SLIDES & TRANSCRIPTS
Tuesday, March 6

Present Status: Non-Surgical Approaches to Therapy
John Macdonald, MD

Clearly, there are going to be chemotherapy regimens and systemic therapy regimens which, in the future, will probably be a benefit, and certainly there is no reason not to continue testing adjuvant chemotherapy. The other thing that is important in designing adjuvant studies for gastric cancer is, of course, to understand something about sites of recurrence. These are data from Len Gunderson, who is in the audience, from the old Minnesota re-operation series. The only point to be made here is that local failure is an important point. Obviously, we are going to have a session on local control. Local failure is an important aspect of what happens in gastric cancer.

TOP

Therefore, local means of therapy, perhaps re-operation, re-resection, or external beam radiation or interoperative radiation are certainly reasonable strategies to test. If one looks at the use of the 5FU, for example, as a radiation sensitizer, plus irradiation, one can see, in patients who have known small-volume residual disease, both from the GITSIG and the old Mayo Clinic studies, that one can actually cure 12 to 20 percent, roughly 20 percent, of patients who have known residual microscopic disease, positive margins, et cetera, small volume residual disease.

TOP

Certainly knowing that there is a high likelihood of relapse locally, knowing that patients with known residual disease can be cured, a small percentage of them, with combined radiation chemotherapy, it was reasonable to test a combined modality arm of therapy. This is the INT-0116, the intergroup study, that was reported at ASCO last year. Again, there are lots of issues here that we can talk about in discussion in the next two days. Certainly the surgical issue is an important one. All that was required here was that patients be deemed free of disease. They had to have negative margins and the surgeon has to say there is no residual disease left. We didn't demand any particular kind of resection, D-1, D-2, et cetera. As you can see this is a 5FU leucovorin radiation therapy arm.

TOP

This is obviously 1990s technology. This is what was appropriate. The arm had been piloted in upper GI cancers at the Mayo Clinic and shown to be tolerable. Six hundred three patients were registered. There were about eight percent ineligibilities. These were patients who, at pathology, were found to have positive margins, et cetera. A reasonable study, a reasonable number, reasonable power, as we can discuss later.

TOP

If one looks at the surgical procedures, and following up on Dr. Karpeh's excellent presentation, you can see here that American surgeons -- one can almost think of this study as a patterns of care study of surgery in the United States during the last 15 years -- American surgeons do less than a D-1 dissection about 50 percent of the time. Formal D-2s, documented dissections, were done about 10 percent of the time.

TOP

The other thing that was important here was the radiation planning. Steve Smolley, who was to be here, and unfortunately couldn't make it, was responsible for reviewing all the radiation ports in the roughly 300 patients who were randomized to treatment. Of interest in patients receiving radiation therapy of the upper abdomen for gastric cancer, 34 percent of the treatment plans that were prospectively reviewed had to be changed. Two-thirds of those would have missed significant node-bearing areas and would have effectively missed significant tumor. One-third of them would have resulted in very significant major organ toxicity. So, radiation oncologists in the United States are going through a learning experience of how best to give this upper abdominal radiation.

TOP

What about the results? Obviously I am skipping a lot of stuff. Just to give you the abbreviated results, this is the relapse-free survival. You can see with a two-tailed log-ranked T test, a highly significant result with median disease-free survival of 30 months in the treatment patients versus 19 months in the observation arm.

TOP

When one looks at the overall survival, again, highly significant, with their 35-month median survival in the treated versus 26, 28 months in the control arm. Bob Mayer informed me last night that this paper has been accepted by the New England Journal. So, it should be published soon.

TOP

What about sites of recurrence? What we can say is that we did not find any statistically significant difference in sites of recurrence. One would hypothesize that perhaps this strategy would decrease local recurrence. As you see here, there is a suggestion that local recurrence was decreased. I think the thing to look at here are the denominators. The way we looked at recurrences was the percentage of patients who had recurred, not the percentage of recurrence of all the patients on the study.

When this study is fully mature -- it has now got about 1,500 days of median follow up, so over three-and-a-half years of median follow up -- when it is fully mature and all the patients who are going to relapse have relapsed, perhaps there might be some differences here. The other thing to point out is that the only thing that was required of investigators was to report a site of recurrence, and the first site of recurrence. They didn't have to report all sites of recurrence. So, the recurrence data is not a complete data set.

TOP

What do we know? Post-operative radiation therapy seems to improve survival, and disease-free survival. The toxicity is really quite acceptable. The main toxicity here was hematologic toxicity and some GI toxicity. That represented really two-thirds of the grade 3-4 toxicity. There were three treatment-related deaths on the treatment arm. One of them, a pulmonary fibrosis, was clearly associated with treatment. The other one was an MI, and the third one was neutropenic sepsis, again, clearly treatment related. Post-operative chemoradiation therapy can be considered, in the United States, a standard for a resected gastric cancer.

