Retroperitoneal Sarcomas

SLIDES & TRANSCRIPTS
Monday, June 17

Retroperitoneal Sarcomas

Burton L. Eisenberg

Slide 1:

Tom Edison was obviously referring to the technology of his day, but I think the same could be said of perhaps this entire conference and, particularly, retroperitoneal sarcoma. I also might add that this is the surgical approach for the disease.

Slide 2:

This is the problem. Clearly, the opportunity here to obtain negative margins in a tumor of this size is almost laughable.

The possibility, then, of discerning whether, indeed, you have negative margins is extremely difficult, because it would take a pathologist the better part of a week to evaluate all these margins. Obviously, this is the type of case you have to start at 8:00 o'clock in the morning.

There are a number of important structures here that are associated with this tumor. The vena cava happens to be in the way. The duodenum is in the way, the liver is in the way, and so forth. So, monumentally, this is a problem in local control and it is a surgical problem in local control as well.

Slide 3:

There are a number of aspects here that are important in understanding the overall problem in retroperitoneal sarcoma specifically. First of all, this says 15 percent of all soft tissue sarcomas. If you are dealing with a rare tumor it is a rare tumor, and that makes it rather difficult to develop paradigms for clinical trials that represent any sort of patient participation.

These tumors are generally asymptomatic. So, they are picked up at a fairly late date. You know, the retroperitoneum is a big place. That is why they get so large. Also, the average size of these tumors is quite impressive, somewhere around 15 to 17 centimeters at presentation. In any series, about two thirds of these are generally primary and about a third are recurrent, which represents a whole other problem.

Sixty percent -- fairly similar to extremity sarcomas -- about 60 percent are high grade and about 40 percent are low grade. The majority are actually two histotypes -- liposarcoma and leiomyosarcoma -- which makes it a little bit easier to concentrate a histotype-specific treatment.

Slide 4:

Surgical considerations are; Complete resection, the absence of gross residual disease occurs in most surgical and institutional series about 50 to 60 percent of the time. Well, this depends on a lot of other factors. Most series, as I said, are recurrent and primary. So, it depends upon the percentage of each in terms of resectability. It is also dependent upon the anatomic site. Lateral retroperitoneum are a little bit more easy to resect than central retroperitoneum. It also depends on the ability of the surgeon to obtain an en bloc type of resection. Some surgeons are better retroperitoneal sarcoma surgeons than others.

Contiguous structures are resected in approximately 80 percent of these patients -- again, relating to their size and position in the retroperitoneum.
Local failure is an extremely important aspect in the biology of this disease -- about 40 to 70 percent five-year local failure rates; this is a continuum. So, this actually increases the longer the follow-ups are. The 40-to-70 percent statistic is based usually on a five-year follow-up. You will see a higher percentage actually if you follow a series of longer duration. It also relates to a problem that surgeons face in this disease in terms of local control. So, we are not doing a great job as surgeons.

The majority of the patients who do continue to display risk of local recurrence after five years are those with low-grade sarcomas.

Slide 5:

Factors associated with tumor-related mortality -- unresectable primary tumor for sure, positive gross margins.

These two are probably the same thing. It is very important in this disease to obtain negative gross margins -- that is, to not leave any gross disease behind. That is true in both short-term and long-term results. In most series, you can see these changes at about three years. You can also see them at eight or nine years.

High-grade histology is also important in this disease for those patients who develop metastatic disease, very much like extremity sarcomas. I might add also to that list, recurrent retroperitoneal sarcomas are a risk factor for tumor-related mortality.

Slide 6:

The single most important variable in long-term tumor controls are complete resection with gross negative margins, and the importance of aggressive en bloc resection of primary disease cannot be overly emphasized.

This is a case of the first time is the best time, which also leads us to explore the possibility of having this rare tumor mostly treated at centers of excellence.

Slide 7:

I take this from Dr. Brennan's wonderful prospective database, again, emphasizing the local control issue here. The gross negative margins -- in patients who have gross negative margins, the median survival rate is 103 months in this series of about 500 patients over a period of 15 years. In incomplete resection, the median survival is only a year and a half.

Slide 8:

These are factors associated with tumor-related mortality after complete resection. So, these are patients that actually had the ideal, that is the tumor was completely removed. So, what happens to these patients and what causes tumor-related mortality in this group? Local relapse is the most common cause of cancer-related death.

Unlike extremity sarcomas, here the local relapse is an extremely important tumor-related mortality.

