By Ben Harder
A Menu of Mutations
Researchers begin to demystify the genomes of two cancers
By Ben Harder
Genetically speaking, cancer cells have a lot in common with the body’s normal cells. It’s the differences—like mutations that vary from one person’s tumor cells to another’s—that can render cancer difficult to treat. Those variations make it hard to predict which patients will benefit from a given therapy.
Researchers want to know which genetic mutations distinguish tumors from healthy tissues and also which mutations mark subsets of tumors, says oncologist and cancer researcher Victor Velculescu of the Johns Hopkins University Kimmel Cancer Center in Baltimore. That information would help them design better diagnostic tests, develop new drugs, and more
effectively select drugs for appropriate patients.
But cancer’s genetic makeup, or genome, has remained largely a mystery.
No longer. A new study, reported by Velculescu and his colleagues in the Oct. 13, 2006, Science
, has identified nearly 200 mutations in breast and colorectal tumors. More than 350,000 Americans are diagnosed annually with these two kinds of cancer and nearly 100,000 die of them.
The researchers examined more than half the genes in cells that had been removed from 22 patients who had either a breast or colorectal tumor. Each tumor had an average of about 90 mutated genes. By analyzing the frequency of each mutation, the researchers picked out a subset of these genes—about 11 per patient, and 189 genes in all—that appear to be prime suspects in the genesis of cancer. Most of the genes on the list had never before been considered suspicious, Velculescu says.
The results indicate that breast cancer and colorectal cancer have distinct genetic blueprints, says Velculescu. Even within breast or colorectal cancers, most of the mutations were unique. That finding, he says, “foreshadow[s] how cancers will be diagnosed and treated in the future. Each patient will be treated differently depending on the individual gene mutations in their tumors.”
The study takes an important step toward identifying “the abnormalities that are critical to the growth of breast cancer and colon cancer,” says Brian Druker, an oncologist at the Oregon Health & Science University Cancer Institute in Portland. But more remains to be done. First, follow-up studies need to separate the genetic changes that drive cancer from those that merely go along for the ride. Second, because of the small number of tumors sampled—only two types of cancer and only 11 samples of each—important genes may yet remain undetected, he says.
Upcoming studies, particularly one undertaken by the National Institutes of Health, will generate similar catalogs for other cancer types, Druker says, and may eventually whittle down the list of genes involved in breast and colon cancers.
Researchers would like to develop cancer treatments based on proteins in the blood produced by a tumor’s mutated genes. But identifying these proteins among many others is sometimes compared to looking for a needle in a haystack, without knowing what the needle looks like, notes biochemist Dwight Randle, the senior scientific adviser to the Susan G. Komen Breast Cancer Foundation in Dallas. That’s because scientists haven’t had a guide to tell them which proteins, or needles, came from the tumors.
“This list of genes actually tells us what the needle looks like,” he says.