We can talk more during the next couple of days about other things in the study. We certainly looked at stage. We had small subsets and they were retrospective subsets, but we looked at stage, we looked at the kind of surgery that was done, and we found treatment benefits in all subgroups, although the subgroups are too small to apply statistics to.

TOP

What about chemotherapy in the year 2000, the year 2001? We have a variety of different combinations that didn't exist in 1990. Certainly CPT-11 and cisplatinum, gemcitobine regimens. Taxanes, of course, can be used in the treatment of advanced gastric cancer.

TOP

I just will show you, as an example, the ECS regimen, David Cunningham's regimen, which was published about three years ago in JCO, using epirubison cisplatinum in 5FU.

TOP

This study is representative of a very well-done, well-powered, advanced disease phase III study in patients with locally advanced gastro-esophageal cancer, and the kind of response rates seen with modern chemotherapy is fairly typical. CPT-11 cisplatinum will, for example, give you this kind of 50 percent response rate. This was a randomized study versus FAM-TX.

TOP

When one looks at the survival curves, one sees that there was a difference in survival, and this actually happened to be significant. However, when one looks out here at the tail at patients with unresected, locally advanced cardia-esophageal junction cancer, one sees that ECF is not resulting in significant disease-free survival. Well, where might we be going? We would like to make our chemotherapy and therapeutic approaches to patients with advanced and localized gastric cancer more specific.

TOP

One way to do that is to look a little bit at targeting. 5FU, obviously, fluorinated primadines are the oldest drugs in GI cancer. We know something about the targeting, we know something about the activation of these drugs, and we know something about the catabolism. Thimadalite synthase does appear to be the target. Of course, DPD, diprimadine dehydrogenase, is the catabolizing enzyme. One could hypothesize that if you could look at tumors with perhaps low levels of target and low levels of catabolizing enzymes, that is where you might get the highest concentration of fluorinated primadines, or its activated nucleoside enablized active forms, and get the best response.

TOP

This is just some work from the Dannenburgs and Heinz-Josef Lenz from USC, just an interesting retrospective study where intratumoral DPD and TS were looked at in tumor specimens using a PCR technique. What was found was that if you had high target or high catabolizing enzyme, the likelihood of response was quite low. Interestingly enough, a small group of patients here with low TS and low DPD had, in retrospective analysis, of course, had a high order of response, suggesting again that this sort of approach may be of value. Of course, what was done here was to look at a tumor effect. Now, these enzymes in tumors, obviously sometimes it is hard to get tumor tissue. Tumor heterogeneity is a fact of life. It may be that when you look at the primary tumor you are not going to be able to predict what is going on in the metastatic lesion.

TOP

This is even more interesting, I think really points toward what we might be able to do in the future. Again, this happens to be work from Anascro Absac from Plumkot(?) from last year. The group at USC is doing the same sort of thing.
This is actually looking at TS polymorphisms in normal tissue, peripheral blood monocytes, to determine whether one can predict the kinds of TS levels that will develop in a patient's tumor, and whether or not that will make a difference in response. Here, you don't have to obviously remember the various polymorphisms.

The point is that in the roughly 10 percent of patients who had a polymorphism of TS which resulted in decreased TS activity, those tumors responded to a high order to fluorinated primadines. The point here is not that fluorinated primadines are the be all and the end all, but if we begin to look at targeting, we are going to be able to select patients who may respond. If we can begin to look, for example, at normal tissue polymorphisms, or normal tissue markers, and predict what is going to happen in a tumor with some regularity, we avoid the whole issue of tumor heterogeneity and getting tumor tissue.

TOP

This is the last slide. I just point that up -- put this up really as something to think about for the next couple of days as to where we are. I think there are some really interesting analogies at the beginning of this century. If one looks at infectious disease at the beginning of the last century, in 1900, you found the agents were things like cyanides, arsinicals, heavy metals. What did they do? They killed bacteria, but they also had significant normal tissue toxicity. Then along came penicillin.

What is the paradigm with penicillin that is so important? Well, penicillin has a selective target. Not only does it have a selective target, the cell wall, but it has a selective target that doesn't exist in ukaryotic cells. So, it is really a classic example of targeting. The whole development of antibiotics in infectious disease has been just that. That is why they are relatively non-toxic and relatively effective. What do we have in oncology now? We have radiation therapy, non-specific. It can be effective in some diseases. These include Alkylators, anti-metabolites, various drugs that affect normal tissue along with tumor tissue.

What do we have on the horizon? What do we have on the cusp of the future? We have an understanding of the molecular genetic differences between normal cells and tumor cells, and we certainly will be talking about things such as tumor suppressor gene products used as drugs, receptor targeting. A number of people this morning have referred to the epidermal growth factor receptor, and antiangiogenesis strategies, anti-oncogene strategies, and a variety of other things that we can do. I think the future is going to be very interesting. It is going to be more specific and hopefully less toxic in regard to non-surgical therapies. Thank you

TOP