Other major prognostic factors

Other major prognostic factors are histopathology, and I’ll go over some of these at the end. Lymph node involvement, which is extremely rare, connotes a very poor prognosis. And you can see that with epithelioids, synovial, rhabdo and angio. For the most part though we don’t recommend doing lymph node dissections because for the most part sarcomas do not go to the lymph nodes. Age; those patients under 60 usually do better than over 60. Females, as for most tumors, do a little bit better than males. And then there’s some literature on P-glycoprotein and Ki-67 expression, and aneuploidy. Debatable literature. Ploidy, you can find in both directions, whether it has significance or not. P-glycoprotein is for measurement of MDR and Ki-67 expression, DNA synthesis. Obviously those patients who have a higher DNA synthesis do somewhat worse.

We have two major staging systems for soft tissue sarcoma. One of them is the Enneking system, put together by Dr. Enneking who is sort of the father of orthopedic oncology, who was at the University of Florida. It’s a bi-gradal system; low grade, high grade and then whether the tumor is either intra-compartmental or extra-compartmental, and then if the patient has nodes or metastases they would be stage III. This emphasizes compartmentalization. It is best suited for extremity sarcomas. It’s not good for other areas of the body, and size and depth are not included. It’s the system though that’s probably used the most by orthopedic oncologists. Now our other grading system is the American Joint Cancer Committee system and this again has been changed in 2006 from a tri-grade to a bi-gradal system. Putting well-differentiated and moderately- differentiated in one group, and poorly and un-differentiated in another. But in addition to that, it looks at size – less than or greater than 5 cm – and it also looks as to whether the tumor is superficial or deep. The tumors that specifically we want to look at, that have probably the highest rate of recurrence and where we possibly could look to give adjuvant therapy, would be in the high grades that are somewhat superficial, but more the high grades that are large and that are deep. Those are the ones that have the most potential for recurrence and where we would think possibly of giving adjuvant chemotherapy. Obviously those that do the worst are those who have nodal involvement or metastasis.

Now this is what a typical soft tissue sarcoma looks like, and I put this up to show to you that it’s critically, critically important that you get this patient to a surgeon who knows what they are doing the first time. Many of these tumors, the majority of these tumors, have a pseudo capsule around them and many many a time the patient gets the “oops” procedure. In other words, the surgeon goes in and sees, “Boy that capsule, this is easy. I’m just going to slide it right out” and the tumor is out. Then the pathology comes back and all the margins are positive. And why are the margins positive? Because there are satellite areas of tumor around that pseudo capsule, so you just don’t want to scoop it out. If possible, you want to get a wide excision with at least 1-2 cm of normal tissue around it. When we do a marginal resection, a shell-out, there is about a 90% incidence of local recurrence.

Even with wide excision, there can be about 50% in older series. When we do a radical, it is less than 15%. Sometimes we still have to do an amputation, and this is really based on the biopsy site and the biopsy being done properly, as I said before, whether there is neurovascular or compartmental involvement, the size of the tumor, and what the anticipated function is. If the patient, after the surgery, is left with very poor function then there is no point in doing a limb-sparing procedure. So in terms of surgical treatment, there has been a transition from doing a shelling out, where there is 90% local recurrence, to an amputation, to now limb-sparing surgery plus radiation, where there is about a 5-15% local recurrence rate. In a trial done at NCI comparing limb-sparing surgery plus radiation, versus amputation, there was no difference in survival.

