Are There Unanswered Questions Remaining in Surgical Management? by Margaret Kemeny, MD

SLIDES & TRANSCRIPTS
Tuesday, February 15, 2000

Are There Unanswered Questions Remaining in Surgical Management?
Margaret Kemeny, MD

Slide 1:

DR. PETRELLI: At this time I would like to introduce Peg Kemeny who will update on the status of sentinel node, and then we will open it up for discussion.

DR. KEMENY: You may have noticed that that the name on the slide is not my name. That is because I haven't done much work on sentinel lymph nodes in colon cancer, and Dr. Saha has and really has done almost all the work, and I appreciate his sharing the information that I am going to present to you.

Slide 2:

There isn't all that much data available on sentinel lymph nodes and colon cancer as opposed to in melanoma and breast, and these are two studies that Dr. Saha has done, which I will share with you and tell you what we know about central lymph nodes right now.

The first study was a study that he did at Michigan State. They looked at 70 patients. This was the first study that he did and he reported on. There were 61 patients with colon cancer and nine patients had rectal cancer.

Slide 3:

Now, the method that they use is important. They use lymphazurin 1 percent and they put in either .5 or 1 milliliter, and it was injected subserosally around the lesion, not in the lesion itself but around the lesion and just in the subserosal area in four quadrants around the lesion.

Then they do the dissection during the operation as they would have before, and during the operation they look at the underside of the mesentery and they marked the first one to three lymph nodes that they saw that were stained blue. Then they continued on with the resection as planned and sent the specimen to pathology.

Slide 4:

At pathology obviously they received the fresh specimen. They saw the suture tagged lymph nodes. Those were dissected out by the pathologist, and then the lymph nodes were sectioned at 3-millimeter intervals. Then they were embedded for microscopic sectioning and they were immunostained for low-molecular weight cytokeratin.

Slide 5:

The results were that in these 69 patients they found one lymph sentinel lymph node in 35 patients, two sentinel lymph nodes in 29 patients and three sentinel lymph nodes in five patients. That is not all that interesting.

Slide 6:

More interesting is that they found 11 patients with positive sentinel lymph nodes, 42 patients with negative sentinel lymph nodes when the non-sentinel lymph nodes were negative. However, when the non-sentinel lymph nodes, and of course they looked at the non-sentinel lymph nodes just as they would have regularly, and when the non-sentinel lymph nodes were positive, they found two patients who had negative sentinel lymph nodes, and 14 patients with positive sentinel lymph nodes.

Slide 7:

So that means that 2 out of 69 patients basically they missed positive lymph nodes, and that is a 2.9 percent miss rate, obviously a small study but this is the first kind of example we are getting of what kind of a miss rate we may be getting.

The other interesting thing, and I think this is quite interesting was that of the patients who were T1 and T2, 26 percent of those patients had actual sentinel lymph node positive. Most papers and most studies it is hard to get a really good take on this, but usually it should be less than 10 percent of patients who have positive lymph nodes if the lesions are T1 and T2.

So probably we are picking up more positive lymph nodes than you would expect and,

Slide 8:

in fact, they looked at 11 patients who had positive sentinel lymph nodes and negative non-sentinel lymph nodes, and 6 of 11 patients had only micrometastasis in 1 of the 10 sections, and 4 out of 6 it was an H&E stain, and 2 out of 6, it was an immunohistochemistry stain which probably means, that you wouldn't have seen these on a regular section of lymph nodes because in general most institutions, and Carolyn can correct me if I am wrong just do one section through the lymph node.

So they probably would have missed this unless they were lucky. So they are figuring that eleven of 69 or 16 percent probably missed lymph node by regular pathology. I mean that is a hard number to come by, but that might be correct.

Slide 9:

Then they did a bigger study. Michigan State did it with John Wayne Institution and they looked at 140 consecutive patients, and they did sentinel lymph node sectioning at 10 to 20 micron intervals at 10 levels in the lymph nodes. They did H&E stains. They did immunohistochemistry this time for cytokeratin and CEA, and then they did a standard pathological exam as before of the rest of the non-sentinel lymph nodes.

Slide 10:

Slide 11:

Here again it was a mixture of colon and rectal cancers. In this study the sentinel node was successfully identified in 137 of 140 patients, so 98 percent ability to identify the sentinel node. That is quite high. Obviously these were all being done by doctors who really knew how to do sentinel lymph nodes, but they basically say that it is fairly easy to identify the sentinel lymph nodes.

Slide 12:

So in the 137 patients in whom the sentinel lymph nodes were noted, there were 250 sentinel lymph nodes, and there were 1,900 non-sentinel lymph nodes.

Slide 13:

Of the 137 patients, 87 had negative central lymph nodes, and 52 had positive sentinel lymph nodes.

