WEDNESDAY May 14, 2003

“A Primer on Molecular Diagnostics: Revealing the Mystery of Proteomics and DNA Array Technology”

Our special guest for the evening is Mona Gauthier, PhD., a post-doctoral researcher at the University of California San Francisco, whose principle work is to contribute to the development and discovery of new biomarkers for cancer. The potential uses for biomarkers include help with diagnosis and prognosis, to determine the aggressiveness of a particular tumor type, as well as predictive markers that will allow us to classify certain tumors and develop specific treatment strategies. Until recently, all breast tumors were treated in the same fashion, with the only distinguishing features being their estrogen receptor status.  One of the new biomarkers, HER2/neu, is up-regulated in certain tumors and can be treated with Herceptin to inhibit the function of that tumor. The ultimate goal of micro arrays and proteomics is to tailor specific treatments to specific disease subtypes. In order to do that, we need to be able to understand and classify the differences in subtypes. If we understand something of the complexity of proteins, we will have a better understanding of why proteomics may be a way to more simply look at the differences in tumors that might be helpful for finding treatments and making diagnoses.

A little biochemistry. Genetic material is housed in the DNA in the nucleus of the cell. Micro arrays look at the expression of a specific cell at a specific time in space and are able to determine whether a specific gene is “turned on” within a specific cell, (part of the messenger RNA). It is then spit out of the nucleus and translated into a protein. The protein is made up of a string of amino acids known as a peptide chain that takes on a 3-dimensional shape when strung together with other peptide chains to make globular units. It is the globular units that are functional. At any given time there are probably about 30,000 proteins within the cell. Some proteins are found in the membrane. They act as the interface between the cell and the outside environment. Proteins found in the nucleus aid in communication inside the cell. We believe that cancer is caused by changes in the genes that make up proteins that make up our bodies. Small changes in a base pair can completely change the function of a protein. Genes are very small things; proteins—relatively speaking—are huge and easier to look at, so it may be easier to find differences on the level of proteins than the level of genes. This may allow us to subtype cancers more easily.

What is meant by communication among cells? Cells communicate chemically, influencing the behavior or makeup of neighboring cells. Proteins that let cancer cells move through the tissues more quickly aid Cancers. There are big protein differences between pre cancer, in situ cancer and invasive cancers. There are more mutations or changes once a cancer has metastasized. Tumors develop by making proteins that allow them to grow a blood supply.

Proteomics lets you look at a spectrum of proteins to see patterns or series of changes, because one change will not move a cell to display a different behavior. The challenge after identifying the patterns will be to determine the specific proteins that make up that pattern.

Are cancer cells and normal cells totally different in morphology? Cells take on different shapes as they become more malignant; a shape that allows the cell to move out of its normal environment. Low-grade cancers look more like normal breast tissue. The higher the grade the wilder the cell will look. The key difference between malignant and non-malignant cells is the pattern of growth, a collection of cells where there should not be a collection of cells. This is for the most part a visual diagnosis. The biggest change in cell morphology occurs when the cell becomes an in situ malignancy, increasing the chance that it will be able to invade someday. Metaloproteases, a class of proteins, seem to allow cells to invade through the stroma and the blood vessels. Targeting that single protein did not yield any positive results in clinical trials, but maybe targeting a group of proteins will allow us to find those that are critical for cancer growth.

It seems that cancer is in our body up to eight or ten years before it is detected. At what point in the growth can this process of change be detected and treated? If we can detect metastases early will it make any difference in survival? Certainly early detection is better, especially when the cancer is still in situ. We don’t know if early detection of metastases makes any difference in survival. Looking at tumor markers hasn’t altered survival. Very sensitive tests may change that in the future.

Does the configuration of an individual’s genes make it more likely that a cancer will recur? No, but it might make it slightly more likely that a new cancer will occur.  The chance of this is not very high; two highly improbably events don’t usually occur together.

If we have the same DNA in all our breast cells, why doesn’t cancer occur in all cells? The theory is that one cell gets into trouble and can no longer respond to the cues from other cells to stop growing. Why it doesn’t occur in a number of cells is not completely understood. Early radiation to one part of the body gives you a greater risk of cancer in that radiated field later in life. But people don’t get multiple cancers; they get one, so clearly single cells are being affected. It is just damage to the right cell at the right time that leads to the process of cancer. This is not a global event or the organism would not survive.

What is known about the breast cancer antigens that are being followed? After a blood sample is taken, the proteins are broken up into peptide strings and funneled through a mass spectrometer. A laser hits each sample. The time it takes for the sample to hit the detector is different for each peptide based on the charge generated by the laser. The detector will generate a set of peaks that becomes the signature of the sample. An unknown sample can then be tested against a sample from a patient known to have breast cancer or one known not to have cancer. In the future, we hope to be able to subclass these samples even further, i.e., estrogen sensitive versus estrogen non-sensitive, and then suggest specialized treatments based on signature profiles.

Are there any current trials that prove the validity of proteomics? No one marker can be linked to a particular outcome. This is the advantage of pattern recognition. The most advanced trials in this context are currently being done with ovarian cancer by the National Cancer Institute in collaboration with M.D. Anderson and UCSF. Data on the ovarian patterning trials are just beginning to come out, breast cancer is not there yet, probably within the next few years.

How useful are the serum markers for breast cancer?  Some cancers make proteins and some do not. The problem is that they are not very specific to a particular kind of cancer and they are not very sensitive. Some people can have metastatic cancer and have normal levels. The tests should not be used regularly for fear of “psychic damage”, creating anxiety. Elevated markers may mean recurrence someday, but we don’t know when, because there is no evidence. Also, we cannot tell if the cancer is in the breast or somewhere else. The idea of using tumor markers to identify tumors is being looked at, but proteins are not than good at binding just to the tumor. Current clinical trials are looking at CEA (carcinoembryonic antigen) in two ways. One is as a labeling radiologic tool to determine if CEA is present in an organ, usually the liver. Another effort is trying to use CEA-based vaccines and antibodies to CEA as a way to treat cancers that make that protein. This will be an enormous step forward in prevention, recurrence and individualization of treatment.

Any research into “souping up” the immune system? The results to date are not encouraging. One of the most important issues is being able to identify a single target protein. The greatest amount of research with breast cancer is with HER2/neu targeting vaccine.

What about finding the cancer early and hitting it harder with chemotherapy? Hitting the disease harder may actually damage quality of life and reduce treatment options over time without benefiting the patient.

Next meeting is Wednesday, June 11^th. Topic: “2003 ASCO: What’s New In Cancer Research?”