DR. ELLIS: If you have any questions for Bob and myself, we can get the discussion going and as soon as David gets on line, we will switch to his presentation.
Based on some of the data that I talked about with Cox-2 inhibitors, would anybody be interested in combining standard therapy for gastric therapy with Cox-2 inhibitors as anti-angiogenic, as perhaps pro-apoptotic in perhaps a phase II study to look at the feasibility and to look at surrogate end points. Is that of any interest to the audience?

DR. LENZ: Actually, we are starting a colon cancer trial where we combine celebrex with 5-FU and CPT-11. I think this is a very interesting approach because of its role of Cox-2, not only in cancer development, but probably also chemoresistance.
With the data shown today, I think this makes even more sense to do that and follow that up. Absolutely, yes.

DR. RADINSKY: Is anyone aware of any studies currently in progress or being planned using specific small molecule inhibitors to EGF receptor for gastric carcinoma?

DR. LENZ: There is an LOI out from CTEP, for OSI-774, which includes GE junction tumor. I am sure that there is a variety of proposals coming in for GE junction and other GI cancers such as HCC, lung cancer and so on, but including GE junction tumors.

DR. ELLIS: Bob talked about all the exciting things that happened years ago with the pancreatic cancer, and that there is a lot of basic science that has come out of that. One of my frustrations, in the clinic, is that I haven't seen a lot of the biologic agents be combined with chemotherapeutic agents in the clinic. Of course, we all know about the anti-VEGF antibody trial with standard chemotherapy for colon cancer as well as SU5416 with standard chemotherapy. Now there are some studies with CPT-11 combined with C225. Basically, considering the breadth and the depth of the biologic agents and the companies that are willing to push them forward into the clinic, I think there is a relative paucity of the clinical trials where we are actually combining biologic therapies with standard chemotherapy. The challenge is, how can we move forward in, first, phenotyping tumors and then using these biologic agents in clinical trials, after we have selected patients who may be good candidates for this therapy. Any thoughts on how we can move this program forward?

DR. LENZ: I think one of the questions you raised, and the problem, is that since we don't even have a standard care of chemotherapy, we are actually dazzled what kind of chemotherapy would be more successfully. As you know, we are working very hard to establish molecular markers to predict chemotherapy for gastric cancer outcome.

I think, in concert of these kind of studies, we should include the new agents and add it to it. As you can imagine, the clinical trial design where you have clinical markers being tested as a predictor of response and survival in addition to new agents which have been not tested created probably a lot of headaches to statisticians and clinic trial designs. I think we just have to start out with some of these prospective clinical trials using the data we have generated in the laboratory to test it out.

I think actually gastric cancer is an even better model than colon cancer, because it could be also an advantage not to have standard chemotherapy. Here, you have the opportunity to actually choose chemotherapeutic regimens without having the discussion of being less suboptimal to standard of care.

Now, our colleagues in lung cancer have actually gone the next step forward. They actually are launching an international trial where they use beta tripoline mutations and ESCC-1 as a decisive marker to do chemotherapy based on taxol or sisplatin.
I think we should do the same thing for gastric cancer. Now, how we include the new data with the angiogenesis markers, or new agent and small molecules and Cox-2 inhibitors, I think we should find a way to do that. First step first. I think we should do the trials which have been started in lung cancer in a similar fashion to gastric.

DR. MAYER: I would just throw in one comment. This conference is sponsored and organized by the GI Intergroup. In large part, many of the newer and more interesting molecules have been initially studied outside of any cooperative group setting.
The intergroup at the present time does not have studies for advanced gastric cancer, for a variety of different reasons. The more that some of these new molecules can come through an LOI through the NCI and become available to the Intergroup for study with a larger base of patients, and a more objective way of defining outcome, so be it. I think that would be a major step forward.

DR. ELLIS: Is it up to the NCI to initiate the process by requesting LOIs? Does anybody from the NCI want to comment on that? Nobody from the NCI wants to comment on that. I think the timing is good because David McConkey is on line. We are ready for your talk.

(Following Dr. McConkey's presentation) DR. ELLIS: That was great, David. Are there any questions for David while we have him on the line? David, your talk complemented some of the other talks. I also talked a little bit about Cox-2 and its role in gastric cancer. If you were to look five years down the line in terms of looking at gastric cancer apoptosis, could you identify a category of compounds, or a particular compound, that you would like to see in a clinical trial?

DR. MC CONKEY: I certainly would love to see more done with the anti-inflammatory agents. I also think that we need to try to look at the death receptors a little bit more carefully. Certainly, as I mentioned, the etiology of gastric cancer probably selects for death receptor resistance. For instance, TRAIL seems to have substantial anti-tumor activity in some of the xenograft studies that have been conducted to date.

I like the idea of trying to combine some of the biologicals with some of the more classic small molecules. I guess with respect to that latter category, I am particularly excited about some of the compounds that you and Bob are working on, some of the growth factor receptor inhibitors, especially perhaps the ones that get inside the cell and inhibit many of the family members.

Also, I like the cell cycle active agents, the CDK inhibitors, the UCNO-1. We are working with a drug called PF341 that inhibits the proteosome, and also definitely is a cell cycle active agent. I think, in the end, it would be useful to try to combine modalities that target these very discrete pathways independently. So, five years from now, I hope we know enough about the biology of this tumor to understand if that kind of assumption is correct.

