Breast tumors can be divided into 10 different subtypes, according to a paper published online this week in Nature. The work is being lauded by TV news media and newspapers as a “breakthrough” study that will “revolutionize” breast cancer care in the “next two to three years.”
However, breast cancer experts I’ve spoken with this week, are a bit more sober in their assessment of the study’s impact.
Everyone agrees the work is exceptional, and the investigators deserve substantial credit for pursuing such a large endeavor: They used genomic and transcriptome analyses to characterize nearly 2000 breast tumors. And the team, led by Carlos Caldas at Cancer Research UK in Cambridge and Samuel Aparicio at the University of British Columbia, Vancouver, was able to use those characteristics to define 10 different breast cancer subtypes that differ significantly in their prognosis.
The catch, though, is translating these findings into something with clinical utility.
Researchers already used expression arrays to define five different subgroups of breast cancer -- luminal A, luminal B, HER2-positive, basal type, and normal-like. “What that has done is given us a shorthand so that we can talk to each other,” said Dan Hayes, MD, Stuart B. Padnos Professor of Breast Cancer Research and Clinical Director of the Breast Oncology Program at the University of Michigan in Ann Arbor. “So when I say luminal A, someone across the hall knows what I mean, ER-positive, low proliferation, not HER2-positive.”
What the community needs now, according to Dr. Hayes, is a better understanding of the basal-type tumors. That label is often considered synonymous with triple-negative (HER2-negative, estrogen receptor-negative, and progesterone receptor-negative), but in truth neither moniker is completely accurate and the groups are not entirely overlapping.
“What we need are better therapies directed towards these so-called triple-negative/basal cancers,” Dr. Hayes continued. “Studies like the one that was just published -- and other people are doing the same thing -- are designed to divide that triple-negative/basal group into even better defined sets of patients, to try to figure out if there are any targets that we can focus on, and make our studies more directed and more efficient.
“Does this paper do that? I think time will tell,” he said.
“Right now this study doesn’t change anything for me in clinic, but I am pretty excited about it. This is an exploratory analysis to find groups for which future trials might be focused.”
Similarly, Larry Norton, MD, Deputy Physician-in-Chief for Breast Cancer Programs and Medical Director of the Evelyn H. Lauder Breast Center at Memorial Sloan-Kettering Cancer Center in New York, noted that the work does reveal some novel genetic aberrations that might be useful as targets for therapy in the future.
More important, in his mind though, he said, is what the work doesn’t do: It does not improve the quality of the prognostic groups.
“Clearly the genomic changes and RNA expression changes they found can be used as a classifier. But it doesn’t look any better or worse than all the other classifiers out there.
“I think that is one of the major overriding points being made by all of this work,” he continued. “Lots of different techniques have been used to separate patients by prognosis and they all separate them roughly to the same degree.
“We are getting lots of different classifiers that are about the same quality. But why aren’t we getting better classifiers?,” he asked. “I have this nagging suspicion that there is something really very fundamental that we are missing. And that is really what we have to wrestle with. What are the pieces of the puzzle we are missing?”
So, yes, the newly published work is interesting, exciting, and possibly will lead to changes in the future. But we shouldn’t hold our breath -- and the media shouldn’t promise patients something that doesn’t yet exist.