Soft tissue sarcomas

Now we’ll first talk about the soft tissue sarcomas, and these are usually solitary, painless and palpable. The majority occur in the extremities. In the lower extremity, 40% of these are the highest amount and 75% of these are above the knee. They can also occur in the trunk where the most common area is the retroperitoneum, and then also rarely in the head and neck area. In terms of sites and histologies, usually in the extremity its MFH, synovial, lipo and fibro and most of these patients develop lung mets. The tumor usually avoids the liver. In the retroperitoneal area, lipos, leiomyosarcomas, there’s intraabdominal spread as well as liver and lung mets. In the chest wall, desmoid, lipo, leiomyo. The GI tract, leiomyosarcomas and then the newer-named GI stromal and GI autonomic tumors, which are extremely resistant to chemotherapy. In the GU tract, leiomyos and mesodermal mixed. Now what’s extremely important when you first see these patients, especially if this is an extremity sarcoma, is that you obtain the tissue in the right format. It’s really recommended that if possible you refer these patients to surgeons who have experience in doing the ultimate procedure. Really, if you are going to do a biopsy, you want to do an incisional biopsy, not an excisional biopsy unless it’s a very small lesion. And you want to make sure that the plane of that dissection is longitudinal to the muscle and not perpendicular. If you have the facilities, probably now at most major centers the majority of biopsies are done as a core true cut. I would say at our center about 95%, 98% of our biopsies are done by our radiologists. The advantage to that is that under CT guidance they can hit different areas of the tumor, because sometimes some of these tumors have necrotic areas and you may get no tissue or you may get a low grade tissue in one area and a high grade in another.

This is a study, an old study. There have been two other studies more recently. One from our institution, one from Memorial, looking at the value of core and needle biopsy compared to FNA. And you can see there is a much higher sensitivity, specificity and accuracy. Really it’s recommended to get at least three core and needle samples. We usually only use FNA’s to confirm a recurrence. Because it’s real important not only to determine, if possible, the exact histopathology of the sarcoma, but also to determine the grade. What’s critically important also is that you have a pathologist who has some experience in reading these and has some familiarity with the different immunohistochemical stains. You need to be aware that for some of these, in terms of helping delineating the type of sarcoma you are dealing with; angiosarcomas having CD34 factor 8; leiomyosarcomas having desmoid muscle-specific and smooth muscle actin; Ewings sarcoma having HB71013. None of these are absolutely specific but they can be very helpful in delineating the type of sarcoma you are dealing with. In addition, more recently now, there have been several different fusion genes based on different translocations that have occurred for some of these sarcomas, that are fairly diagnostic. One of them being the translocation of 11-22 for Ewings or primitive neurectodermal tumors. Also for synovial sarcoma, a translocation of X and 18. Looking at SID and SS-1 and SX-2 and we have found now that not only can these be somewhat diagnostic for diagnosing the sarcoma, but in this instance with synovial sarcoma it can tell whether you are dealing with a biphasic or a monophasic tumor. The biphasic being SS-1, the monophasic being SX-2 And not only is it helpful in delineating that but also in prognosis. The monophasic tumor being all spindle cell, having a much better prognosis than a biphasic tumor where there is an epithelial component as well.

Now what do we do in terms of a work-up? Well, first of all we want to find the local extent of the tumor and probably the best test for extremity, trunk and head and neck is an MRI. But in addition to that, if it’s in the retroperitoneal area or an intraabdominal sarcoma, we want to get a CT because we want to look at the abdominal contents and also evaluate the liver where the tumor can metastasize to. Then in addition, for all these sites, they have the potential to metastasize to the lung. So we want to get a chest CT as well. Now there are several major prognostic factors. Histologic rate is extremely important. Low grade tumors tend to stay locally, tend not to metastasize, tend to have a much better prognosis. We need to know the extent and location of surgical margins and those should be delineated by the surgeon. We want, if possible, to get margins of at least 1-2 cm. That is not always possible, depending where the tumor is. It’s much harder if it’s in the retroperitoneal area, much easier in the extremity area. Size becomes important. Those tumors greater than 5 cm and even greater than 10 cm are much more likely to recur and have a worse prognosis. The primary site becomes important. Distal tumors usually do better than proximal tumors, they are picked up earlier, usually easier to resect. Those tumors that are subcutaneous usually do better than deep. Those tumors that are intra-compartmental usually do better than extra-compartmental. Those tumors that are in the extremities usually do better than the trunk, and head and neck usually do better than the retroperitoneum. Retroperitoneum tumors are usually picked up late because they are usually very big, the patient doesn’t have any symptoms, then once they are picked up they are usually very near vital structures and it’s very hard to do an adequate full resection to remove all the tumor. In fact, for most patients who are at major centers getting surgery for retroperitoneal tumor, approximately 75% of the time a vital organ or part of a vital organ has to be removed in addition to the sarcoma.

