Thank you, Vern. I also want to thank the organizers of the meeting. It is quite a group of melanoma experts to assemble. I know it takes a lot of work.

I want to talk on the state of RT-PCR and, specifically, about how it applies to probably the new standard of care for at least surgically treating melanoma, and that is the sentinel node biopsy.

TOP

We know that nodal status of the melanoma patient, like a lot of solid tumors, is the most powerful predictor of recurrence and survival.

It was kind of natural to try to develop better ways to at least nodally stage patients with melanoma. The first advance was this technique of pre-operative lymphoscintigraphy, which is simply a radial colloid injection around the melanoma, here in the right arm.

If that colloid has the right particle size, it is taken up by the lymphatics. Then you can scan the patient and actually get a map of where the cutaneous lymphatic flow is from any primary sites on the body.

So, this right upper arm melanoma not only goes, as you would predict, to the right axilla, but it also drains to a node in the posterior triangle.
It has been shown that, since both nodal and basin are equally at risk for disease, they have to be dissected to accurately stage the patients.

TOP

Then, the surgeons take the information from pre-operative lymphoscintigraphy to the operating room and actually use a combination mapping technique -- usually, the first mapping technique is a vital blue dye procedure.

Actually, it was initially described by Dr. Morton and his group. It has the right particle size, goes to the regional base, and then actually will color the first node blue.

TOP

Then we use a second mapping technique, radiocolloid, and these handheld gamma probes to actually follow the migration of the radial colloid to the central node.

Obviously, that makes it hot compared to surrounding tissue.

So, we can pretty much know where these sentinel nodes are located through the skin without making any sort of skin incision. This is just a picture of the gamma probe that we use interoperatively.

TOP

So, the sentinel node is any sort of node in the basin that is directly connected to this primary tumor.

The hypothesis put forth by Dr. Morton and proved now by probably more than 15,000 melanoma patients in the literature is, if these sentinel nodes are negative for metastatic disease, then the rest of the nodes in the basin should be negative.

All of a sudden, you can get full nodal staging with a simple lymph node biopsy procedure.

Then, it becomes obvious that you can better apply more sensitive assays for metastatic disease to one or two nodes in the regional basin rather than the 15 or 20 nodes with our typical complete node dissections or elective lymph node dissections.

TOP

So, it became obvious that certainly that would be possible, to apply more sensitive assays.

The assays may involve more sections, just serial sectioning the sentinel node, or histochemical staining of the sentinel node, or actually molecular biology for metastatic disease, the disease based on PCR.

There is evidence to suggest that, just as the blue dye and the radial colloid will come into this first node, so do the metastatic cells come into this first node.

Actually, the sentinel node, at least in melanoma, is relatively efficient as far as confining the metastatic disease in the regional basin.

If there is any sort of metastatic disease in the regional basin, it is confined to the central node anywhere from about 85 to 94 percent of the time.

TOP

This is just the experience that we published from Moffit Cancer Center when I was there, over an eight-year period of time.

The standard of care would be to take all the patients back to the operating room who have a positive sentinel node for a complete node dissection.

When we did that, only 9.7 percent of patients had disease beyond the sentinel node. So, 90 percent of the disease, the disease, when it exists in a regional basin, is confined to the sentinel node.

A large percentage of those patients are actually one node positive, with just microscopic disease.

The other thing you have to point out is that 90 percent of patients cannot possibly benefit from this more radical complete lymph node dissection.

TOP

Again, the lymphatic mapping allows now for a more detailed examination of the sentinel node. You can take out these one centimeter lymph nodes and actually bivalve them, and look inside them.

Sometimes you can see one to two millimeter deposits of metastatic melanoma. So, certainly this is grossly visible disease.

Hopefully, this is not the type of disease that is really not missed by our routine histologic examinations, even though this is still pretty low volume disease in the sentinel node.

TOP

What we find, when we do a lot of these procedures, is that metastatic disease in the sentinel node can be very low volume disease.

This is a sentinel node that may has 15 to 20 metastatic melanoma cells here, kind of outlined with the S-100 stain here, which is an immunostain for metastatic disease.

If you look at the standard histologic examination -- and that was removing 15 to 20 lymph nodes from a regional basin and perhaps making one or two slides out of each one of those nodes -- by doing so, the routine examination, which really was the standard throughout the world, examined one percent of the submitted material.

The pathologists would make their diagnosis, and we would make treatment decisions based on a pretty superficial examination of the regional basin.

Certainly, if you are doing that superficial examination, you are going to miss this low volume disease that is present in some patients.

