromosome probes, single nucleotide polymorphism arrays, and array comparative genomic hybridization, all of which are becoming more commonplace in medulloblastoma diagnostics. Pathologists will also need to identify non-WNT/non-SHH molecular subgroups of medulloblastoma, as these don’t fall within the standard risk category.

Our study has identified around half of all standard risk patients who could potentially benefit from less intensive therapies. The next step will be to run trials that test whether their favorable prognosis is maintained once therapy intensity is reduced – a goal that could alleviate neurological late effects, such as reduced growth and IQ.