Out of sequence?

Scientists debate whether it’s time to tackle tumor genomes and epigenomes

On a cold January afternoon, Eric Lander sits in his office at the Broad Institute, explaining a new effort to sequence the DNA of tumors, when his cell phone rings. He answers and listens.

“Do you want me to come over to the hospital tonight?” he asks, then covers the mouthpiece and whispers, “I just want to make arrangements to visit my cousin. Unfortunately, he has a very serious cancer, so this is quite relevant to our conversation.”

Lander’s cousin was diagnosed with a type of bile duct cancer at the end of December. This form of cancer is relatively rare, and without known therapies.

Getting off the phone, Lander notes that “we could try to use therapies that were developed for other cancers to treat my cousin, but we don’t know which genes are involved, and that’s very frustrating.”

Illustration: James O'Brien

The Broad is participating in a new federally funded venture that could help patients such as Lander’s cousin. The National Cancer Institute recently kicked off a pilot project to sequence the DNA of tumors from patients with particular types of cancer. If the pilot succeeds, the government might fund a massive cancer-sequencing project to determine the genetic components of all types of cancer. Many involve mutations to the same genes, and the resulting information would help researchers map relationships between cancers.

But some scientists feel that NCI should wait to launch the project when federal funding for biomedical research expands. The National Institute of Health’s budget doubled between 1998 and 2003 but is now dropping in real dollars each year. Whitehead Member Robert Weinberg suggests that grant funding should emphasize small, investigator-initiated projects rather than large, collaborative ones.

“I’m not at all convinced that, given the current NIH grant funding climate, we can afford to invest enormous amounts of money in sequencing cancer genomes,” says Weinberg. “The larger-scale projects generate large databases, which are useful to smaller-scale scientists, but on their own, larger-scale projects hardly yield as much conceptual bang for the buck as smaller-scale projects,” he declares. “I think there are still some major unsolved conceptual problems that only smaller, more focused research efforts can address.”

Lander, who is both a Whitehead Member and Director of the Broad, counters that the pilot project will cost just 0.5 percent of the NCI budget.

“I can’t imagine why you wouldn’t spend 0.5 percent of the budget on cancer genome sequencing, because it would make the other 99.5 percent of the budget twice as efficient,” he says. “And the Human Genome Project taught us that setting a goal often causes costs to drop as the technology grows more efficient. This makes more science possible.”

Sequencing the epigenomes

A group of scientists recently proposed another giant project—sequencing several human epigenomes—to advance cancer research. The epigenome lies above the DNA sequence and includes methyl marks as well as proteins that package DNA (for more information see "The Unusual Suspect”). Aberrant methylation plays a role in many types of cancer.

As president of the American Association for Cancer Research, Peter Jones helped to develop a blueprint for an international project to sequence several human epigenomes. Eventually, he hopes to compare the epigenomes of cancer patients with those of healthy individuals.

“The epigenome is the missing piece between genes and proteins, and we need the sequence to fully understand cancer and make accurate diagnoses,” says Jones, director of the USC/Norris Comprehensive Cancer Center. Like Lander, he believes that larger-scale projects make smaller-scale projects more efficient.

But the proposal faces critics. Some wonder if we know enough about the epigenome to launch such a monumental project. Labs are still uncovering layer upon layer of epigenetic marks that current sequencing technologies miss. And the project requires sequencing numerous epigenomes, because each cell type contains a different set of epigenetic marks.

“My feeling is an epigenome project is a bit premature,” says Emma Whitelaw, of Queensland Institute of Medical Research. “We could save a lot of money by waiting a few years, by which time we should know more about what to look for and where to look for it.”