SLIDES & TRANSCRIPTS
Tuesday, June 18

Breakout Group B Summary and Recomendations with Group Discussion: Therapuetic Intervetion/Targeted Therapies for Systemic Disease

Laurence H. Backer, MD

We wanted to begin -- I think the whole purpose of this conference is largely as a result of GIST and imantinib. So, we wanted to begin with, what lessons have we learned from that, so that we could begin to identify new targets?

We wanted to understand how much evidence did there need to be considered before a target would be considered to be a target, and I think the breakdown you just saw from Jonathan is a superb way of thinking about this.
What, if any, targets that exist currently are worth exploring? And we will give you some specific recommendations and some basis for that. And then, what about the clinical trial design necessary to be specific for targeted therapies?

TOP

Much has been said today and yesterday about the use of cDNA microlibraries as they look at sarcomas. I think it is fair to say - and, if it is not, someone will replace me -- the paper with the highest score for the sarcoma oral session at this last ASCO-and I should point out this is quite a year for sarcoma people, we had a specific session at AACR, we had a specific session at ASCO, and now this. You know, life doesn't get better.

This is Bob Maki's work on behalf of Memorial Sloan-Kettering and the genomic center at Columbia. This happens to be a gene expression profile for PDGF-A and by various histological subtypes that were outlined.

This is his last slide, and he said, is this over-expression -- if you will, increased expression -- of the specific oncogene a target for imantinib? He left that question for us. You will notice that a couple of the patients that most expressed this were this subtype malignant fibroushistiocytoma.

TOP

There is a consortium of nine institutions that calls itself the North American Consortium of the Connective Tissue Oncology Society that has been engaged. Five of those member institutions have been engaged in doing a trial of Gleevec in 10 histological subtypes, as shown here. The idea of this was to try to accrue a sufficient number of patients in each subtype to be able to make some assessment as to the value. You can see angiosarcomas, Ewing's, fibros, et cetera.

I won't bore you with how these 10 types were chosen; but I will tell you that seven were chosen because they simplistically represented fusion proteins that were known. Some were chosen because they were common. Some were just chosen.
At any rate, in five months time, these five institutions have accrued 76 patients. I want to tell you about one of those patients in particular.

TOP

This is a woman who had malignant fibroushistiocytoma of her thigh. She was treated with a neoadjuvant protocol that included adriamycin/ifosfamide. She underwent wide local excision and radiation therapy. Unfortunately, despite that, within a year, she developed pulmonary metastases. The pulmonary metastases were treated then with high-dose ifosfamide and then with yet another protocol, none of which affected the tumor.

Here you have CAT scans of this rather large nodule that was biopsy-proven to be MFH, identical to the thigh lesion. You can see, in a couple of months, there is significant shrinkage.

TOP

That in itself is not important, but perhaps this explanation may be more important. So, in that particular patient, you see immunohistochemical staining for PDGF-A, which was positive, for PDGF-B, which was negative, for the receptor alpha, which was positive.

So, you have the ligand and the receptor being positive, and perhaps a downstream recognition of that phosphorylation of Akt. This may, indeed, be an explanation for this patient's response.

TOP

DR. CHRIS FLETCHER: Larry, I am going to be very, very rude. Go back one slide, because it is really important. This is when the audience gets to understand this thing.

I will be very blunt and I know it is offensive -- but that immuno is garbage. You have actually got normal osteoclastic giant cells there, stained with a marker that you think is relevant. They are not even tumor cells. Every cell sitting there is positive. That is the whole problem.

DR. BAKER: Actually, that is not true. Why don't you let me finish, and I am sorry I said what I did about MFH, but I am going to continue.

Here is another patient. This is a patient with a desmoid tumor. This patient also, for reasons that I won't belabor, was a patient who was given Gleevec. You can see, on a T1-weighted image, a significant reduction in the volume of the tumor, both here and almost disappearance of it, as it is anterior to the issue.

As a result of this, and as a result of something that was sort of overlooked, as Jonathan said, it has been reported that 90 percent of GIST patients are c-kit positive. You should remember that the Gleevec trials that we have been talking about, by definition, had to be c-kit positive. So, now there are a couple of cases that emerged that seem to be c-kit negative by immunohistochemical staining, and indeed, the question is, would they respond to Gleevec? So, this trial that I told you about, that had 10 histological subtypes, will be expanded to include desmoids and c-kit negative GIST.

I should also tell you that all the patients that are accrued to this trial, by eligibility criteria, had to have progressed to several other chemotherapies, as I have illustrated with that one respondent.

TOP

Finally, there was discussion, by way of background, about other potential targets. This is tissue microarray data that looks at three of the four of HER2/neu family of genes - EGFR, HER1, 2, 3 and 4 -- by histological subtypes.

What one sees is that, in some subtypes, like osteosarcoma, that no expression of HER2/neu could be seen in this laboratory, whereas in synovial sarcoma, there was increased expression of HER2 and HER1.

I should point out that Marc Landanyi in our group also actually could tie this in to SSX1 expression versus SSX2 expression. So, there seems to be greater increased expression of this gene family in those synovial sarcomas that had the SSX1 variant. In any event, these are the reasons -- these are some of the reasons -- we thought that this was a target ready to go forward.

TOP