Approximately 75 percent of primary retroperitoneal sarcomas die in the absence of metastatic disease, another important factor in consideration for local therapy and for the combination of surgery and enhanced local treatment. Thirty percent of patients with high-grade histology will eventually develop metastatic disease.

Slide 9:

This is long-term recurrence and survival, "long term" meaning probably greater than five years. There are very few series that follow these patients long term with any numbers, but we can get a sense that the recurrence rate is approximately five percent per year from the time of the initial surgery. That is, this is a continuum, and it depends on the number of patients at risk.

Forty percent of the recurrences at 10 years occur in patients who were disease-free at five years. So, all of those patients who luckily make it five years out after surgery then some other form of treatment, will survive five years with no evidence of disease and will eventually develop recurrence between 5 and 10 years.

The five-year disease-free interval, then, is not necessarily a measurement of cure, which makes it somewhat difficult to develop trials in this disease, because of the necessity of this long-term essential morbidity. The patients who do develop long-term local recurrence are more likely to have the primary, low-grade, large liposarcoma.

Slide 10:

In order to extrapolate data, because there is a paucity of data in retroperitoneal sarcomas, we often extrapolate from the extremity experience, and many of the recommendations for retroperitoneal treatment are based on our understanding of the biology of extremity sarcoma. We can do that, for the most part, because there are very important similarities but there are also some differences.

The grade is fairly similar in terms of risk of metastatic disease. The size of these, as I said, retroperitoneal tends to be larger. Margin status in the extremity -- margin status is certainly an issue in terms of biologically aggressive disease. However, one can always amputate an extremity if there is local occurrence. The problem here is the margin status for retroperitoneal sarcomas is very problematic. In many cases, the margins are either close or microscopically positive.

The importance of local relapse -- again, in extremity, the local relapse may or may not mean death from sarcoma; but certainly in the retroperitoneum, it is an important contributor to tumor-related mortality.

Histology is fairly specific in the retroperitoneum for the large histotypes, and response to adjuvant radiation and chemotherapy -- just suffice to say that we do a lot of extrapolation from extremity studies in order to justify some of the adjuvant theories that we have for retroperitoneal sarcomas.

Slide 11:

So, in radiation therapy, can we actually do this? Can we extrapolate from the extremity experience? Is it fair to use our knowledge base of extremity sarcomas in response to adjuvant radiation and apply that to retroperitoneum?

We know, for instance, that in the extremity experience, there is some pretty good data to support adjuvant radiation therapy in terms of positive effects of local control of these tumors. The problem is that the dose that is used in the extremity is difficult to deliver in the retroperitoneum because of sensitivity of the surrounding viscera.

Slide 12:

Treatment factors, preoperative radiation here has some advantages. The reason is that we have a defined target, that we can use smaller fields, that sometimes, as I showed in that second slide, the tissue tumor itself can serve as a tissue expander to prevent radiation of important and vital surrounding structures.

We have a sense that giving preoperative treatment allows us to diminish, then, the GI toxicity that is often associated with post-operative radiation in these patients because of the factors related to the surgical manipulation of organs -- the fact that organs are fixed, they are much more difficult to get out of the radiation field.

Probably something that needs to be explored in much greater detail is this combination with a boost, the ability to deliver adequate radiation therapy by a combination of external beam plus some sort of internal radiation or a boost dose that is given post-operatively with some way of doing a bioprotecting procedure.

Slide 13:

I picked out of the most recent literature to demonstrate that there is some data with pre-operative radiation therapy in retroperitoneal sarcomas.

Most of these are principally single institutional series and are small because of the nature of the disease.

However, in this series, pre-operative radiation was given and the result was an overall feeling that this contributed to better local control rates but had no absolute effect on overall survival, if you can say that from single institutional series.

These three series from MGH, Mayo, and University of Toronto emphasized the use of boost dose radiation, some in the form of electron beam intraoperative radiation, and some in the form of giving pre-op radiation and then brachytherapy as a post-operative boost.

What these series did prove was that this was feasible, that the overall GI toxicity from pre-operative radiation in this disease site is relatively controllable and that biologically important doses can be delivered to these patients without overt morbidity.

Slide 14:

Of course there are also, as you would expect, data relating to single institutional series for post-operative radiation as well, again addressing the local disease control.

I have to remind everyone that there was a randomized series published beginning in 1993 from the NCI looking at this question and particularly looking at intraoperative radiation. This was surgery -- intraoperative radiation, post-op radiation. Half the patients received low-dose post-op, half received high-dose post-op, "high dose" meaning 55 Gray. Half the patients got intraoperative, the other half didn't. Obviously, the ones with high dose did not.

