DR. SONDAK: Okay, thank you. Now we get to the fun part. Our first pair of commentators, and we gave the honor of the first commentator, the optimist viewpoint, to our eternal optimist, Scott Saxman.

I just want to take this public opportunity to thank Scott for all that he has done on behalf of melanoma since he has taken over at CTEP, both for the cooperative groups and in setting up this meeting, and in a number of other things. Scott, it has really been a pleasure working with you and, thank you. We are looking forward to what you have to say.

DR. SAXMAN: Thanks. I hope people can appreciate how intimidating it is for a clinician to stand up in the middle of the biologists, after the biology talks, and talk about targeted therapy.

For the first pro session, Vern requested that I focus not on one genetic abnormality, or even on the genetic abnormalities that have been discussed this morning, but sort of generally and broadly overview the way we at CTEP have been thinking about clinically targeted drug development, and what I think some of the main issues are that need to be considered as we collaborate on biologic and translational research in the context of both phase I and phase II trials, and considering large scale randomized trials.

I would like to consider targeted therapy in the context of definition A here, a mark to shoot at, with the main consideration being, I think, as we think about clinical investigations of targeted therapies, of characterization of the mark, and the mark needing to be in the tumors of the patients that we are treating.

I really think that is the main message I would like to get across here. I think if we do that, and we work with our biologic colleagues in that context, then we can think about targeted therapy as a goal to be achieve and hopefully, by the end of this, the con speaker won't turn me into definition C.

TOP

So, what makes a good target? This is from sort of a general clinician's point of view, and some people have touched on this, this morning, already.

A good target should be an essential aberrant cellular process. A very nice summary article from some Australian investigators sort of put the molecular information of the important oncogenic events into three categories.

Proliferation, the cancer cell needs to be able to proliferate and bypass the usual growth check mechanisms that occur in normal cells, permanence, bypassing the apoptotic mechanisms that occur in normal cells, and position. Cancer cells have the ability to move and go from one place to another.

Now, this may be the causative molecular abnormality or oncogenic event. However, I think we need to consider -- and a couple people touched on this, this morning, as well -- that it has to continue to be an important pathway for the malignant cell

I think the best example of this is probably chronic granulocytic leukemia. The BCR ABL rearrangement is clearly critically important in the oncogenic event and, in chronic phase leukemia, a CGL, Gleevec targets that, and 90-plus percent of those patients respond to that therapy.

Once you enter an accelerated or blastic phase, other genetic events become equally or more important and those targeted therapies don't work nearly as well when that occurs.

So, it is not enough to say, well, we have identified the oncogenic event. It is also critical to say at what point in time that helps us in terms of directed therapy toward that event.

The expression of an expected target does not necessarily equate with pathogenesis. I am not the first person to make this observation here this morning, but I think it is a critically important observation.

It is not enough to say that VEGF is over-expressed or this is over-expressed. Again, we need to get back to whether this is an essential aberrant cellular process.

So, we shouldn't throw drugs at diseases, particularly melanoma, just because something happens to be there.

TOP

There should be minimal bypass or redundant mechanisms or pathways that the cell can easily bypass the therapy.

The crucial component of the pathway, I think, for drug development should be quantitatively measurable, both at baseline and when perturbed.

This is going to be a critical feature, I think, as we develop drugs in melanoma.

TOP

So, in thinking about this clinically, there are actually clinical considerations of target identification and selection that I think we can learn from and that are important.

If we think about this, there have actually been several major developments or major advances in treatment of certain diseases over the past five or six years that are targeted therapies, imatinib in GIST and CML, trastuzumab in breast cancer, rituximab in non-hodgkin's lymphoma and ATRA in promyelocytic leukemia.

Thinking about how these drugs were developed in these trials clinically, the main feature here is that these drugs were developed in these diseases for the patients who had those targets.

For example, if we had looked, as we usually do in sarcoma -- and I am very familiar with this because we were part of those trials -- just giving imatinib generally to patients with sarcoma, what would have happened? We would have missed it.

We would have missed it completely because GIST is a very rare subtype of patients with sarcomas. There are usually one or two of those patients in those trials, and half these patients respond. So, it is very likely that that would have been missed.

The same is true with trastuzumab in breast cancer. Those trials were done in patients who were HER2 positive.

If they had been done in general groups of patients with breast cancer, the effect of herceptin would have been missed. We have run those numbers, actually, and those trials would have been negative rather than positive.

So, the critical feature here is that we need to be using targeted therapies in patients who have the target.

I think a very good example of an unsuccessful strategy has been Iressa, or ZD1839, in non-small cell lung cancer.

For those of you who may not be familiar, there have been two large phase II trials with Iressa in non-small cell lung cancer, with fairly respectable response rates in the 15 to 18 percent range.

