Audience: I passed around a petition to sign for an ovarian cancer awareness stamp. There are plenty of sheets, so if you guys would all sign before you leave today it would really help in getting that stamp passed.

Michael Birrer: Let me start by thanking the Gynecologic Cancer Foundation and David for their gracious invitation to me to be here today to talk to you. It is really an honor and pleasure for me to have a chance to chat with you.

The topic I was asked to cover was the molecular biology of ovarian cancer. And if there is nothing that you take home from my lecture other than my belief that we are on the cusp of a relatively dramatic increase in our knowledge and understanding of the way this tumor develops that will have a fairly dramatic impact on the way we handle this tumor clinically over the next five years, if that is the only thing you take home without any of the details, then I would consider my talk a success.

I think that the way in which we will clinically impact this cancer is by understanding the way in which it develops. Ultimately knowledge in this case will cure. Now why am I so optimistic? Part of it is that we have undergone an enormous revolution in biotechnology in the last 20 years. There have been a lot of achievements in terms of our understanding of genetic structure, its contribution in human disease, and I have just put up here one milestone, which you are all familiar with. This is the Science magazine from 2001 that declared that the human genome had been sequenced. The bottom line is we now know essentially all of the genes in a human or mammalian cell. We are trying to work out the structure and the function, but the truth of the matter is we are in what most people now refer to as the post genome era. I believe that as this is applied to the problem of cancer, in particular ovarian cancer; you are going to see some really exciting developments that will have an impact in the clinic. Now some of this data has already been accumulated in cancers in general over the last twenty years. We now know that human cells grow, but they grow under very regulated conditions. For example, here is a normal cell, it has a growth pathway, the growth factor binds to a protein on a cell surface, it stimulates signaling enzymes. This tells the nucleus to do things and the cell will grow. Very tightly regulated. This is what I would call the gas pedal. The more you push this system the more the cell grows.

On the other hand there are the brakes, which are what we refer to as tumor suppressor genes. These inhibit these growth pathways. All cells have them. All normal cells have these brakes if the are functioning normally. We get a cancer cell by one of two processes. One is that there is too much gas, meaning that those pathways, which stimulate growth, become mutated, and you can see that over here. This is a mutated oncogene and you get uncontrolled growth. Or, you lose the brakes. So here you have the tumor suppressor gene and it's missing and you get a cancer cell.

Pretty well worked out for most epithelial cancers. It's now being translated into our understanding of ovarian cancer. So really it comes down to a question of how do you go from something like this, which is a relatively well-behaved ovarian epithelium, to an advanced adenocarcinoma of the ovary? We now know that this is not an instantaneous process. It most likely takes some time. It is not clear for ovarian cancer how long it takes. It may be shorter than other tumors but it still takes some time. It is probably a multi-step process. It starts with a cell that looks pretty normal, it mutates, perhaps, a suppressor gene, the cells begin to grow and then you go through a series of mutations in other genes and eventually you get cancer.

The "take-home" message, and this is something that's often said but probably can't be said enough, is that ovarian cancer is a genetic disease. I will repeat that. Ovarian cancer is a genetic disease. There may be things in the environment that we can change to modulate the way in which the tumor will appear or will grow. But the truth of the matter is the mutations within genes gives rise to ovarian cancer.

Over the last five to ten years, laboratories across the United States and other parts of the world, some laboratories represented right here, have worked very hard at trying to identify what those genes are for ovarian cancer. I would like to give a quick summary of it. I only have about ten minutes, so write fast. Here are some key players. What I really want to emphasize is that after I give you the list I want to tell you why it's important to know this list, because it's going to have clinical impact. So here are the pedals. Remember the gas pedal gene. These are the ones when they are over expressed the cells grow.

We know that there are cell surface receptors primarily in the EGF receptor family. Including ErbB one, two, three, and four which are over expressed in ovarian cancer. Presumably, they play a role in the genesis of this disease. Next are what we call signaling proteins. These are not located in the membrane; these are in the cytoplasm of the cell. Work at UCSF, MD Anderson and in South Florida have identified an enzyme called PI3 Kinase, which is an important player in the genesis of this disease. Work by Joe Tessa and Tom Hamilton at Fox-Chase has identified AKT2 as a gene that is amplified and over expressed in ovarian cancer. This is also an important gene.

Work that we've done in our lab showing a decrease in a phosphatase that resulted in an increase in this whole signaling pathway is probably important in ovarian cancer. And lastly, I just want to point out that the proteins don't have to be in the membranes or in the cytoplasm. They can be in the nucleus of the cell. For instance, Cyclin E is an important critical protein for driving the cell to multiply and recent data shows that this gene is amplified in ovarian cancer.

Let me quickly show you what I mean by this. Here is an ovarian cancer cell that has been immunostained using an antibody to one of the EGF receptor family members. These are the cells and all of this brown staining is over expression of ERB2 (Her-2-neu). This would be a high expressor and the concept would be that the growth of this tumor is at least in part driven by this protein.

Here is our friend PI3 Kinase. This is a western blot. I don't expect you to know this. The bottom line is, let me explain very simply, this band here is the enzyme. These are all cancers. It's here, and then you go to the normal ovarian epithelium, it is very very low, and that's because in several of these specimens the gene is actually amplified. You have more than two copies. This is a slide that was loaned to me by Dr. Gordon Mills from MD Anderson.

