Cancer is a disease of genes gone wrong. When certain genes mutate, they make cells behave in odd ways. The cells divide swiftly, they hide from the immune system that could kill them, and they gain the nourishment they need to develop into tumors.
Scientists started identifying these cancer genes in the 1970s, and their list slowly grew over the years. By studying them, scientists came to understand how different types of cancer develop, and in some cases they were even able to develop gene-targeting drugs. Last May, for example, the Food and Drug Administration approved a drug known as Tarceva to treat lung cancer in which a gene called EGFR has mutated.
The National Institutes of Health, hoping to speed up the identification of cancer genes, started an ambitious project in 2005 called the Cancer Genome Atlas. It analyzed 500 samples from more than 20 types of cancer and found a wealth of new genes. The data have helped scientists discover more of the tricks cancer cells use to thrive at our expense.
“The Cancer Genome Atlas has been a spectacular success, there’s no doubt about that,” said Bruce Stillman, president of Cold Spring Harbor Laboratory.
But now, as the Atlas project is coming to an end, researchers at the Broad Institute of MIT and Harvard have published a study in the journal Nature that has scientists debating where cancer research should go next. They estimated that scientists would need to examine about 100,000 genes — 10 times as many as the $375 million Cancer Genome Atlas has gathered — to find most of the genes involved in 50 cancer types.
“We now know what it would take to get a complete catalog,” said Eric S. Lander, founding director of the Broad Institute and study co-author. “And we now know we’re not close to done. We have a lot left to learn.”
For their new study, the scientists examined cancer samples from the Cancer Genome Atlas, as well as cancer samples from the Broad’s own collection. All told, they analyzed 4,742 samples from 21 types of cancer.