Classifying cancer tumors by their molecular structure rather than the tissue or organ where they were found, such as the breast or bladder, may lead to more accurate diagnoses and potentially better treatments and outcomes for patients, a new study finds.
In the largest undertaking to analyze and compare different cancer types based on genomic sequencing, researchers found at least 10 percent of tumors - and possibly as high as 30 to 50 percent - would be identified differently if oncologists determined their diagnoses by a tumor's molecular makeup.
"The old system classifying cancer by the tissue of where it arose is outdated. It's been in existence for over 100 years now, and we know it doesn't merit the true nature of the cancer," said Dr. Christopher Benz, a professor at the Buck Institute for Research on Aging in Novato and co-senior author of the study.
The results were published online Thursday in the journal Cell.
The project is part of a national effort, the Cancer Genome Atlas project, which started in 2006 and is funded by the National Institutes of Health. It brings together more than 150 scientists from around the country, including researchers from the Buck Institute and UC Santa Cruz, with the goal of studying the genomic changes in more than 20 cancer types.
12 types of cancer
In this latest study to come out of the effort, researchers analyzed more than 3,500 samples from 12 cancer types that had been sequenced. Those included breast, kidney, bladder, brain, colon, endometrial and lung, among others.
They found that in many of the cancer types, such as an aggressive form of brain tumor known as glioblastoma and a type of leukemia, the tumor samples matched up well with the tissue classifications, suggesting that a tumor's location is still important for certain types of cancers.
But with other cancer types, tumor samples on a molecular level appeared to look more like unrelated cancers. For example, a significant number of squamous head-and-neck cancers looked more like some squamous-cell cancers found in the lung.
"How we use that information for treatment is anybody's guess right now," said Josh Stuart, professor of biomolecular engineering at UC Santa Cruz and a senior author of the paper.
The discovery that some tumors shared molecular similarities with tumors from other parts of the body was particularly striking in bladder cancer as well as an aggressive subtype of breast cancer scientists call "basal-like" but is more commonly known as "triple negative."
Explains patient response
Researchers said the results may help explain why some bladder cancer patients respond much differently to treatment than others.
In the triple-negative breast tumors, named as such because they lack the three hormone receptors targeted by the most successful therapies, the researchers learned the molecular wiring looked so different from any other breast cancers that they think the disease may warrant a new classification.
"It actually looks like a different cancer completely," said Benz, who is also a breast cancer specialist at UCSF. "It's a totally independent type of cancer from a molecular point of view."
Scientists say more work needs to be done before doctors are able to make different treatment decisions based on the new research. Many cancers, such as prostate, liver, melanoma and cervical, have yet to be sequenced, and adding them to the genomic "map" may change results.
"What we want to do with these maps is figure out all these different ways of looking at these (tumors) molecularly and putting more of them in the map," Stuart said.
But Benz said the research may help scientists know how to design new clinical trials to figure out, for example, whether an existing therapy for one cancer will work for a subtype of a seemingly different cancer.
"We have to keep doing this for all these cancer types," he said, "to completely reorganize and form a new textbook on oncology."
Victoria Colliver is a San Francisco Chronicle staff writer. E-mail: vcolliver@sfchronicle.com Twitter: @vcolliver
0 comments:
Post a Comment