Again, MKP1 is the phosphatase. As I described before this protein is down regulated in ovarian cancer. It disappears in ovarian cancer and I demonstrate this here. This is actually the protein in a normal ovarian epithelium, and a Stage I tumor. When you get to advanced tumors it completely disappears. This difference in expression we think is critical for the progression of the tumor.

Finally, another example of a gene that plays a role in ovarian cancer is Cyclin E. What you're seeing here is fluorescent in situ hybridization. You're actually looking at the gene itself within the cell. Of course since I am colorblind I can't really see what I'm looking at. But, I can read. So, there are red dots here, but the key is there are a heck of a lot more red dots in here, and this is a reflection of the fact (well trust me I'm from the government) in this particular sample that Cyclin E is amplified to about eleven copies per cell. And, that is not normal and that is driving the cell. Now, the critical question, particularly from a patient's standpoint, and from a physician's standpoint is, what does this all mean?

Let me remind you, I mentioned the brakes on the system--the tumor suppressor genes. There has been a fair amount of work done here. In ovarian cancer we all know the role of BRCA 1 and 2 inherited ovarian cancer; there are mutated DNA repair genes in HNPCC, whose patients also have a slight increased risk of ovarian cancer. The major player for what I call sporadic ovarian cancer is P53. P53 is a major tumor suppressor gene and is mutated in the vast majority of ovarian cancers and it looks like this. Here is some stroma, here is a normal cyst. You should see no staining here for P53, but this tumor has a mutated P53 and it shows up gangbusters. So let me go back then.

What does allof this mean for the patient? What does it mean for us in the clinic? This is where the excitement is going to be. In the next couple of years you are going to see a lot of these discoveries translated into the clinic. What I mean by that is that the applications of molecular biology are really protean in terms of clinical practice. They are involved in early detection screening, prognostic markers, markers of disease activity. Some of these markers can tell us if the disease is coming back, or if it is going to go away. And finally, they play a role as targets for a lot of novel interesting therapy, some of which you will hear about in a few minutes from some of the other speakers. Let me give you a couple of examples. Go back to our cell surface molecules. These are clearly good targets for novel therapies. Samir Khleif will talk a little bit about vaccine development, because these are proteins that are exposed to the immune system. Of course, if they are shed into the blood stream, they could be markers for early detection or recurrence. I will give you another quick example, but I think one of the other speakers will talk in more detail about this. We have all heard about HER/2neu cell surface protein in the EGF family. If you have one copy, the cell grows normally. If you have multiple copies it becomes a cancer cell. Several years ago through the work of a lot of laboratories and drug companies a monoclonal antibody was developed which, in fact, when combined with this receptor can inhibit cell growth.

The story for ovarian cancer with Herceptin is more complex than that. I've talked about this before. We did a large study with GOG9404 that actually examined 139 Stage III and Stage IV cancers from a randomized prospective trial. And, the take home message is the amplification of this gene and the expression of the protein is extremely low. So I don't think that the use of Herceptin in ovarian cancer would be something that would be used frequently, at least as a single agent, but there is optimism, and again it comes from the science. What is interesting is if you look at those same specimens where you can see Her-2-neu standing you could also see another family member, EGFR. In fact, if you total it up, these are the total amounts of staining for Her-2-neu and EGFR. This is Her-2-neu and this is EGFR, add this all up and you are up around 60 percent. So it may be that while we can't necessarily use Herceptin by itself, we have a panel of monoclonal antibodies we might be effectively able to approach this disease with.

Before I finish, I would just like to mention a couple of other applications. Nita Maihle from the Mayo Clinic has looked at secreted forms of ErbB protein. Some of these receptors actually break off and flow into the bloodstream. Their levels correlate with disease activity. They may be very helpful in diagnosis and early detection. This is an area of very interesting research. If we go inside the cell into activated pathways, remember I mentioned some of our friends like PI3 Kinase and Cyclin E, these could play a role as prognostic markers and also targets for small molecule inhibitors. This is a prognostic marker study that we did with the GOG. We were looking at Cyclin E. Cyclin E, in fact is amplified. In those patients, who have a single copy of the gene, this has a considerably longer time to disease recurrence than those who have amplified genes. In fact, it translates into almost eleven months of disease for survivors.

So let me just finish up and say that for some of the enzymes I quoted, these are actually inhibiting molecules that are available to us in the lab. PD stands for Park-Davis, LY stands for Eli-Lilly. These molecules have been made. They inhibit the enzymes. These enzymes are overactive in ovarian cancer. My prediction is these will come into the clinic; a version of these will come in to the clinic in probably a couple of years.

Finally, back to our brakes on the system, our tumor suppressor genes. Can we deal with tumor cells that are missing tumor suppressor genes? The truth of the matter is there may be some clever ways to do this. They are mutated proteins so they may be targets for immunotherapy and vaccine. It is also possible through gene therapy that we may be able to replace the growth inhibitory activity, and there is active research in both of these areas.

So in conclusion, I hope that in the next five years we see a translation in some of these very interesting scientific studies in to the clinic and you will see a dramatic change in the way in which we approach this disease, and we hopefully will be more successful. Thank you.

David Mutch: We are running a little behind. We will take some questions at the panel discussion time if that's okay. The next speaker will be Samir Khleif who is head of the vaccine portion of the NCI.