Sarcoma

Sarcoma
A sarcoma (from the Greek ‘sarx’ meaning “flesh”) is a cancer of the connective or supportive tissue (bone, cartilage, fat, muscle, blood vessels) and soft tissue. This is in contrast to carcinomas, which are of epithelial origin (breast, colon, pancreas, and others).
Sarcomas are quite rare tumors, tumors that you don’t see that often. There are approximately 7,800 soft tissue sarcomas occurring in the United States, and this is about equivalent to the number of non-Hodgkin’s lymphomas, a little bit more than Hodgkin’s disease, but it’s a much rarer tumor than colon, lung, breast which are 170,000 to 180,000 a year. There can be the classical type of extremity and trunk, visceral involving the GI tract the GU tract, also KS and mesothelioma, and then even rarer are the bone sarcomas, approximately 2,600. The major one being osteogenic sarcoma but also chondrosarcoma, Ewing’s and MFH of bone. Although these tissues arise from about 75% of the average body weight, the just represent less than 1% of all adult and 15% of pediatric malignancies.

In terms of etiologies, exposure to radiation is certainly a cause of the development of sarcomas. Anywhere from 2000 to 7800 centigray and mostly we see osteogenic sarcomas, but also MFH, angio and fibrosarcomas. There is literature on chemical exposure and this comes from Vietnam and also from the farm belt in which exposure to herbicides and dioxin and oxyacetic acid can increase your chances of developing sarcoma. But there is quite a lot of debate in the literature about this. Also exposure to vinyl chloride and arsenic evolving into an angiosarcoma. Patients who’ve had renal transplants, chronic lymphocytic leukemia, autoimmune hemolytic anemia, can develop Kaposi’s. Then in terms of viral etiologies, the Kaposi’s sarcoma virus, HHV8, and also some literature looking at Epstein-Barr virus in children. Those children exposed to Epstein-Barr virus having a higher incidence of smooth muscle cell tumors. There is some literature, small usually, case reports of different injuries; scars, burns leading to fibrosarcoma. Implants and bone infarcts leading to osteogenic. There is the postmastectomy lymphedema, Stewart-Treves syndrome in which you develop an angiosarcoma. There’s a higher incidence of osteogenic sarcoma in those patients who have had Paget’s disease, and then certain benign forms of bone disease can also lead to osteogenic sarcoma.

In addition, in terms of genetic predispositions, those patients who have had a retinal blastoma develop about one thousand times more incidence of osteogenic sarcoma, and there is even a further risk if those patients have been exposed to radiation therapy. But you can also develop osteogenic sarcomas outside of the radiation therapy site. Rp53 17P _ in which there is a high incidence of sarcomas, breast, leukemia, brain tumors. The murine double minute 2, which is an oncogene on chromosome 12, and this can inactivate both P53 and a retinoblastoma and therefore gives you similar effects. Neurofibromatosis, a higher incidence when you have both a genetic defect of 17Q, where there would be a higher incidence of neurofibromas, and then adding on the P53 17P with an increased incidence of malignant schwannomas and neurofibrosarcomas. Also with Gardeners, the APC5Q, there is a higher incidence of intraabdominal desmoids.

Off of a clinical trial

Off of a clinical trial, what we now recognize, is that there is an intermediate level of hormonal sensitivity that we call, for lack of a better term, androgen- independent-but-not-yet-hormone-refractory disease where patients will respond to alternative hormonal manipulations. So typically what we do is, for a patient who is on combined androgen blockade, we will first stop the anti-androgen, Casodex or flutamide, because in about 25-50% of patients with stopping that drug the patients will go into a remission, generally of short duration. The mechanism here is not completely known but it is felt to be mutations that occur within the androgen receptor that allow the androgen receptor to become somewhat promiscuous and be stimulated by the anti-androgen. When you withdraw the anti-androgen, those clones will die off and that’s what one see clinically.