Slide 14:

Of the 85 with negative sentinel lymph nodes, 78 or 91 percent, all of the non-sentinel lymph nodes were also negative. Seven or 8.2 percent some of the non-sentinel lymph nodes were positive.

So instead of the 2.9 percent, now it has gone up to 8.2 percent.

Slide 15:

Of the 52 patients with positive sentinel lymph nodes, 25 had some positive non-sentinel lymph node and 28 had negative non-sentinel lymph node. You can make what you want out of that.

Slide 16:

They then, in this study, took the first 25 patients and looked through all of the non-sentinel lymph nodes in those patients doing multilevel sections on the non-sentinel lymph nodes instead of just the one or two sections that they would have, and in doing that they found that in the non-sentinel lymph nodes that were negative before 2 out of 319 lymph nodes, that should be, had positive non-sentinel lymph nodes. So it means they are probably not missing too much by not microsectioning these lymph nodes.

Slide 17:

I just put this up here because this is one of the other papers that has been published. This is a study from the Netherlands and this study was somewhat different than the two studies I have just shown you. It was a smaller study. It is 50 patients, but in this study only 70 percent of patients did they find the sentinel lymph node. So 35 out of 50 patients they could not find the sentinel lymph node.

What is the difference? Why are we finding it in 99 percent, and they are finding it only in 70 percent? That is a little bit disturbing. One of the big problems Dr. Saha thinks is that this study used India ink and lymphazurin, and he feels that that makes a big difference, and that certainly could. We use lymphazurin really for all the sentinel lymph node procedures including the breast and the melanoma.

Also in this study it was injected into the tumor, not around the tumor, in the serosa around the tumor, and that may explain why there is this difference, and I guess this is something we won't know for sure until a study is done of this.

Slide 18:

So what are the conclusions that we can draw from this? It is an easy and safe technique. It is definitely safe. These patients had no problems. You are not even entering the bowel cavity. So there shouldn't be any reason why these patients should have any problems from injecting, you know, less than a cc of lymphazurin into the area around their tumor, and you are doing the dissection as usual. So we are really not doing anything different in these patients at this point.

So there is not a safety issue for this. Certainly there is no extra expense since lymphazurin is cheap and that is really the only thing extra that is being done at this moment aside from the pathology expenses, obviously which is something else.

So the question is are we getting more accurate staging from this. Obviously we are getting somewhat more accurate staging because we are looking more closely at things, but is it important? That we are not sure about, and does this mean in the future that we could section less lymph nodes, make the pathologist's job a little bit easier? I don't think we know that yet, and how important is immunohistochemistry staining to add new information in the future, and I certainly don't know what the conclusions are for that.

Slide 19:

Nick already said this, but I just think it is terribly important to remember this is completely, and I will go as far as to say completely different than sentinel lymph nodes in breast and melanoma because the real impetus for both of those was morbidity from the lymph node dissections, for breast obviously for axillary dissection for which you can get lymphedema of the arm and even more important perhaps is the inguinal dissection from which you can get quite bad lymphedema with melanoma which is a quality of life problem for these patients.

So there was a big impetus to do this in these two diseases, and that was the main impetus for this. That they may be getting additional pathologic information, of course, we are not really sure of, and that is still something that has to be proven even in these two diseases.

So in the colon, there is no impetus as far as morbidity is concerned because it really doesn't matter, as Nick was saying, whether or not you take out all the lymph nodes or not as far as in the wedge that you are taking out.

It doesn't add to the morbidity, and it really doesn't add to the operation length much, and so the only reason to do sentinel lymph nodes would be that the sentinel lymph nodes will give us increased information.

Slide 20:

So I think we are left with these questions. Is there a prognostic significance to micrometastases in sentinel lymph nodes, and these are the questions that have to be answered in the future, and is there a prognostic significance to immunohistochemistry positive sentinel lymph nodes that are H&E negative?

Two things that we are going to be finding if we study sentinel lymph nodes and certainly the first one I think will be very interesting as far as staging is concerned because we may find, and this may be as simple as a set of whether they are p53 positive or DCC, may be as simple as they have positive lymph nodes. That is why the Stage II are doing worse, that this group have positive lymph nodes and they were just missed, and the immunohistochemistry I think we really have to find out about.

Slide 21:

So obviously we need to go on to a large trial, not a large randomized trial, but we need to add sentinel lymph nodes probably to the trials that we are doing. I think that would probably be the easiest thing. It would have to be blinded obviously because you wouldn't be able to change what is happening to the patient because of the trial, and I think it could be very easily done without upsetting any of the trials since most, as I said, in most pathology they would only need one slice of the lymph node and the rest of the lymph node could be sent to a central area to look at these lymph nodes for positivity, but we certainly need a large trial to validate this new technique, and that is pretty much where we are right now.

Thank you.

DR. PETRELLI: Thank you, Peg.

(Applause.)