Posted 8 a.m. Fri., 02.05.10 - "No child should die in the dawn of life," said Danny Thomas, the late entertainer and founder of St. Jude Children's Research Hospital, a world leader in research and treatment of childhood cancer.

To further this goal, St. Jude has teamed up with the Genome Center at Washington University School of Medicine. In a project announced Jan. 25, researchers at the two institutions will collaborate to sequence the genomes of more than 600 childhood cancer patients.

The estimated cost of the project is $65 million over three years. Kay Jewelers, a long-time supporter of St. Jude, is contributing $20 million. This effort, the largest research project to date devoted to pediatric cancer, could lead to a better understanding of cancer's genetic origins in both children and adults and possibly to more effective treatment.

Whole genomes - a new tool

In 2008, researchers at Washington University's Genome Center became the first to sequence an entire cancer genome and compare it to the sequence in the same patient's healthy cells. In this case, the female patient had acute myeloid leukemia (AML), a type of blood cancer. They identified 10 mutations in the patient's cancer cells, eight of which were previously unknown in AML.

The study demonstrated that whole genome sequencing can be a powerful tool in finding mutations relevant to cancer. Many such mutations would likely be missed using older, more piecemeal approaches to the genetic study of cancer.

Since then, the team has sequenced cancer genomes and the corresponding healthy genomes of the same person from more AML patients as well as patients with other types of cancer including breast, lung, ovarian and brain.

This expertise led St. Jude to choose Washington University over other institutions with large genetic sequencing centers.

"We did look around at other genome centers, but there was no question in our mind that the best place for us to partner was Wash U.," said James Downing (right), St. Jude's scientific director. "They were the clear leaders in doing whole-genome sequencing of cancer samples."

Washington University will provide the computing power to sequence the genomes, and St. Jude will provide the samples to be sequenced. Since the 1970s, St. Jude has gathered a massive library of pediatric cancer samples.

"St. Jude is a remarkable place," said Timothy Ley (right), professor of medicine at Washington University and one of the researchers involved in the project. Ley called the research hospital a tremendous resource - "unique in the world in terms of their samples from patients."

Map the battlefield

Ley compares the current fight against cancer to fighting a war without a map of the battlefield. "We're just firing Howitzers blindly at the horizon right now," he said. "We're trying to establish the genetic map of cancer and you really need whole genomes to do that fully. What we don't know about pediatric cancer is whether it will be the same or different from the adult map."

Past research by these and other groups suggest that childhood cancer is likely somewhat different from adult. As an example, Ley points to adult vs. pediatric AML. The diseases can look identical, but mutations that are common in adults with the disease are rarely seen in children. "They seem to have an overlapping but different set of mutations," Ley said.

In addition, Ley suggests that childhood cancer may be simpler than adult cancers. Since pediatric cancer patients are younger, Ley explained, "They will have had less of a chance to acquire random mutations in their genome that are just a consequence of aging." These "background" mutations, more common in adults, complicate analysis of the genome because they can be in the tumor, but not relevant to the cancer.

Another layer of complexity will come when the researches investigate the epigenome of the cancer genomes they sequence. Richard Wilson (right), director of the Genome Center at Washington University, calls the epigenome a "secondary code." It does not alter the gene sequences themselves, but it is likely relevant in cancer because it influences how the genes are expressed.

As with any massive project, the first question is where to start. Ley and Downing say they are still discussing which cancers should be sequenced first. Downing says they want to focus on diseases they can learn the most from, that are among the most common and that have the poorest outcomes.

Illustration courtesy of St. Jude

St. Jude and Washington University will sequence the genomes of normal and cancer cells to find genetic mistakes that lead to childhood cancer.

In terms of blood cancers, a high priority will likely be acute lymphoblastic leukemia, a cancer of the white blood cells. It is the most common pediatric cancer, according to statistics from St. Jude.

"If I had come down with acute lymphoblastic leukemia as a child, I would have had a 5 percent chance for a cure," said Downing. "Today a child who walks in with it has a greater than 90 percent chance for a cure." Downing says he would like to study that last 10 percent that continue to have poor outcomes. "Let's sequence those and see if we can learn something to help us better treat them," he said. Other cancers likely to be sequenced include other types of leukemia, brain tumors, muscle tumors, and sarcomas (tumors of connective tissues).

Targets and prognostics

In the long term, these researchers hope a cancer map might provide "targets" for future drug development, leading to designer molecules that might interfere with pathways that let cancer cells grow and spread. "But drug development takes 10, 15, 20 years to go through all the clinical trials and testing and FDA approval," Wilson said.

Fortunately, Wilson says there are shorter term opportunities as well. It can be useful, he said, to think of mutations not as drug targets, but as "genetic signatures" that are associated with a good or bad prognosis. With a large database of cancer genomes, sets of mutations that are present in a significant number of those genomes are likely playing some role in the cancer. If those mutations are also associated with a good or bad outcome, the information becomes useful in a clinical setting.

Right now, "physicians treat most AML patients about the same up front," explained Elaine Mardis (right), co-director of the Genome Center. Subsequent steps in treatment, though, are determined by what chromosomal changes are present in the tumor cells. If those changes are bad, the patient would get a more aggressive treatment. In most cases, however, who will do well and who will do poorly is unknown. According to these Washington University researchers, whole genome sequences identifying all the mutations will improve this classification scheme.

Indeed, Wilson says such information "can give an oncologist immediate feedback on the most appropriate treatment for a patient using the drugs that he currently has." This group has already identified one set of mutations in AML that is associated with a poor outcome.

If a patient has this set of mutations, the doctors can know right away to treat that cancer aggressively, perhaps considering such invasive procedures as a bone marrow transplant much earlier. Likewise, a patient without the mutations may be given the traditional chemotherapy regimen. In short, such information is powerful even without new cancer drugs or therapies.

After scientists at Washington University and St. Jude complete and verify the genome sequences, they will make the data publicly available. Then, others in the scientific community can use the information to speed research against childhood cancer.