'The Time Has Come' for Medulloblastoma

Zosia Chustecka

May 31, 2011 — A huge research project focused on medulloblastoma has taken more than 20 years to collect a sufficient number of tumor samples, but now the analyses are complete and the results are out. They show that this tumor has 4 distinct molecular variants, each with a different clinical pattern, which has immediate treatment implications.

The results are outlined in 4 separate papers, published simultaneously in the Journal of Clinical Oncology. An accompanying editorial declares: "the time has come for many things in medulloblastoma."

The hope is that this research will lead to better treatment of medulloblastoma, said Scott Pomeroy, MD, PhD, professor of neurology at the Children's Hospital in Boston, Massachusetts, who is senior author on 2 of the papers.

Current treatment is craniospinal radiation with chemotherapy; although this has a good survival rate (around 80%), the damage to the developing brain leads to cognitive and other adverse events. This means that as these children grow into adulthood, they are unable to live independent lives, hold down jobs, and have families of their own, Dr. Pomeroy explained. "They usually continue to live with their parents or in sheltered accommodation," he noted.

Identifying subtypes with a good prognosis raises the hope that the treatment can be scaled back in these children; for instance, lower doses of radiation can be used, he said. In the subtypes identified as having a poor prognosis, treatment can be intensified to improve survival rates, he added.

Although there has been some move toward different treatment based on clinical signs and age of the child, these new genotyping studies have "defined a new landscape," Dr. Pomeroy said, adding that "the conversation has changed."

Speaking last week at a Cancer Genetics Media Fellowship at Harvard Medical School in Boston, he said this medulloblastoma project is an example of the impact that genetics research can have on clinical practice.

"I think we are ready now to start making clinical decisions based on these new data," Dr. Pomeroy noted.

"We should move forward quickly," write the 3 editorialists from Stanford University, Palo Alto, California: Michelle Monje, MD, PhD, neurooncology fellow; Philip Beachy, PhD, professor of developmental biology; and Paul Fisher, MD, director of neurooncology.

"Patients waiting in the clinic right now are entitled to better risk stratification than that of the anachronistic system we are currently using," they add.

Huge Undertaking

"This was a huge undertaking," explained Dr. Pomeroy. Medulloblastoma is a rare childhood brain tumor. About 400 cases are seen annually across all the United States; his department sees about 10 cases each year.

It took more than 20 years to collect enough tumor samples that were of sufficient quality for genotyping, he said. "It's not as if we weren't trying."

Dr. Pomeroy and colleagues conducted genotyping on 194 tumor samples. Two other groups report analyses from sets of 103 and 207 samples.

The evidence from these independent studies "clearly points to the existence of at least 4 molecular subtypes of medulloblastoma with differential outcomes," write the editorialists.

The 4-marker pattern was validated in 294 medulloblastoma tissue samples, which confirmed the risk stratification.

Although genotyping of the tumor tissue was done to obtain some of these results, the 4 subtypes can be determined with standard immunohistochemistry.

This is "remarkable," write the editorialists, because this technique is both readily accessible and relatively inexpensive.

"This could allow for rapid transition of a molecular medulloblastoma classification system into clinical practice," they add.

Different Approaches to Treatment?

Because each subtype of medulloblastoma has a different prognosis, each may warrant a different approach to treatment.

For instance, the Wnt pathway active subtype, associated with monosomy 6 and occurring in all age groups, has an excellent prognosis, the editorialists note. It is possible that these patients are currently being overtreated, they write, and their treatment-related sequelae could be minimized.

The Hh pathway active subtype, associated with desmoplastic histology and occurring primarily in infants and adults, has a good prognosis, they add.

Both of these subtypes have clearly defined targets for specific therapy; drugs that target the Hh pathway already exist, and drugs for the Wnt pathway are under development. Clinical trials with these new agents can be designed to include only patients with the appropriate subtype, the editorialists explain.

However, the c-MYC-amplification subtype, associated with anaplastic/large-cell histology and occurring primarily between the ages of 3 and 10 years, carries a dismal prognosis, they report. These patients might require a more aggressive therapy than the current standard of care, they add.

The current treatment of medulloblastoma is under scrutiny, Dr. Pomeroy explained. Already there have been moves to reduce radiation doses in "good risk" patients, from 3600 Gy to 2400 Gy. Now, there is a clinical trial underway that is looking at an even lower dose, comparing 1800 Gy with 2400 Gy.

"The question that we are asking ourselves is: How low can we go?" he said, especially in the newly identified Wnt subtype, where the survival rate is better than 90%.

There are also moves underway to reduce the dose of chemotherapy, he said. In fact, there is little solid evidence to show that chemotherapy is necessary after the radiation, but it is a standard of care and difficult to test, he said. A randomized clinical trial comparing radiation alone with radiation plus chemotherapy was stopped because no parents were prepared to have their child go without the chemotherapy, he added.

The identification of the poor-prognosis c-MYC subtype is perhaps "the most novel finding of our paper," Dr. Pomeroy said. This subtype accounts for about 10% to 20% of all medulloblastomas; however, in contrast to the other subtypes, the survival rate for these patients is only about 20%.

These patients warrant the maximum doses of radiation and chemotherapy, "but we can't push these treatments any further, so we will have to look for something different for this group."

J Clin Oncol. 2011:29:1395-1398, 1400-1407, 1408-1414, 1415-1423, 1424-1430.