Next, one can introduce an alternative anti-androgen, specifically high dose Casodex or flutamide, and one can see patients respond, interestingly, to an alternative anti-androgen when the other anti-androgen was used. And finally, one can use an anti-adrenal agent. My drug of choice here is ketoconazole or Nizoral, which is tolerated by most but not all people. Some develop fatigue or GI symptomatology. In general, one can use 200 mg three times a day of this drug, although higher doses have been used with greater toxicity. If one tells the patient that one should take it without an H2 blocker on an empty stomach, one can get by with lower doses because one needs acid in the stomach for adequate absorption of ketoconazole. Aminoglutethimide has been used but it has greater toxicity than ketoconazole but similar response rates.

Now what about chemotherapy in this disease? Chemotherapy had been felt not to be very active in this disease for many years as a result of pilot studies, phase II studies, randomized studies, etc. until this study was done. Very clever study done by Ian Tannic and his colleagues in Canada. A study that looked at quality of life as its primary endpoint, in 161 patients with symptomatic advanced prostate cancer. Patients were either treated with 10 mg of prednisone per day or alternatively with prednisone plus mitoxantrone at 12 mg per meter squared every three weeks. Once again, primary endpoint was pain palliation. What they found was there was a difference in the rate of pain relief and the duration of pain relief, both of which were statistically significant. Another study that we did in parallel to this study in CLGB, almost 250 patients randomized to hydrocortisone plus mitoxantrone or hydrocortisone alone. Unfortunately we looked at survival as a primary endpoint. No difference in overall survival but we also tallied quality of life endpoints and in this particular study we did see a difference in quality of life associated with mitoxantrone. So on the basis of this study and the Canadian study, mitoxantrone was approved as a first chemotherapy drug for use as palliation in patients with advanced prostate cancer. Since that time, we’ve recognized that the old drug, estromustine – which was initially made as a designer drug to be targeted to hormone sensitive cells deliver a nitrogen mustard – very modest activity by itself but when combined with microtubule inhibitors generated response rates that were very respectable. With 50% reductions in PSA seen in over 50% of patients.

So these remain the most active regimens and are now being used clinically but have never been compared to either mitoxantrone/prednisone or to no therapy, and there have been no chemotherapy agents or regimens that have ever been demonstrated to improve survival in this disease. Although we do suspect at the present time that if one did a randomized study using Estromustine-based regimens that one would see a benefit in survival. So in summary, with regard to chemotherapy, mitoxantrone is used in this context for palliation. Estromustine-based regimens have greater activity but we are still uncertain, in terms of their impact in overall survival – since no randomized studies have yet been done – and certainly no randomized studies have yet been done in the context of earlier disease using chemotherapy, although those are planned now as well. Clearly we have to move beyond chemotherapy in this disease and there are a number of strategies, experimental trials, in the clinic right now that offer potential for exciting advances in this disease.

Treatment of advanced prostate cancer

Let’s move on to the treatment of advanced prostate cancer. Hormonal therapy remains the standard and the strategy here involves lowering testosterone levels or blocking androgen levels. The tried and true methods for doing this are orchiectomy or LHRH analog. Some more recent innovations that are not standard are peripheral androgen blockade, using an antiandrogen like flutamide or Casodex with or without a 5-alpha reductase inhibitor. There are a number of phase II studies, no comparative studies. This is not a standard in the United States, nor is intermittent hormonal therapy where a patient goes on orchiectomy, LHRH analog or combined androgen blockade for some period of time, stops it, starts it, stops it, starts it. No randomized studies yet although there is one randomized study that is accruing patients right now, looking at this strategy. Combined androgen blockade remained the standard in the United States for many years for patients with advanced prostate cancer. Based on a large Intergroup study that was published in the New England Journal in 1988 comparing daily injections of Lupron versus Lupron plus flutamide demonstrating an improved relapse rate and survival rate associated with combined androgen blockade. Amazingly, since that time, there have been 28 or 29 other randomized studies, including this one that has looked at the concept of combined androgen blockade. This is the largest study that I know of. Patients had metastatic disease, randomized to orchiectomy or orchiectomy plus flutamide. In this study there was a slight improvement in the ability to render the PSA less than 4 with combined androgen blockade, but no overall difference in progression free or overall survival. When one combines all the studies in a metaanalysis there appears to be a very very modest improvement associated with combined androgen blockade, which is statistically significant but arguably, is it really clinically significant?