Now, the challenge was to really determine whether this low volume disease was clinically relevant disease.

There were a number of perhaps immunologists and clinicians who suggested that perhaps this low volume disease would never become clinically relevant, that the immune system would be capable of getting rid of some of this low volume disease.

I think that is the challenge in a lot of these solid tumors, and that is, determine the relevance of the upstaging that we can do with the lymphatic mapping technique, and now, a more detailed examination of the central node.

TOP

We looked at better ways to do assays for metastatic disease in the regional basin.

We initially started, when I went to Moffit, with a lymph node culture technique.

We would just bivalve the lymph node, and we would put half the node in tissue culture and submit half the node for routine histology.

Lo and behold, in about 20 percent of the nodes that were called histologically negative by the pathologist, we were able to grow melanoma cells out of.

Then we showed that those patients in which we were able to do that had a decreased survival and an increased recurrence rate, to suggest that there was missed micrometastatic disease, and that this missed micrometastatic disease was clinically relevant disease.

Lymph node culture took about four weeks. So, it was a little clumsy to do, and now the PCR assay has now pretty much supplanted that sort of work.

TOP

You can actually -- the PCR assay and all these molecular techniques are better than doing serial sectioning of your lymph nodes.

Really, the rate-limiting factor still here is the number of sections you make and examine, and you can pretty much look at the whole lymph node for any evidence of metastatic disease with the PCR assay.

TOP

So, the PCR assay that we have described and other groups have described is based on the tyrosinase marker. All cells of the body will have the gene for tyrosinase, but only cells that produce pigment will express the messenger RNA for this particular gene.

Initially, we used a double round PCR assay for this particular marker. The feeling was that, if you found evidence of the gene product for tyrosinase in the lymph node, that would be pretty good evidence that there was metastatic disease there.

TOP

I don't think I have to really spend a lot of time on how the PCR assay is done. This is the whole gene sequence for tyrosinase

Obviously, we isolate the entire messenger RNA from the lymph node. Then we make a cDNA copy and we pretty much isolate just a sequence that we want to look at for tyrosinase, and then make multiple copies of that particular gene, to see if there is any evidence of that signal in the lymph node. So, there you have the PCR assay.

TOP

New assays are good but, unless you really show clinical correlation, you really can't use it to make decisions.

We tried to do it in our first 326 patients, in which they underwent -- these were patients with all tumor thicknesses of melanoma. Actually, they were greater than .76 millimeters in thickness.

They had the mapping procedure. The sentinel node was harvested. We looked at the sentinel node with our PCR assay.

What we showed was that, in those patients whose sentinel node was histologically negative, S-100 negative with the immunostain, as well as PCR negative, did fairly well.

This is looking at disease free survival. The five year survival on this group of patients was about 96 percent.

That is for all comers of melanoma. So, no matter what your tumor thickness was, no matter if your tumor was ulcerated or not, if there wasn't any sort of evidence of metastatic disease in your sentinel node with very sensitive assays for metastasis, you did fairly well.

You came a lot closer to identifying patients who really were surgically cured of your disease than we could with the old staging systems that we used to use.

The other interesting group of patients here in yellow are those that you upstage with the PCR assay. So, these patients here in yellow, their sentinel node was histologically negative, S-100 negative, but PCR positive.

You can see that those patients are doing significantly worse, again suggesting that missed microstatic disease occurs with routine histology, but also that it seems to be a clinically relevant disease. In patients in red here, their sentinel node was histologically negative.

It does seem that missed microstatic disease is relevant, but also what it shows is that the volume of disease in the sentinel node seems to be a prognostic factor.

If you have higher volumes of disease that can be detected by just routine methods, you do worse.

TOP

Now this is being tested in the Sun Belt Melanoma Trial that Kelly McMasters from Louisville is heading.

Basically, they are starting with melanomas one millimeter or greater in thickness. The lymphatic mapping and sentinel node biopsy procedure is done.
The sentinel node is examined with routine histology with S-100 stains. If that is negative, the PCR assay is performed. If that is negative, patients are observed.

If the sentinel node is histologically negative, but PCR positive, they are randomized to either observation, complete node dissection, or complete node dissection and interferon to now, in a national trial, determine the relevance of the upstaging that you can do with these molecular assays. Also, if it is clinically relevant disease, what is the best way to treat those patients?

TOP

On the Sun Belt Melanoma Trial, it is a little bit more conservative. In order to call someone PCR positive, you have to be positive with a tyrosinase marker, as well as positive with one of the other three markers for metastatic melanoma.

