Scientists have reported that through exome sequencing, they have been able to identify changes that characterize each of the four recognized subtypes of the brain cancer medulloblastoma.

How much of a problem pediatric medulloblastoma is, Scott Pomeroy told BioWorld Today, "depends on your lens."

In terms of sheer numbers, at around 500 cases a year, it is dwarfed by the numbers of lung, breast and colon cancer cases, each of which account for more than 200,000 new illnesses a year.

But specifically in children, cancer is second only to accidents as a cause of death, and medulloblastoma is both common and aggressive.

Furthermore, the disease is treated quite aggressively, with long-term consequences for its victims. "If you irradiate a child that is less than 5 years old, it has pretty devastating consequences," Pomeroy said.

And so, while long-term survival has improved, many of those survivors will never be able to live independently.

In the July 22, 2012, issue of Nature, Pomeroy, who is neurologist-in-chief at Boston Children's Hospital, and his colleagues reported the results of an exome sequencing study that identified chromosomal changes that tend to underlie each of the subtypes.

The "quick punchline" of the paper reporting the studies, Pomeroy said, is that the overall mutation rate, particularly the p53 mutation rate, of the tumors is fairly low, and that those mutations segregate pretty clearly into the subtypes.

That low mutation rate may seem surprising. If pediatric patients have had fewer years than adults to accumulate mutations, and the mutation rate of their cells is low, then their having cancer is against the odds. The cancer rates in 6-year-olds are, of course, lower than those in 60-year-olds. But Pomeroy said that precisely because their brains are developing at a dizzying speed, children likely also have a population of cells that are "in a state of differentiation where it doesn't take much" to push them into a cancerous state.

And, he added, the data are a piece of a broader emerging story of what goes wrong in pediatric medulloblastoma – and perhaps also in other tumor types. "The data this last year has really changed how I think about the disease," he said.

In pediatric cancers, he elaborated, there is "a surprising emphasis not on signal transduction mechanisms per se, but on the regulation of transcription." For example, changes in histones are common in pediatric but not adult medulloblastomas.

Many oncologists think of changes in transcription as being a consequence of genetic mutations, "sort of a squiggle at the end," Pomeroy noted. But pediatric patients, at least, can have "an amazingly normal-looking genome," and for them, "it's the squiggle at the end that's the problem.

"It's become increasingly clear that medulloblastoma has a greater degree of heterogeneity than you can see under a microscope," and the current study builds on earlier work that showed medulloblastomas break down into four subtypes. That situation is "not unique to medulloblastoma," he added. "But it is clearly playing out in this tumor type."

In their work, Pomeroy and his team used exome sequencing to look at the four subtypes in greater detail.

They found that one subtype, which affects the sonic hedgehog pathway, is characterized by chromosomal gains and losses. Another, which affects the Wnt pathway, has very few such gains and losses with the exception of chromosome 6, which is pretty invariably affected. Two other subtypes, three and four, are "a little fuzzy but still distinct" and tend to have many chromosomal gains and losses, and can be further divided into subtypes.

Survival outcomes are very different between the subtypes. While patients with mutations that impinge on the Wnt pathway have an overall five-year survival rate of more than 90 percent, for one subtype of group 3 tumors with a gain of c-Myc, that survival rate is around 20 percent.

Given the low quality of life that is currently the lot of many medulloblastoma survivors, Pomeroy said, his team's findings may open up the possibility of easing up on treatment for patients with relatively more benign tumors, although the question of what, exactly, to cut back on is, of course, a fairly fraught one.