So for patients with advanced prostate cancer, that is D2 disease – overt metastases in the bones or lymph nodes above the aortic bifurcation – there is probably only a modest impact on overall survival. So the use of an anti-androgen combination with an orchiectomy or an LHRH analog is particularly important in the symptomatic patient in order to block the flare pattern early in the disease when an LHRH analog is used. The combination approach may promote the selection of clones that have an altered androgen receptor, which is a problem, but the use of combined androgen blockade is still a wide open question in earlier forms of the disease, locally advanced disease, the rising PSA patient, etc. But for D2 disease we are moving away from the routine use of long term anti-androgen in addition to orchiectomy or LHRH analog.

This is a fairly typical scenario of a patient with advanced prostate cancer who was treated with hormonal therapy. The patient had a PSA of 600. Started on combined anti-androgen blockade. His PSA rapidly decreased to near undetectable levels. The decrease in PSA is a reflection of several things. One is an artificial decrease in PSA. The PSA gene is partially androgen stimulated so when one removes androgen, the biochemical production will go down automatically without anything happening to the cells themselves. Part of this reduction is due to cell loss. There is some apoptosis that is going on and there is some dormancy as well. So this is some combination of effects that include a biochemical effect as well as a true cellular effect causing dormancy or apoptosis. The PSA is an incredibly reliable test in predicting relapse in this disease after hormonal therapy. The PSA will remain at some level for some period of time, which may be a few months or many many years. The earliest sign of relapse will be a rising PSA which will predate the radiographic or symptomatic progression. In this case, by a few months. Sometimes by many many years. Very good predictor of relapse and obviously a controversial issue is when to institute further therapy here. There is no consensus on this right now. In general patients should go on clinical trials when this happens.

Prostate cancer

A slightly different story with radiation therapy. Prostate cancer

A slightly different story with radiation therapy. Three RTOG studies looking at hormonal therapy followed by radiation, versus radiation therapy alone. I just want you to look at these two studies down here, comparing Zoladex or LHRH analog versus no concomitant radiation therapy. Difference in local control rates in disease free survival, so far no difference in overall survival when combined androgen blockade was used in combination with radiation. Once again, local control rates improved, disease free survival rates improved. As of yet, no difference in overall survival. The most important study though was a study published in the New England Journal last year, a European study that looked at high risk, localized prostate cancer treated with radiation versus radiation therapy plus three years of androgen deprivation; 455 patients. Local control improved with hormonal therapy, relapse rates improved with hormonal therapy and there was a 20% actuarial five year difference in survival. So the concept of using hormonal therapy plus radiation appears to have some support at the present time. It appears that longer term hormonal therapy is required. Whether that will work for surgery or not remains to be seen. But the standard of care in the United States for locally advanced prostate cancer, I think at the present time – this is T3 tumors, high grade tumors – remains combination therapy with hormonal therapy plus radiation therapy or more, which I’ll come back to.

So we can sort of design how patients should be treated based on their risk group. Good risk patients – lower PSA’s, lower grade tumors, lower clinical stage, fewer biopsies that are positive. If cure is necessary, that’s a hard question to answer, but if it is necessary the standard of care remains radical prostatectomy but what we would like to do is minimize morbidity, and that’s why brachytherapy or seed implants have become popular over the past few years because of the sentiment that this therapy may in fact be equivalent in terms of outcome or close to equivalent, and less morbid. There are some patients in this subgroup that can be watched and external beam radiation still remains an alternative therapy here. For intermediate risk patients; some patients will be cured with radical prostatectomy, 50% will not and therefore multi-modality therapy should be contemplated here. Either standard multi-modality therapy with hormonal therapy plus external beam therapy, or in the context of a clinical trial. For high risk patients; these patients are not going to be cured, for the most part, with local therapy with external beam radiation or surgery. So the current strategy is these patients should be enrolled in clinical trials, with some experimental agent, hormonal therapy, chemotherapy, etc. prior to surgery or prior to radiation therapy.