TOP

This is just preliminary data from the Sun Belt Melanoma Trial, showing pretty much what we showed at Moffit Cancer Center.

That is, if your sentinel node is histologically negative, S-100 negative, PCR negative, you probably are close to being cured of your disease.

If you are upstaged with the PCR assay, you do significantly worse, again suggesting that missed microstatic disease occurs, and these patients do significantly worse.

Now, not all these patients are going to recur and die of their disease, but not all the patients who are just identified as being node positive with routine methods recur and die of their disease.

You know, surgery, in and of itself, probably cures about 60 percent of patients who have microscopic one node positive disease.

Surgery, in and of itself, actually cures even people with grossly positive disease, to the extent of about 15 to 20 percent.

So, not all these people are going to recur, but again, they didn't recur and die all the time down here.

TOP

Now, this PCR assay has been supported by a number of other laboratories in the country, and actually the world.

Jim Goydos is here, and he has certainly published data showing that you can upstage patients who are called histologically negative with a PCR assay for tyrosinase, and those patients do worse than if you are negative with the PCR assay.

Again, we have a number of John Wayne people here who have shown the same findings.

TOP

There are other lines of evidence to suggest that this low volume disease in the sentinel node that is only identifiable with more sensitive assays of the sentinel node, like immunostains, you can inject this low volume disease in animals and grow tumors nicely in animals to suggest that these are viable cells and, at least in animal systems, they are tumorigenic.

TOP

So, how do we find metastatic disease in the nodal basin? We know that routine histology, if you just consider all comers of melanoma, will identify 15 to 20 percent of patients with stage III or metastatic disease to the regional basin.

If you routinely add immunohistochemistry -- and this is only possible with the lymphatic mapping technique -- you upstage about eight to 10 percent more patients that would have been called histologically negative by the pathologist.

With the molecular biology assays, you upstage an additional 10 to 15 percent of patients.

TOP

This is interesting data actually from the American Joint Committee on Cancer, that really looks at the way we stage the patients to be node negative through the years.

The data base involved over 17,000 melanoma patients with a mean follow up of about five years. In the 1970s and 1980s, we just did wide excisions of melanoma and staged patients to be clinically negative with just a clinical examination.

We have looked at that database. If you stage all comers with melanoma with thicknesses greater of a millimeter to be node negative, they had this survival.
Then, in the 1980s and beginning of the 1990s, a lot of groups actually were doing elective lymph node dissections to stage patients.

In the yellow here are those patients staged to be node negative with elective lymph node dissection. You can see that their survival is significantly increased, to suggest that, with clinical examination, you probably miss a lot of people that really have stage III disease, and that would decrease their survival.

At least with elective lymph node dissection, where you take out 15 or 20 nodes and do one section of each node, you do a better job as far as staging the nodal basin.
Look what happens if you stage melanoma patients to be node negative with the lymphatic mapping technique. They have this survival, to suggest that, even with elective node dissection and a superficial examination of the regional basin, you probably are missing a lot of people with stage III disease.

The sentinel node procedure will be the best way to not only stage patients to be node positive but it also will be the best way to stage patients to be node negative.
Just think, perhaps we can get a similar increase in staging when we add the PCR assay now to the lymphatic mapping technique.

This data is generated just from doing routine histology and maybe some immunostaining. Perhaps the survival of patients that are staged to be node negative with lymphatic mapping and PCR will be up in here.

TOP

[No text is associated with this slide].

TOP

So, my time is up. I just want to show you how this PCR assay is being performed in at least a regional trial, looking at the sentinel node. The sentinel node is being examined with routine histology with immunochemistry as well as PCR.

If you are positive with any one of these markers -- this is a multiple marker quantitative method -- then you are eligible for this second part of the trial, where you randomize to just interferon versus complete lymph node dissection and interferon, to examine the role of complete lymph node dissection in patients with melanoma.

This is a little bit different than the elective lymph node dissection trials, because the only patients that are eligible for this part II are those that actually have solid evidence of metastatic disease in their regional basin.

TOP

That is about it. I show this slide to look at some future considerations. Certainly, we really haven't determined whether we need multiple markers or just the tyrosinase marker.

It would be nice to develop RT-PCR assays on fixed tissues. The question is, is this an independent prognostic factor with all the other prognostic factors? It seems to me, at least from our data.

You can develop the concept that ultrastaging, where actually we look at the sentinel node and the peripheral blood and bone marrow, to really try to identify patients that are probably closer to being surgically cured of their melanoma and don't need any sort of other therapeutic intervention. Thank you very much.

TOP