In the large randomized trials, they chose unselected patient populations, and those trials were stone cold negative.

There have been lots of reasons hypothesized as to why that may be true, but in my own opinion, that is true because they chose unselected patients.

Some patients had the molecular abnormality that the drug targeted, but many did not. So, the effect was washed out in those who did by those who did not, and I think this is a critically important feature as we develop those trials.

TOP

This is a statistical representation of this. I am not going to go into great detail, but just to show you the numbers -- and Boris Friedlin at CTEP and Janet Dancey generated these numbers.

With the molecular heterogeneity of these diseases, now, that we have to consider, we need to think about the number of patients or the percent of patients that actually have the target.

If that number is small, even with very high hazard ratios, even if we double the survival, but the target is only found in a small population of patients, it requires an enormous number of patients in that clinical trial to see the effect.

If we don't either select patients appropriately, or do enormously large trials, we are going to miss those effects.

TOP

So, while the paradigm itself may not be different, that the substrate that a drug interacts with in a tumor cell must be present for the drug to have activity, the need to characterize the target, and the dynamics of the interaction between the target and the agent being tested are much more critical than in the past.

So, in ancient times, when chemotherapeutics targeted DNAs or ubiquitous proteins, there was less need for characterization and individualization of therapy, because all cancer cells have DNA.

That is no longer true. Now we are dealing with more precise weapons. Thus, the presence of genetic and phenotypic heterogeneity within the cancer cells and within a histologic type become much more important, and I think a very critical consideration.
I would hope that this would seem obvious. However, I will say, at a recent meeting, I made that argument and this is what I was presented with.

We don't require markers of benefit for standard cytotoxic agents. So, CTEP is being illogical to demand this for new agents. Why is there suddenly a different standard.

I think there is a different standard, and I hope that I have been able to explain that a little bit.

TOP

I think melanoma is a good model for targeted therapy. I think we have reason to be very optimistic that we can do this in melanoma.

Melanoma has the availability of precursors and early lesions that are identifiable and available. We can get to them.

That is not true in most diseases and most malignancies.

Many patients have easily accessible tumor tissue for molecular studies. This is a double-edged sword, I understand, but we don't have effective therapies for patients with melanomas.

So, the confounding effects of co-administration of chemotherapy with the targeted agents, and what effect does the chemotherapy have on the target and those sorts of things we don't have to worry about, quite often, in melanoma.

We can treat chemo-naive patients with single agent therapies or minimally pre-treated patients. It gives us a much better opportunity, I think, to look at the perturbations that occur with these targeted therapies.

TOP

Do we have potential targets in melanoma? Well, you have heard a few of them this morning, and I am not going to go through this list.

This is from a paper that was published a year and a half ago or so, by Dr. Brown and Kirkwood. We certainly know that there are a lot of genes with altered expression in melanoma.

TOP

So, hopefully, somewhere on this list, we are going to find these genes that actually do drive the melanoma cells that are critically important.

The onus is on our biological colleagues here to tell us which of these we should be targeting in our clinical investigations.

The reverse of that is, the clinical colleagues need to conduct the early trials such that we can determine if these genes are perturbed with the targeted therapies and, if so, how, and correlate that with clinical outcomes.

TOP

CTEP, just to let you know, is conducting several trials with targeted agents in melanoma. STI571, or Gleevec, you are familiar with. Bevacizumab is an anti-VEGF antibody.

R11577 is a farnysyl transferase inhibitor. Rapamycin analog, and so on -- I am not going to go through each of these, but I can tell you that, in almost all of these studies we are also conducting translational research activities, to look and see what the interaction is, and whether it is predictive of the clinical outcome.

TOP

So, here is the way I think that we need to think about this. We need to have rationally designed, target-based biologic and translational research that is conducted by our biology and translational research colleagues, that leads to empiric but focused studies with mechanistic investigation of clinical end points.

Investigation of those clinical end points needs to feed back to our translational research investigators, so that they can continue to look at these target-based biologies.

TOP

So, it is really a circle here, and I would even overlap these now, that our futures really rely on each other.

Part of the purpose for this meeting is to get people from both camps into the same room, and to think about how we should develop these things, moving from the test tubes or the petri dishes and the mice into the patients, and back to the mice.

TOP

So, taking the pro position, I think we have many opportunities in melanoma. I think we need to continue to focus on molecularly targeted therapies and molecular targets within the melanoma cells that we can use in our armamentarium.

I think that has been neglected in melanoma to some degree because of some of the other immunologic activities that have been going on, also important, but I think we need to pay attention to this as well.

I think we need to stop arguing for our limitations, because then they become us, and look at the opportunities that we do have. So, that is the optimistic position.

TOP