Gene expression patterns and breast cancer

The latest advances in cancer genetics have led researchers down a path that has the potential to alter oncology clinical practice. Using collections of thousands of human genes arranged on gene chips known as DNA microarrays, doctors may one day have the ability to distinguish between many different types of cancers based on the presence or absence of numerous proteins that are expressed within different cancer cells. The use of these DNA microarrays for this purpose is commonly known as gene expression profiling. Research concerning gene expression profiles is ongoing in both hereditary and non-hereditary forms of several cancers. Recent studies have examined the expression of genes in leukemias, lymphomas, melanomas, colorectal cancers, pancreatic cancers, ovarian cancers, prostate cancers, and breast cancers.

Current Breast Cancer Classification System

Generally, prognosis and options for the treatment of breast cancer depend upon the classification of a breast tumor in terms of its grade and stage. Diagnostic classifications of breast cancers have traditionally looked at microscopic characteristics of the cells themselves and at the presence or absence of cancerous tissue in the nearby lymph nodes. Although these characteristics provide important information about the breast cancer’s aggressiveness and its tendency to spread, limitations exist that can complicate predictions about outcome. Subtle differences in the microscopic characteristics of cancer cells can be challenging to distinguish. For this reason, there are some tumors that have ambiguous classifications. In addition, the examination of lymph nodes requires invasive surgery and cannot determine whether a cancer that has not yet spread is on the verge of spreading.

Other prognostic indicators that are currently in use in breast cancer, such as the presence or absence of estrogen, progesterone, and human epithelial growth factor (HER2) receptors on cancer cells, examine only a few markers of disease category at a time. Because of this, the amount of information that can be gathered for making predictions about outcome and response to treatment is limited.

It is clear that patients would benefit from a more accurate classification of breast cancers.

What is gene expression profiling?

Normal cell function is dependent on the availability of a variety of proteins. Different types of cells, for example skin cells, nerve cells, and liver cells, perform different types of jobs and therefore require different types of proteins. All proteins are made from the expression, or reading, of instructions found on the DNA of the genes within each cell. Although all human cells contain the same genetic information, different genes are expressed depending upon a certain cell’s functional needs. For example, when skin cells are damaged, proteins that can trigger the growth of new cells are initially made, followed by other proteins that can trigger a slowing of cell growth as a wound heals. Thus, an examination of the expressed genes from any given cell can provide a sort of fingerprint that is characteristic of the growth, development, and even death of that particular cell type.

When tumors form, DNA changes are found in the tumor cells themselves. These affect the expression of protein-forming genes, which in turn affect the function of the cell. Because of this process, tumor cells will have different protein (or gene expression) profiles when compared with normal cells. In addition, different types of tumors can behave in different ways and are expected to have their own personally identifiable gene expression profiles.

To create a gene expression profile, researchers remove the expressed genetic material (that material which will become a functional protein) from a tumor cell for analysis. This expressed genetic material, in the form of messenger RNA (mRNA), is then labeled with a fluorescent marker so that researchers can detect it. The next step involves the use of DNA microarray technology. Simply put, DNA microarrays are ‘chips’ that contain more than 6000 normal human genes attached in a grid-like pattern. The labeled mRNA from a tumor cell is applied to the microarray ‘chip’. Then, like a child’s matching game, the labeled mRNA’s find their matching genes on the chip and ‘stick’ to them once a match has been found.Next, based on the location of the fluorescent signals on the microarray ‘chip’, researchers can determine which genes were expressed in a given tumor cell.

Cells with similar properties are expected to have similar fluorescent patterns, or gene expression profiles, whereas dissimilar cells are expected to have dissimilar profiles. Because thousands of genes are in question, statistical programs run by computers are being developed to distinguish unique profiles.

Why is gene expression profiling important?

This emerging technology has been used in a number of studies involving various cancers in the last few years. Regarding breast cancer, researchers have already used gene expression profiles to distinguish between tumors with different estrogen receptor status – and, they have done this with greater than 90% accuracy. There has also been some success in distinguishing between cancers that have and have not yet spread to nearby lymph nodes. More recently, scientists at the National Human Genome Research Institute, led by Dr. Jeffrey Trent, used this technology to identify differences between breast cancers caused by inherited mutations in BRCA1, breast cancers caused by inherited mutations in BRCA2, and sporadic breast cancers. “These findings illustrate the potential diagnostic power of gene expression profiling. We hope the technology will translate into improved patient care treatments, for example, by identifying those patients best suited for a particular therapy ” says Dr. Trent.

In spite of numerous recent research efforts, gene expression profiling has not yet been introduced into clinical practice. The existing studies show that gene expression profiles can help distinguish between tumors with previously known clinical characteristics. This, by itself, is not particularly helpful because there are already methods in place for making such distinctions. More promising is the presence of a few studies that have attempted to identify new classes of cancers based on the results of gene expression profile analysis. However, large studies are still needed that link different tumor gene expression profiles to particular clinical outcomes such as survival and response to treatment.

As DNA microarray researcher Leping Li of the National Institute of Environmental Health Sciences at the National Institutes of Health has proclaimed, gene expression profiling may one day ‘revolutionize cancer biology’. Although it is unlikely to replace our current methods of cancer classification, it is possible that we will soon see this technology used in conjunction with currently accepted methods. It might prove to be particularly helpful in cases that are ambiguous, and may help physicians provide more accurate information about prognosis to their patients. In the future, it may also be possible to target breast cancer treatments based on the particular set of genes that is expressed within a given tumor. Finally, scientists may gain a wealth of information about the ways in which different cancers develop and progress by examining the types of genes that are expressed in cells at different stages of breast cancer.

Researchers at Johns Hopkins are involved in ongoing efforts to create cancer tissue DNA microarrays for the future study of gallbladder and bile duct cancers. Similar projects focusing on pancreatic and prostate cancers are also underway. In addition, there are gene expression research studies being carried out at the National Institutes of Health (NIH). For information on current studies at the NIH, visit the NIH clinical research studies website at http://clinicalstudies.info.nih.gov.

Lori Hamby, ScM

References:

Gruvberger S, Ringner M, Chen Y, et al. (2001)Estrogen receptor status in breast cancer is associated with remarkably distinct gene expression patterns.Cancer Research 61(16): 5979-5984.

Hedenfalk I, Duggan D, Chen Y, et al. (2001) Gene-expression profiles in hereditary breast cancer.NEJM 344:539-548.

Li L, Pedersen L, Darden T, and Weinberg C (2001)Class prediction and discovery based on gene expression data.http://dir.niehs.nih.gov/microarray/datamining/public_html/cbgi.pdf -

Ramaswamy S, Tamayo P, Rifkin R, et al. (2001)Multiclass cancer diagnosis using tumor gene expression signatures.Proc. Natl. Acad. Sci. 98(26):15149-15154.

West M, Blanchette C, Dressman H, et al.(2001)Predicting the clinical status of human breast cancer by using gene expression profiles.Proc. Natl. Acad. Sci.98(20): 11462-11467.