The findings here were based on very small numbers but did show a statistical advantage to local control in the patients who had intraoperative boost radiation.

Additionally, there were different toxicities, as you might expect. The group that had the IORT had more toxicity especially neurotoxicity because of overlapping fields and large field sizes with the intraoperative cones, and the morbidity -- radiation-induced morbidity -- in the post-operative radiation group was more GI toxicity.

There are some series that look at boost plus post-op radiation. So, the surgery is there, the boost, and then radiation is given post-op. Most recently, a series from Memorial looking at high-dose rate IORT followed by post-op radiation in lower doses. We must be reminded that there is a dose response here. Just like there is in the extremity, there is a dose response in the retroperitoneum. It seems to be clear that approximately 45 Gray has to be delivered in order to look at some sort of local disease control benefit. In many of these series, the dose-limiting toxicity is related to doses that are higher than that.

Slide 15:

So, conclusions based on what is known about radiation therapy and local control of this disease is that these are single institutional series for the most part. The addition of a boost of IORT or brachytherapy -- or I might add there are surgical techniques that might protect, putting in mesh or omentum or whatever is available tissue expanders (we have had some experience with that) to protect the sensitive normal structures -- appears to enhance local control without excess toxicity, but no evidence of a survival benefit to date. I might say there also may be some data to suggest that we are delaying recurrence in these patients but not necessarily eliminating that recurrence.

Slide 16:

Role of chemotherapy? If one were to extrapolate from the extremity experience, one could potentially create quite an argument. Arguably there is some data that suggest that high-risk profile patients will benefit with adjuvant chemotherapy in the extremity.

There is no data to suggest this in the retroperitoneum. There are studies that have looked at chemotherapy as radiosensitizers, in a sense giving it pre-operatively in combination with radiation.

Slide 17:

Conclusions, then: Surgical resection: resectability rates are marginal. Margins generally are inadequate. Local recurrence is common even after five years and is a major contributor to tumor-related mortality.

Slide 18:

Radiation: no proven benefit but data suggests that local control is improved with radiation of some type. However, there is a problem of dose-related toxicity. Combined external beam radiation therapy and boost appears to be superior for response.

Slide 19:

Chemotherapy: no proven benefit in primary disease, may be useful in combined modality therapy.

Slide 20:

These are the challenges for the clinical research and perhaps challenges for discussion. I think that this disease site and this type of particular soft tissue sarcoma group finds a necessity for combined-modality therapy. With this, there is an absolute necessity for multi-institutional Cooperative Group trials to define standard of care.

We have to come up with a rational, standard type of radiation in addition to standardized surgical procedures to address the question of local control. The only way you are going to do that is with a Cooperative Intergroup study of the type that will provide enough numbers to substantiate that.

Additionally, I think we can take advantage, there has been a lot of discussions of tumor biology and I think there is an opportunity to take advantage of the fact that an important histotype of retroperitoneal sarcoma is liposarcoma, particularly myxoid liposarcoma. If one could develop the small molecule cytostatic agents that took advantage of increased knowledge of the genomic control of initiation and differentiation and proliferation of these tumors, it might make an interesting combined treatment scheme, particularly as we understand more about the transcriptional control, the fat maturation of whatever stem cell these come from.

One could envision a time when we could turn these malignant, large, expanding tumors in the peritoneum into quiescent big globs of fat that essentially create cancer as a chronic, but non-lethal disease.

Slide 21:

I just wanted to, while I am here and have the podium, I just wanted to tell you that there is actually a Cooperative Group trial for retroperitoneal sarcoma at the present time.

This is a concept that was basically developed out of some pilot data from the M.D. Anderson hospital in cooperation with the Mayo Clinic, and they are a proponent of giving intraoperative radiation therapy or boost radiation, if you will.

Slide 22:

This is RTOG study 0124. This is the design of the study. These patients will all have essentially high-grade, either primary or recurrent, retroperitoneal sarcoma, as well as pelvic soft tissue sarcomas. They will receive a fairly intensive chemotherapy neoadjuvantly -- up to four cycles -- followed sequentially by external beam radiation therapy in a therapeutic dose range, followed by surgical resection and a radiation boost du jour. We could not come to a conclusion about what kind of radiation boost one would want to perform, depending on where you are practicing.

So, we gave everybody the opportunity to use their favorite type, including IORT where that was available, and brachytherapy, and also a variety of bioprotectant external beam radiation therapy post-operatively -- hopefully to areas of disease concern, and marked out carefully by the surgeon who was doing the procedure. That is it. That is all of my talk.