DR. BORDEN: Questions? That was an excellent summary.

DR. HEINRICH: Can I make a comment? We discussed this quite a bit and we really couldn't make a recommendation, but I think it is worth throwing out there, is that there may be the conceptual issue -- and you could say that ALK for IMT and anaplastic large cell lymphoma is the test case in which you have a compelling target that met all of the group A -- Jonathan's -- criteria, but may never be a target for a drug unless it happens to be a side effect activity of some other drug.

So, one developmental thing or programmatic thing that we have to think about in the future is that, if we develop these targets and no pharmaceutical industry cares about them, what do we do at that point? What kinds of creative intellectual property agreements can we make as organizations or the government? Or what kind of library resources for inhibitors could we do?

It is great -- we all believe that if we understand the biology, we will have a treatment. But what if that is not true -- what if we understand the biology and nobody cares about developing a drug?

That is something I don't see that we solve today; but if we make this big infrastructure -- and especially sarcomas which are rare -- we all know that, and if we talk about a drug that only works in a subset of sarcomas, if it doesn't work in other disease, what do we do with that?

DR. VERWEIJ: In Europe, we wouldn't have any system in place to help you there, apart from the clinical trial system that we do have. I think here, NCI may be facilitating the possibility to develop drugs like that. I don't know whether hairy cell leukemia is that frequent anywhere. That is another example of where we have another drug for a very rare disease, and subsets of soft tissue sarcomas may be like that as well.

I fully agree. I mean, if we can get some support there to develop agents for a very rare disease where we understand the biology of it, we should pursue that.

DR. BELL: Jaap -- just to comment -- one of the issues that may be more specific to sarcoma is that developmental signaling pathways that are active during limb generations may become reactivated.

That seems to be true in chondrosarcoma with hedgehog signaling, fibromatosis. I am wondering whether we should be trying to develop links with developmental biologists who have expertise in the area, understanding which pathways are active, during limb generation, to see if that paradigm of recapitulation of limb-generating signaling pathways is a more general effect in sarcoma development.

I think that is one area that, with the development of cDNAs, sometimes you pick up unexpected signals related to signaling pathways that offer potential for interaction with drugs that are already known and available.

This is not an area that is in the forefront of research right now, and I wonder if it should be something we should push.

DR. VERWEIJ: I agree, we should be pursuing that as well.

DR. BAKER: Also, stem cell biology is a very fundamental biology to this as well.

DR. BORDEN: Going back to Dr. Heinrich's comment -- I think it may be worth considering some kind of thoughts that relate to pointing out how sarcomas may elucidate some important pathways. This may not be something the industry will pick up on initially. It may be something that we need to think about as a recommendation that would be a scientific recommendation.

DR. VERWEIJ: This is something that NCI, from what I can see, could be of support.

DR. BAKER: We may not want to frame it directly for the NCI, but this is the kind of thing that is needed, I think -- might be very helpful -- because I think Dr. Heinrich did bring out a point that I heard a lot of discussion about in the group yesterday.

DR. BLUM: Larry, the group as a whole, one of my concerns is the ability to operationalize, at this point, certain things that came out in the meeting. I mean, it is clear that we need to look at hypotheses and test them in specific cell types. Jaap, in your comments it seems as though there is the opportunity to operationalize in a couple of diseases.

Certainly, the issue of synovial cell and the opportunity, perhaps, to take an EGFR inhibitor, which there are several out there, to move forward in a collaborative way just as we did with the GISTs. The issue of taxanes and angiosarcomas, I think, is certainly something that could be tested in a formal way through the Cooperative Group's access to the drugs and the data, and these are diagnoses that are fairly unequivocal.

So, my question, then, is, what are we doing, what can we do, what should we be doing to recommend operationalizing clinical trials with the appropriate biomarkers?

DR. VERWEIJ: I think to start that, you need a clinical trial consortium to do it. We started an EGFR targeted study in synovial sarcoma in EORTC. That is a fairly common subtype that we can study within this large group.

If you want to study angiosarcomas, then it gets even more difficult. So, I think we really need an international consortium to be able to do studies like that.

DR. BLUM: In the response to Gleevec, it was clear that we could organize, at least in the United States and certainly in Europe, an Intergroup protocol that had a very short lead time. Why can't we do that at this point, using existing structures?

DR. BAKER: I think Ron that one of the important factors -- history depends on who survives and who gets to write it. The reason that, at least in large part in my mind, why the Gleevec trial was such an enormous success is that many of the clinicians in this room and not in this room have large populations, relatively large populations of patients that we had heretofore said had no therapy.

It is a miracle of communication that this data became exploded in front of us and said, here is a disease and here is a new drug, and the only way you get this drug, in this last trial, the intergroup trial, was by entering patients into that study. That was, in my mind, a major motivator. No clinician could, in good conscience, not do whatever it took to get that drug. In part, that trial was designed to make available, on a compassionate basis, this drug to the American public and to the European public.