Now if a patient is treated with surgery or with radiation and they are not cured, as evidenced by a rising PSA, this presents a huge clinical dilemma for the physician and a quandary for the patient as well. A very heterogeneous group of patients, how do we determine what to do with these patients? Should they undergo immediate hormonal therapy or delayed hormonal therapy? Well, the things that help us make a decision here are the rate change of the PSA and the tumor grade which will clearly determine prognosis on the basis of several studies now. The absolute PSA level is less predictive, although we use that as a benchmark for patients in terms of saying, let’s say, “When your PSA gets to this level then it is reasonable to start therapy.” But probably the dominant determinant of when therapy is instituted is the balance between the anxiety of the patient with a rising PSA versus the quality of life aspects of starting a patient on hormonal therapy. So it’s still a wide open area and there are no answers here yet, but we individualize therapy as best we can here.

Now what about treatment for this disease?

Now what about treatment for this disease? Well, hopefully we have progressed a little bit from this “voice-mail” answer for patients a few years ago. Once again, there are a number of different ways to treat this disease. There are some patients that don’t need to be treated. There are no randomized studies that compare treatment to no treatment. The standard in the United States though, for organ-confined prostate cancer, remains a radical prostatectomy and different forms of radiation therapy remain an alternative, and in general a quality of life alternative. Hormonal therapy could be used in some patients and cryosurgery as well.

Prostate cancer

Now how do people do with treatment? Well, we can say something about their likelihood of being cured. There are now two large studies, one from the Cleveland Clinic and one from our institution, looking at the comparison – this is not randomized data, but controlled retrospective data looking at different factors at baseline – and looking at the comparison of external beam radiation to surgery. What one can say is that no matter what one does with localized prostate cancer, the likelihood of being alive at five years is good and equivalent between both of these different types of treatment. And the same can be said at 10 years as well. The outcome appears to be similar with surgery and with radiation in these retrospective, non-randomized types of studies. What the outcomes will look like at 15 and 20 years after treatment remains to be seen. Another thing that we can tell patients is something about the morbidity of treatment. So we can tell people the likelihood of being cured, we can tell people the morbidity of different treatments – we have some good data there – what we cannot tell people is what is their need to be treated. That remains largely unknown. This is a study that we did. This is the only prospective study that looked at treatment outcomes after radiation and surgery in which questionnaires were given to the patients prior to treatment as well as after treatment. The leader of the study was Jim Calcott in our group. We are only looking at radical prostatectomy and external beam radiation here. When asked the question of whether the patients are developing urinary incontinence, radiation does not cause urinary incontinence where surgery does. Fully one-third of individuals two years after having a radical prostatectomy were still wearing absorptive pads because of concern of leakage or true leakage. And 13% of men said that they had a lot of incontinence. One of the disappointing findings in this is that sexual dysfunction was very high with either form of therapy. With radical prostatectomy the majority of patients were rendered impotent as a result of surgery, by the definition of inadequate erections for sex. It really did not matter whether or not they had the nerve-sparing procedure. It is only the subset of individuals who had bilateral nerve-sparing procedures, who were relatively young, where the results were reasonably high. About 50% of those individuals maintained potency. Radiation has its problems as well, although radiation does not cause immediate impotence, it does cause comparable levels of impotence which occurs slowly over a several-year period of time.

An issue that comes up in the treatment of early prostate cancer in under-staging. This is best illustrated, I think, from this paper that was published in JAMA a few years ago. Once again, out of the SERE cancer registry looking at 3,000 men who had undergone a radical prostatectomy in the late 80’s, early 90’s. The important finding here is that over one-third of patients who were treated with radical prostatectomy required some other form of therapy within the next five years; either hormonal therapy or radiation therapy because they had relapsed or there was some concern of relapse. So what had become popular 5-10 years ago were strategies to combine hormonal therapy with either surgery or radiation to try to improve the cure rates of these particular forms of local therapy. In the context of surgery, there were three randomized studies comparing surgery to preoperative hormonal therapy followed by surgery. In these studies hormonal therapy was given for a short amount of time, generally three months. In all the studies there was a reduction in the likelihood of having positive margins when hormonal therapy was used, but no difference in disease free or overall survival. So at the present time, preoperative hormonal therapy is not the standard of care. Now it may be that this concept is completely flawed, or alternatively one requires longer term hormonal therapy in order to see an effect in relapse and overall survival. But that remains unanswered.