By Christina Frangou

image In a finding that is expected to revolutionize future approaches to the treatment of breast cancer, researchers have confirmed the presence of four distinct subtypes of breast cancer, each with a unique heterogeneous mix of genetic and molecular abnormalities.

The findings, published online Sept. 23 in Nature, add to the growing body of evidence suggesting that tumors should be catalogued and treated based on the genes that are disrupted, rather than the tumor location in the body.

As well, the researchers reported a breakthrough finding about triple-negative breast cancers: These breast cancers more closely resemble ovarian cancer than other breast cancers.

Basal-like breast tumors and ovarian tumors could potentially be treated with the same therapeutic approaches, said the investigators.

“With this study, we’re one giant step closer to understanding the genetic origins of the four major subtypes of breast cancer,” said Matthew J. Ellis, MD, PhD, Anheuser-Busch Chair in Medical Oncology, Washington University School of Medicine, St. Louis, in a statement.

“Now, we can investigate which drugs work best for patients based on the genetic profiles of their tumors. For basal-like breast tumors, it’s clear they are genetically more similar to ovarian tumors than to other breast cancers. Whether they can be treated the same way is an intriguing possibility that needs to be explored.”

Surgeons, oncologists and other physicians who treat patients with breast cancer say the study adds significantly to knowledge of breast cancer, and will shift from the traditional way of thinking about cancer, in which tumors are classified by body part, to a more modern system that relies on a tumor’s genetic signature.

“I believe that this study is highly significant because it confirms what many of us have predicted for some time: Our current breast cancer phenotypes are far more complex than the current classification system suggests, and the genetic abnormalities extend to primary tumors in other organ systems,” said Richard J. Bleicher, MD, director of the Breast Fellowship Program and associate professor of surgical oncology, Fox Chase Cancer Center, Philadelphia.

At a Glance

Physicians who treat patients with breast cancer say the study will shift the way they think about cancer.

Triple-negative breast cancers more closely resemble ovarian cancer than other breast cancers.

The new research is part of The Cancer Genome Atlas Project which brings together leading genetic sequencing centers to identify and catalogue mutations involved in many common cancers.

The researchers gained key insights into previously defined gene expression subtypes and confirmed the existence of four main breast cancer classes.

“Data of this type will be what reshapes our understanding of cancer, how we classify it and how we tailor therapy.”

The new research is part of The Cancer Genome Atlas Project, which brings together the country’s leading genetic sequencing centers to identify and catalogue mutations involved in many common cancers. The effort is funded by the National Institutes of Health.

A nationwide consortium of researchers analyzed tumors from 825 women with breast cancer. The scientists used six different technologies to examine subsets of the tumors for defects in DNA, RNA and proteins. (In comparison, most other studies use one, perhaps two techniques to perform genetic analysis of cancer tumors.) The techniques included Agilent mRNA expression microarrays, Illumina Infinium DNA methylation chips, Affymetrix 6.0 single-nucleotide polymorphism arrays, miRNA sequencing, whole-exome sequencing and reverse-phase protein array data. Nearly 350 tumors were analyzed using all six techniques.

By integrating information across platforms, researchers gained key insights into previously defined gene expression subtypes and confirmed the existence of four main breast cancer classes: luminal A, luminal B, HER2-enriched and basal-like.

Charles Perou, MD, co-author of the paper and the May Goldman Shaw Distinguished Professor of Molecular Oncology, University of North Carolina School of Medicine, said researchers were able to collect “the most complete picture of breast cancer diversity ever.”

“This study has now provided a near complete framework for the genetic causes of breast cancer, which will significantly impact clinical medicine in the coming years as these genetic markers are evaluated as possible markers of therapeutic responsiveness.”

Across the four subtypes, mutations in only three genes—TP53, PIK3CA and GATA3—occurred in more than 10% of patients’ tumors. But the scientists found unique genetic and molecular signatures within each of the subtypes.

