Genomic Characterization in Breast Cancer

Perou Discusses Genomic Characterization in Breast Cancer and Understanding Basal-Like Disease
Laura Martin @OncEditorLaura
Published Online: Friday, May 8, 2015
Charles M. Perou, MD
Understanding and classifying tumors based on their genomic makeup is an essential component of precision medicine. Genomic characterization can be used to understand why some cancers respond to standard therapies or not, and allows oncologists to begin tailoring treatment to each patient’s disease, said Charles M. Perou, PhD, May Goldman Shaw Distinguished Professor of Molecular Oncology, professor of Genetics, and Pathology and Laboratory Medicine at Lineberger Comprehensive Cancer Center, University of North Carolina School of Medicine.

Perou has dedicated his career to the field of genomic characterization and was instrumental in the discovery of the five intrinsic subtypes of breast cancer: luminal A, luminal B, HER2-enriched, basal-like and claudin-low. He was the first to identify the basal-like/triple-negative breast cancer (TNBC) subtype. Today, he focuses his research specifically on basal-like disease.

In an interview with OncLive, Perou explains the importance of understanding and treating each subtype differently and why oncologists should be aware of the similarities between basal-like breast cancer and other non-breast cancer tumor types.

OncLive: How has genomic technology revolutionized the classification of breast tumors?

Dr Perou: Gene expression profiling has been around for well over a decade. It allows us to look at the whole genome of expressed genes at once. By using this wealth of data, we’ve been able to examine different types of tumors in an unbiased fashion. For example, we now understand that, in breast cancer, it is not one disease but five or more distinct tumor types, including the basal-like subtype. Through the use of the genomic technology, we have a good job of finding these subtypes of breast cancer and many other cancers, as well. Through gene expression, these findings have also been corroborated through DNA sequencing and proteomics.

How does better understanding these subtypes inform treatment decisions?

These subtypes predestine prognosis. Some subtypes are associated with a negative outcome, while others have a fairly favorable outcome. In addition, there are many drug targets that cosegregate subtypes. If we know a patient has a particular subtype, we can predict which therapies will work best. For example, if a patient has the estrogen receptor subtype, we will give them endocrine therapy, or if a patient has the HER2 protein we would give the patient trastuzumab.

Many drug targets and specific pathways are altered in certain subtypes and not in others. We are learning how to use that information to craft a personalized medicine approach to treating cancer. We can target the biology of each patient using this subtype information.

What have you learned about the basal-like subtype of breast cancer? How does it relate to TNBC?

Eighty-percent of TNBCs have the basal-like phenotype. Therefore, when you are studying TNBC, the huge majority of those are basal-like. Sometimes the terms are used interchangeably, but they aren’t perfectly interchangeable.

The basal-like subtype may share more genetic features with ovarian cancer and lung squamous cancer than it does with other breast cancer subtypes. We don’t know exactly why this is, but nature is giving us a pretty strong clue. It suggests to me that these tumor types arise from a similar cell type of origin that has a particular function.

For basal-like breast cancer, some data suggest that it may arise from a cell with some stem cell–type characteristics that may impart some of its aggressive nature and metastatic potential. The basal-like subtype, ovarian cancer, and lung squamous cancer have significant similarities. Basal-like is not a subtype of breast cancer, but actually its own disease type that happens to occur in the breast with another disease type, ER-positive breast cancer. People had long assumed they were two different flavors of the same disease, but they are not. Instead of thinking of them as a red apple and green apple, we should really be thinking of them as a red apple and an orange.

How does this understanding of basal-like breast cancer impact how it is treated?

Cisplatin or carboplatin are standardly used to treat lung squamous patients and ovarian patients. Now that we have seen the link between those and basal-like breast cancer, we are considering if those treatments might be useful for basal-like as well.


How common is the basal-like subtype?

About 10% to 15% of patients with breast cancer have this phenotype. Basal-like breast cancer is more common in young women than older women, and it is more common in people of African descent than people of European descent.

You get a very diverse range of outcomes with basal-like breast cancers. Some are very responsive to chemotherapy, and that alone can cure these patients. Those that don’t respond have much worse outcomes and very aggressive disease. That is where we desperately need new treatments. TNBCs really have a unique genotype. It is different on every level. The 10% to 15% may sound like a minority, but because breast cancer is so frequent, TNBC by itself is about number five on the list of cancer killers of women in the United States.

What are the next steps in this research?

We are conducting and trying to set up trials testing immune checkpoint inhibitors for basal-like breast cancer. We are using mouse models of basal-like breast cancer and other types of TNBC to figure out what the immune cell and tumor cell interaction is in a controlled fashion. We want to learn about how the immune system responds to these tumors. We need to determine what antigens the immune system is mounting a response against. It would be really valuable to know what the immune system is specifically attacking.