Classifying and Understanding Pediatric Medulloblastoma

Classifying and Understanding Pediatric Medulloblastoma

Matthias A. Karajannis, MD

The understanding that medulloblastoma is comprised of 4 distinct molecular subtypes of disease, each representing a unique opportunity for new drug development, has set the groundwork for a wave of advances in the treatment of pediatric brain cancer, according to Matthias A. Karajannis, MD.

In research conducted by the Medulloblastoma Advanced Genomics International Consortium (MAGIC) that established the 4 subtypes, somatic copy number aberrations were analyzed from more than 1000 medulloblastoma samples. The WNT and SHH subgroups were identified as the most actionable while the other two groups (known simply as Group 3 and Group 4) were less defined, with few actionable therapeutic targets.

In the analysis, the WNT group did not have any significant gene deletions or focal gains, suggesting a less aggressive subtype. However, on the other end of the spectrum, the SHH tumors exhibited multiple somatic copy number aberrations, including those in thePI3Kgenes,PTEN, andIGFsignaling genes.

To shed more light on these findings,Targeted Oncologyspoke with Karajannis, pediatric oncologist, NYU Langone's Laura and Isaac Perlmutter Cancer Center, on a shift toward a more personalized treatment approach for medulloblastoma.

TARGETED ONCOLOGY:Can you talk about the importance of findings from MAGIC?

KARAJANNIS:Medulloblastoma is a rare and aggressive pediatric brain tumor. The main accomplishments of MAGIC were to pool together a large number of international investigators, pool samples from medulloblastoma, and conduct a number of groundbreaking genomic studies to understand the tumor genetics and the biology of medulloblastoma.

Medulloblastoma has turned out to be a heterogeneous disease with at least four different subgroups that have fundamentally distinct biologies, as well as clinical features, and different outcomes.

TARGETED ONCOLOGY:What were the implications of the findings, in terms of therapy development?

KARAJANNIS:MAGIC, and other groups, have identified a subgroup of medulloblastoma called the WNT subgroup that has a particularly good response. Patients with this subtype of medulloblastoma do exceptionally well with standard therapy. This knowledge has formed the basis of a new generation clinical trials that is currently emerging. The trials are attempting to deescalate therapy for these patients, because we believe they may not require as aggressive of a therapy. This therapy includes chemotherapy and radiation therapy.

The goal is to identify particular patients who have a good risk profile and trying to customize their therapy and reduce long-term toxicity of chemotherapy and radiation therapy in these patients. We believe we will be able to select patients with a good risk profile and whose outcome may still be excellent with reduced therapy in particular. There are efforts to eliminate radiation therapy, or at least reduce it in those patients. Radiation therapy usually carries a significant risk of short-term and long-term toxicities.

TARGETED ONCOLOGY:Do you feel as though there is potential for molecular targets for other patients who need more aggressive treatment with other subtypes of medulloblastoma?

KARAJANNIS:A wealth of information has emerged from this data. There is another subtype of medulloblastoma, sonic hedgehog (SHH), where recurrences are predominantly local. Efforts from upcoming clinical trials may need to focus on escalating therapy locally, at the primary site of the tumor.

There are other subgroups such as Group 3, where poor survival is mostly a function of metastatic disease recurrence. For those patients, the focus of therapy will probably have to be on the metastatic component. A lot of this biology is still emerging with studies coming out almost on a monthly basis, which are further mining this genomic data.

There are also new biology studies that help us to understand what drives the tumor recurrence in the different subgroups.

TARGETED ONCOLOGY:Are there any findings in the pediatric setting that have been translated to adult brain cancers, or could be translated to adult brain cancers?

KARAJANNIS:It turns out that pediatric and adult brain tumors don’t segregate at age 18. We do have certain types of tumors that only occur in young children or only occur in older adults, but this is a small slice of pediatric brain tumors. The majority of brain tumors and subtypes of brain tumors occur across a relatively wide age spectrum. So these are tumors that occur in a 16-year-old, but they can also occur in a 36-year-old and biologically, they are indistinguishable.

There has been a lot of overlap between pediatric and adult neuro-oncology. Now that we are learning more and more about the molecular underpinnings of the disease subtypes, we are also recognizing that for many of these subtypes there is a much wider age spectrum than we previously believed.

There is a group of pediatric brain tumors that are characterized by a specific genetic mutation called K27M, which was originally identified in pediatric, high-grade gliomas exclusively. It recently turned out that the same mutation and the same disease can also be found in spinal cord high grade gliomas in adults. This is one example where a tumor that was thought to be specifically pediatric actually turns out to be present in adults.

There is a lot of shared biology for this example and hopefully there will be a lot of shared therapies for these patients.

TARGETED ONCOLOGY:Looking ahead, what are you most optimistic about in terms of treatment in pediatric brain cancer?

KARAJANNIS:We have now begun to actually understand the biology of the diseases. As recent as 5 or 10 years ago, we literally had almost no idea what underlies the growth and progression of pediatric and adult brain tumors. Most of this knowledge is very recent, so we’ve done the first step of understanding the biology. The next step, and this is what we’re most excited about because this will most benefit our patients, is to be able translate those findings into treatments.

We are already at a stage where the knowledge is allowing us to classify tumors much better at the time of diagnosis, which gives us a lot more information about what treatments might work for a given tumor. We are also much better at predicting patient outcomes based on biology.

Many of the targets identified relatively recently in pediatric brain tumors are becoming targets for therapies that are being developed. The challenge is going to be to bring these new drugs that are being developed to the clinic and to find the right patient population that will benefit most from them.

I have no doubt there will be tremendous progress made over the next 5 to 10 years.