Deadly cancers driven by relatively few mutations, study finds

Most common tumors are driven by just two to six DNA mutations, according to a genetic analysis of 3,000 tumor samples that brings scientists closer to devising a list of all key mutations that cause or contribute to cancer.

In the study, researchers from Washington University in St. Louis analyzed 3,281 tumors from 12 cancer types, including breast cancer, lung cancer and colon cancer. They found 127 mutated genes in those tumors that appeared to be involved in either in cancer initiation or progression.

The results indicate that a relatively small number of genetic changes fuel tumor growth, said Li Ding, senior author on the study.

“This is a really small number. This suggests that only a few driver mutations are required for developing tumors,” said Ding, assistant director of the genome institute at the university, in a phone interview.

In the long run, cataloging cancer’s genetic changes “will help develop personalized treatment strategies for each patient.”

The new work, published in the journal Nature, is part of the Cancer Genome Atlas project, a U.S. National Institutes of Health-funded effort to discover what changes make a normal cell cancerous and pinpoint more effective treatments.

After performing several studies examining mutated genes found in specific tumor types, researchers in the project are now scanning multiple cancer types simultaneously to find patterns of genetic abnormalities common to different tumors.

The ultimate goal is to treat patients with therapies tailored to the specific mutations involved in each individual’s cancer.

The new study moves researchers closer toward devising a comprehensive list of the key genes involved in cancer, said Ding in a phone interview. Her study is not the final word because it looked only at certain types of common mutations and didn’t examine other genetic changes in which large chunks of DNA are rearranged abnormally.

Still, given the rapid rate at which researchers are generating cancer genome data, “we have a reasonable chance of identifying most of the core cancer genes” in three to five years, Ding said.

One promising result of Ding’s study is that it identified several mutated genes that are bad prognostic factors across a number of cancers, said Thomas Hudson, a genome scientist at the Ontario Institute for Cancer Research, in a phone interview.

If confirmed, the findings could help doctors develop new prognostic tests that indicate which patients need the most aggressive treatment, he said.

“Having markers of bad prognosis is really critical” as it may help doctors decide which patients may benefit from drug treatment after surgery to prevent cancer from recurring, said Hudson.