DR. ELLIS: Would that require phenotyping of the tumor before treatment or would that be for all tumors?

DR. MC CONKEY: Well, either phenotyping a tumor before treatment, or perhaps -- I know you are not a big fan of this -- trying to see if the therapy is working soon after it is initiated. That depends upon tissue accessibility and the availability of biomarkers that are reliable. I would certainly like, in the short term, to understand if we are actually hitting our targets and if, in hitting our targets, some of our predictions about pathway resistance hold up or fall apart in the patients. Any other questions for David before we let him go? David, that was great. We appreciate your participation. Any other questions for Bob or myself?

DR. MACDONALD: Just one. I sort of lost the role of beta fibroblast growth factor. The reason I ask it, it is interesting. We do have agents that act as inhibitors of beta fibroblast growth factor, particularly suramin and other hepranoids. I am just curious. Again, I sort of lost in the morass what it was.

DR. ELLIS: You know, basic fibroblast growth factor, as you know, was the first angiogenic factor identified by Judah Folkman and Michael Klagsburn, back in the 1980s. It has fallen out of favor and it waxes and wanes. With identification of other angiogenic factors which seem to be more of a driving force, such as VEGF and second, on the list right now I would rank the angiopoietins, I think basic FGF has become further down the list, and it is somewhat confusing. In certain tumor systems, low levels of basic FGF have been associated with tumor progression. So, it is confusing.

What was also somewhat surprising in our studies was that we did not find any of the four receptors for basic FGF on endothelial cells in gastric cancer. Yet, the receptors were on tumor cells. Even though there are some non-specific inhibitors for basic FGF, I don't think that those compounds are ready to go into the clinic, and the biology is interesting, but it didn't pan out.

DR. FUCHS: I was just curious what your thoughts were. We obviously have an example with GI stromal tumors of a dramatic effect of C-KIT inhibition, but that is really probably the only solid tumor where targeted agents have really had a dramatic response. Probably the other agents will have some marginal benefit on top of our existing therapies. Do you think it is possible, in gastric cancer, that we could achieve some similar novel therapeutic agent that could have the same kind of effect?

DR. RADINSKY: Probably not, because it is not a single mutation. I think what is more encouraging is the possible combination use of the biologics with the cytotoxics. What you see in vivo are not only additive, but synergistic effects. So, when you combine the cytostatic, anti-EGFR agents with radiation or chemotherapy you get much better effects than either agent alone. I think that is more encouraging.
Now, who knows? Will we discover one key mutant gene protein that leads to gastric carcinoma progression, metastasis, et cetera? It is highly unlikely, but the combination approach is exciting.

DR. O'DWYER: Again, going back to that model of the C225 with gemcitabine, and the difference between the in vitro effects where you saw maybe a 20 percent decrease in growth rate versus the very striking in vivo responses, and you showed that there was clearly an effect on angiogenesis which was really interesting, but the other thing that you looked at was the cell cycle related end points. In the in vivo model, if I read it right, the PCNA went from 70 percent down to around four percent in the combination-treated group, and maybe down to 20 percent in the C225 alone-treated group. Do you think there is something more in the cell cycle effects in vivo than there is in vitro? Is that explaining part of the major difference in vivo?

DR. RADINSKY: Potentially it could. We know, in vitro, that when we treat with C225 or a small molecule, that the cells stall in G-1-S. When we remove the agents, the cells then kick back into the cell cycle. In vivo, what I think is happening, you are getting the combination of both the cytotoxic and the cytostatic, whereas the cytostatic may be working in the G-1-S, where the cytotoxic may be stalling the cells in G-2-M. Possibly that combination may result in the observed effects. It is currently a hypothesis right now.

DR. MANSFIELD: In those studies, did you remove the agents after a period of time and then looked to see if there was a delayed growth, or were all animals sacrificed? Did you do separate studies?

DR. RADINSKY: That is an excellent question. That is one of four individual experiments. We did not test your question in a rigorous manner, but we did have a number of mice that we took off therapy and the tumors actually regrew. With the agent alone, the anti-EGFR agent alone, tumors regrew. We could put them back on therapy. I think Christiana put one or two animals back, and they did come back down. In the combination group, when we looked histologically, you had very, very small microfoci of tumor cells. Now, that was after four weeks of therapy.
If we had continued on, would we truly abolish all tumor cells? We don't know the answer to that. That would be something to go back and look at.

DR. SCHWARTZ: I think Peter hit it on the head with that comment regarding those studies. I think there is a growing body of data that cell cycle arrest does not equal cell death. Cell cycle arrests will cause cytostatic effects, but with withdrawal of drugs, cells will then resume growth. It appears that cell cycle arrests, however, in the presence of a DNA-damaging agent, will be sufficient to tip the balance in favor of apoptosis.

I wouldn't be surprised if that is what is happening with the combination of the antibodies plus chemotherapy. I will be talking more about this tomorrow, actually, in my talk on apoptosis induction with combination therapy. I think that really probably does explain it, Peter. I wouldn't be surprised. The growth arrest does not equal death, but with chemotherapy it can. I think this leads to new directions in cancer therapy, and gastric cancer, too.

DR. ELLIS: I think that is probably a great ending point and introduction to tomorrow. We will end this session and reconvene at 3:00 o'clock.
[Brief recess.]

TOP