Identifying a signature of outcome for breast cancer patients
Breast cancer is the second leading cause of cancer-related death in women in the United States. There are several types of breast cancer, with the most common being estrogen receptor positive (ER+). This means that the cancer cells grow and divide rapidly in response to estrogen, a hormone that has fluctuating levels throughout a woman’s life.
Estrogen is one of the hormones that controls breast development during puberty and pregnancy, but when damaged cells respond to it later in life, this can lead to accelerated tumor growth in ER+ breast cancer.
The growth advantage that these tumor cells utilize is also a key to their demise. Physicians use this mechanism to prevent ER+ breast tumor growth by starving these cells of estrogen. There are several drugs that are widely used to do this, with tamoxifen being the most well-known example. This is a more targeted therapy than chemotherapy, which prevents all of the body’s cells from dividing by damaging their DNA nonspecifically.
While tamoxifen and other endocrine therapies have been very successful in the treatment of ER+ breast cancers, the downside is that there is a high rate of developing resistance to these drugs. Almost 30 percent of women who are treated with tamoxifen will have a recurrence of their tumor within five years. Currently there is no way to identify which patients will have a recurrence early on in treatment. The way that physicians determine which patients receive endocrine therapy is by taking a biopsy specimen, cutting into it, and staining for the presence of the estrogen receptor in the nucleus of the tumor cells. If there was a way to match a breast cancer patient to the best-targeted therapy for her tumor, side effects of nontargeted drugs would be avoided and lives would be saved.
A collaborative effort between researchers at the University of Illinois at Urbana-Champaign, Dr. Benita Katzenellenbogen, an expert in breast cancer endocrinology in the Department of Molecular and Integrative Physiology, and Dr. Rohit Bhargava, a chemical imaging expert in the Department of Bioengineering, resulted in a multidisciplinary approach to identify the reasons why endocrine therapies fail. They also designed a method to determine which patients would be long-term responders at the time of the first biopsy. The Illinois team's research was published in May 2014 in PLOS One.
They hypothesized that, because estrogen affects normal breast development through interactions with the non-glandular breast tissue, or stroma, there must also be a role for the stroma in response or resistance to endocrine therapy. In order to approach this problem, a novel model system was developed that combined ER+ breast cancer cells and normal mammary fibroblasts, the cell type responsible for maintaining homeostasis in the stroma, in a three-dimensional culture system. This allowed the cells to grow together in a more realistic environment, while still maintaining a controlled experimental system.
What the researchers found was that, when the breast cancer cells were grown in close proximity with fibroblasts, the growth of the tumor cells was no longer dependent on signals from estrogen, even though the estrogen receptor was still present when the cells were stained. This finding suggests that there are patients who will be diagnosed as having an ER+ tumor but they will not respond completely to endocrine therapy. Furthermore, the tumor cells did respond to factors that are markers of inflammation, which might lead to these tumors becoming more aggressive.
After the identification of this novel mechanism for resistance of some breast cancers to endocrine therapy, the researchers began developing a method to determine which patients will have tumors that respond this way. This approach was two-pronged: a gene expression profile was determined, and a label-free image-based method was developed.
First, the factors involved in the tumor-stroma communication were isolated and identified. A signature of 46 secreted factors that were significantly higher in the endocrine-resistant cells was defined. The advantage of identifying secreted factors is that the signature eventually may be detected in patients with a simple blood test, as opposed to taking a biopsy sample. The factors present in the signature are secreted by both the tumor cells and the stromal cells, which confirms that both cell types are involved in regulating the growth of breast tumors.
A genetic signature, called iSig, was also derived from this data. This allowed the researchers to compare the expression of iSig among different groups of breast cancer patients using publically available genetic data derived from hundreds of patients, testing the universality of their data among patients from all over the world. This confirmed that patients who had high expression of the iSig also had more invasive and aggressive tumors.
The secreted protein signature and gene signature iSig may be used in the future to identify patients who are likely to become resistant to endocrine therapies. However, the current gold standard for breast cancer prognosis and therapy is through the analysis of biopsy specimens. Although current methods can identify the presence of certain proteins that are correlated with therapeutic outcomes, such as the estrogen receptor, these methods cannot tell the pathologist if those proteins are functional. If a patient was expressing the estrogen receptor, but had a tumor whose growth was regulated by the stroma, as the researchers showed using their experimental system, she would likely be placed on endocrine therapy even though her tumor may not respond in the long term.
Professor Bhargava is an expert in spectroscopic imaging, where the inherent chemistry present within a specimen can be visualized without the use of dyes. His group has used this method in the past to improve current pathology methods, aiding physicians when they look at tissue samples and providing them with additional information, such as cellular state and tissue composition. Profs. Katzenellenbogen and Bhargava used one type of spectroscopic imaging, Fourier Transform Infrared (FT-IR), to look at breast tumor biopsy samples from patients with varying levels of ER expression. They found that the imaging method was sensitive enough to distinguish between the different groups of patients and surprisingly, that many of the chemical differences that were identified were found in the stromal tissue of the tumor. These preliminary results suggest that FT-IR imaging may be used to complement the gold standard pathology method in order to aid the pathologist in making decisions about which therapy may be best for the patient and minimizing the chance of recurrence.
This collaboration between Molecular and Cellular Biology and the Beckman Institute for Advanced Science and Technology at Illinois has led to a multifaceted approach to identifying both a novel mechanism of therapeutic resistance in breast cancer and also a preliminary method for identifying patients who will become resistant to therapy at the time of their first biopsy. This method could potentially be used to decrease the rate of recurrence and improve patient outcome.