Very similar, Sharon said this earlier today, in a previous Intergroup activity, when ifosfamide was experimental, it was phenomenally easy to do clinical trials in sarcoma to get your hands on ifosfamide.

When ifosfamide became commercially available, that was a major impediment to going forward.

In part, the answer to your question is being able to have access to a drug that is generating extraordinary excitement among the clinical community. They will do almost anything, which I think is a great testament to physicians in general.

DR. DE MATTEO: That is true now for EGFR. There is none commercially available at this point, so that might be a strong statement to make.

DR. BAKER: That is why the CALGB trial and the bcl anti-sense -- those people who want to get access to that anti-sense molecule have no choice. I think that to tie back to something that was said before, about access of drugs, I have literally begged the Pfizer Corporation which has a pan-HER inhibitor. So, it is inhibitive of EGFR, HER2, HER3, HER4, to look at it in synovial sarcomas.

They said, no, we are going after breast cancer, lung cancer. Maybe that is why they make billions of dollars and we are all here.

DR. SAXMAN: Just let me comment, just quickly, on what Ron is saying. I don't want to get in the middle of this argument either at this point in time. I am not sure it is the appropriate place. I would say that one of the things that I had mentioned earlier, the Cancer Trials Support Unit, is a resource that we are bringing to bear to allow broad participation in histology-specific studies.

So, one of the important aspects, I think, of the CTSU, which started out being just a mechanism -- or not just -- but a mechanism for phase III clinical trial participation across multiple Groups, has made a commitment to do that in the context of histology-specific phase II trials.

DR. BAKER: That is a great experiment to do. You can test this hypothesis exquisitely easily. That is, have a trial of angiosarcoma with Taxol and have a trial of any of the number of experimental drugs against synovial sarcoma and look at recruitment to CTSU. What you will find is, if you have a new drug that is not available, CTSU is going to be a phenomenal mechanism. To get Taxol, my prediction would be you will see nothing.

DR. CHRIS FLETCHER: I just had a sort of comment in response to what Mike said which is, since our major goal is to identify new targeted therapies, and that was the purpose of the whole meeting, then I would like to reiterate a slide I put up yesterday which is, given we are dealing with a rare disease, we have to cling to this paradigm that, among solid tumors sarcomas have brought far more mechanistic information than many of the other tumor types in terms of signaling pathways, in terms of having discrete genetic lesions.

That is the thing we have to capitalize on when we try to get industry to be engaged, I think.

The other thing is that if we want to keep industry engaged, we can't expect them to invest large sums of money unless we make absolutely sure that we validate our targets.

Without wishing to harp on it, we really have to have the data to support what we claim is a target in the tissue.

Yes, Larry's pitch is what got me going; but that is the biggest problem right now -- is that that is why we are going to have wild, wacky, negative trials, because it is not properly validated data in the first place. We have to be really careful. Otherwise, industry won't invest a penny, I shouldn't think.

DR. BORDEN: Jonathan has laid out very clearly things that would make us say that something would be an excellent target in his presentation this morning, and George also did it yesterday.

DR. WHARAM: Sharon gave us a kind of hierarchy of targets going from expressed to expressed plus, active, et cetera. It is worth remembering that there is a therapy that hasn't been mentioned here, which is the use of radioisotope labeling.

In fact, we don't necessarily require the elegance of the greatest or the highest hierarchy. If a target is expressed well enough and even not in every cell, that with isotope therapy, the targeting may be sufficient to induce response. A model for the concept, if one doubts the efficacy of the therapy, is the fact that we can fairly routinely cure metastatic adenocarcinoma if it comes from the thyroid gland and if the mets are small. So again, there is also a problem with industry not having the where withal to radiolabel the drugs that might serve to target these lesions. But it is still important to remember what potential that therapy has.

DR. LADANYI: Larry, I just wanted to raise again something I raised in the breakout session, and it actually ties in with what Chris said. Again, there is the fact that the empiric approach is really not a high-throughput approach to evaluate new agents. I think we are in a special situation now where we have just a handful of agents and we can afford to try them out with relatively little pre-clinical data, relatively little target validation on different relapsed or recurrent patients. I think that the emphasis for the long term is to figure out the most efficient way to identify targets when you have got 50 agents that are available, and that you cannot possibly test empirically.
I think we need to figure out the infrastructure and the systems, the level of evidence that we need to take these agents into clinical trials.

DR. BAKER: I couldn't agree with you more, but I would want to re-emphasize this point. The reason we showed these things was that a year ago, there would have been many people who said that drug couldn't work in any condition other than c-kit positive GIST. We have now learned that there are some other conditions where -- fortunately, for those patients involved, that is not true.

The issue, then, is to see what you can learn in a well-studied patient, and we can argue about how well-studied the patient might be. In a well-studied patient, is that going to give you insight to a new target?

DR. BORDEN: Any other comments? Again, this has been excellent discussion. Okay, thank you very much, Jaap and Larry.

TOP