Compared with other subtypes, basal-like and HER2 tumors had the highest mutation rates but the shortest list of significantly mutated genes. These genes are thought to be the major drivers of cancer progression. Eighty percent of basal-like tumors had mutations in the TP53 gene, a known marker for more aggressive disease and poorer overall survival. About 20% of the tumors also had inherited mutations in the BRCA1 or BRCA2 genes, which are known to increase the risk for breast and ovarian cancer.

“This suggests that it only takes a few hits to key genes that drive cancer growth,” said co-author Elaine Mardis, PhD, co-director of The Genome Institute at Washington University.

Overall, the genetic profiles of basal-like and ovarian tumors were strikingly similar, with widespread genomic instability and mutations occurring at similar frequencies and in similar genes. Among the similarities, both tumors have BRCA1 inactivation, RB1 loss and cyclin E1 amplification, high expression of AKT3, MYC amplification and high expression, along with the high frequency of TP53 mutations.

“The common findings of TP53, RB1 and BRCA1 loss, with MYC amplification, strongly suggest that these are shared driving events for basal-like and serious ovarian carcinogenesis,” the authors reported.

“This suggests that common therapeutic approaches should be considered, which is supported by the activity of platinum analogues and taxanes in breast basal-like and serous ovarian cancers.”

Finding new drug targets for this group is critical, said the authors. Basal-like tumors account for about 10% of all breast cancers and disproportionately affect younger women and those who are black.

Another previous study from Dr. Ellis’ group showed that women with basal-like tumors do not benefit from anthracycline-based chemotherapy, commonly used to treat breast cancers (Clin Cancer Res 2012;18:2402-2412). The new data indicate that clinical trials should be designed to avoid the use of these drugs in basal-like tumors.

Instead, patients with mutations in the BRCA genes may benefit from poly (ADP-ribose) polymerase inhibitors or platinum-based chemotherapy, which already are used to treat breast cancer.

“It’s particularly exciting when research delivers important findings with an immediate impact on treatment,” said Marisa C. Weiss, director of breast radiation oncology at Lankenau Hospital in Philadelphia, and founder of the nonprofit information website

The study showed that luminal cancers had the lowest mutation frequencies and longer lists of significantly mutated genes, suggesting that defects in multiple genetic pathways lead to the development of luminal breast cancers.

Luminal A tumors are the most common form of breast cancer in the United States and the primary cause of breast cancer deaths, accounting for 40%. The study helps provide a much better picture of the genetic causes of this subtype. The most common mutation in luminal A tumors occurred in PIK3CA, and was present in 45% of these tumors. TP53 mutations occurred only in 12%.

Patients with luminal B tumors generally do well after treatment but many experience recurrence years after treatment. The study showed that the most common mutations in these tumors occurred in TP53 and PIK3CA, which may explain the disparate results seen in patients with this subtype.

“Now, we’re much closer to understanding the true origins of the different types of breast cancer,” said Dr. Ellis. “With this information, physicians and scientists can look at their own samples to correlate patients’ tumor profiles with treatment response and overall outcomes. That’s the challenge for the future: translating a patient’s genetic profile into new treatment strategies.”

For now, experts say that the findings are not changing clinical practice.

“Although this is interesting and fascinating research, it does not really change how we manage the patient with breast or ovarian cancer who is diagnosed today,” said Deanna Attai, MD, a breast surgeon at the Center for Breast Care, Burbank, Calif.

“We are able to get more information than ever before on the genetic makeup and biological behavior of an individual patient’s cancer, but we still are not at the point where we can offer truly individualized treatment. And the more we learn, the more questions remain.”

The study offers “lots of promise and potential for future treatment, but no real changes today in how surgeons and medical oncologists approach patients with breast or ovarian cancer,” Dr. Attai said.

But, physicians can use the report to help patients understand the complexity of breast cancer and the need for a multidisciplinary response, said Dr. Weiss.

“The message here is quite clear: The constitution of a cancer is complex, heterogeneous; the cells are not all identical,” she said.

“Each woman’s individual cancer is unique. It’s made up of a range of different cells that work in different ways. In order to get rid of them, you need to use different treatments that address different vulnerabilities of that cancer.”

Drs. Ellis and Perou are inventors on patent filing for PAM50 and have equity interest in Bioclassifier LLC.