By Oliver Sartor, M.D.,
Chief, Hematology-Oncology Section;
Director, Stanley S. Scott Cancer Center,
LSU Medical Center, New Orleans
Edited from PCRI Insights November, 2005 vol. 8 no.4
It is possible that hormones can sometimes cure, but unfortunately, that seems to be true only in a minority of patients. Certainly if the disease has become significantly advanced at the time that hormonal therapy has begun, the probability of that cure is greatly diminished. There is clear data to indicate that the duration of response to hormonal therapy is inversely related to the volume of the disease.
Initially after androgen deprivation, there is a marked response in the vast majority of men. We now understand that the nadir PSA is an increasingly important prognostic marker. Based on a series of studies recently presented by Dr. Anthony D’Amico and others, a PSA of = 0.2 ng/ml level is prognostically important.1
A very pragmatic definition of androgen-independent or hormone-refractory prostate cancer involves a patient with progressive prostate cancer and serum testosterone of less than 50 ng/ml. At this point the disease is classified as either androgen-independent (AI) or hormone-refractory prostate cancer (HRPC), whichever term you prefer. I am not saying that that a testosterone level of less than 50 ng/ml is optimal, but it does represent a pragmatic definition. Surprisingly, no large clinical trial indicates conclusively that a lower testosterone results in better outcomes. However, this question has never been examined in an intellectually rigorous manner.
How does one define progressive disease? After all, the term, progressive, must be defined in precise terms. To establish progression of disease, there are three types of parameters that should be considered: (1) clinical parameters, (2) laboratory parameters, and (3) radiographic parameters. When dealing with a patient, the physician should listen carefully and understand how the patients feel. What are their symptoms? Is there pain or swelling of the extremities that could be edema from lymph node involvement? Are they experiencing appetite or weight loss, fatigue, or loss of energy? The symptoms of prostate cancer are actually fairly well defined. Though it is possible for urinary symptoms to predominate, more commonly prostate cancer symptoms in the hormone-refractory disease state relate to either general elements related to tumor bulk (fatigue, loss of appetite, or weight loss), or symptoms of metastases to bone (bone pain).
PSA is the most prominent laboratory marker, of course, but there are also tests like prostatic acid phosphatase (PAP), chromogranin-A (CGA), and neuron-specific enolase (NSE) tests that are sometimes useful in individual patients. The chromogranin-A and NSE represent markers of neuroendocrine differentiation.
Radiographic tests that are useful include bone scans, CAT scans, and MRI’s. These can be used to detect metastases. Most metastases occur in the bone (over 90% of patients with far advanced cancer will have bone metastases). The second most common site of metastases for patients with advanced disease is evidence of metastases in the lymph nodes (LN’s). When lymph nodes are involved and able to be detected with CT and/or MRI testing, the most common locations are near the aorta, in the abdomen, deep in the pelvis, or in the chest. LN’s greater than 2 cm in size occur about 25% of the time in patients with advanced disease. The only lymph node (about 3% of patients with advanced disease) found on exam is just above the left collarbone at the base of the neck. This is called a Virchow’s node. Other sites of metastases (lung, liver, adrenals) occur less than 5% of the time. Brain metastases occur in approximately 1% of patients.
Only when we put all of these variables together do we have the best way to assess a patient. So while PSA is the single best marker for asymptomatic patients, assessment requires more than just PSA, particularly in those who have advanced disease. For most patients, however, PSA is the single most sensitive marker of disease progression and it is typically the first detectable evidence of progression. At the time that PSA begins to rise (PSA <5 ng/ml) in patients who have failed prior therapy with radiation or surgery, the typical CT, MRI, and bone scan evaluation in that patients reveals no evidence of metastases.
There is a changing spectrum of hormone-refractory patients. When I started to work in this field in 1989, PSA was just becoming available. Sometimes we obtained the very first PSA that any patient in the clinic had ever had. Patients would come in with bone metastases and pain, and CT scans would not infrequently (about 20% of the time) reveal large lymph nodes. I calculated the average PSA in our clinic at the National Cancer Institute when I was working with Dr. “Snuffy” Meyers, and the average PSA in our clinic was approximately 550 ng/ml. Clearly that’s not true today. Now, we have patients who are just beginning to have their PSA rise after hormonal treatment. I often see patients with a PSA of 0.3 or 0.4 ng/ml and rising. I also see patients who may not have had a PSA nadir occur at an optimal degree.
Progression usually involves a rising PSA (but not always). We do see patients who progress with PSA’s that are extremely low and show progression in other ways. Right now, I have four patients in my clinic with very advanced disease who have a PSA less than 0.5 ng/ml. It clearly is possible that bone scans or CT scans can reveal progression in the absence of rising PSA; it is just not likely to occur.
How Should We Treat Hormone-Refactory Disease?
The following is the menu from which I choose:
1. Anti-androgen withdrawal (and other withdrawals). Anti-androgens include agents such as flutamide (Eulexin®), bicalutamide (Casodex®), and flutamide (Eulexin®).
2. Anti-androgen administration
3. Adrenal suppressives, such as ketoconazole
4. Corticosteroids such as Decadron® (dexamethasone), prednisone, and hydrocortisone
5. Estrogens like DES
7. External beam radiation therapy
8. Intravenous bone-seeking radiopharmaceuticals (samarium-153 or Quadramet®, strontium-89 or Metastron®)
9. Bisphosphonates (only zoledronate or Zometa® is FDA approved in prostate)
10. Chemotherapy (e.g., docetaxel or Taxotere®, and mitoxantrone or Novantrone®)
11. Experimental therapies
Withdrawal Responses in HRPC
Withdrawal therapies are interesting. We put a patient on a drug to help him, and then the drug somehow turns and becomes his enemy in a subset of cases. If we then take the drug away, some patients get better. Actually, withdrawal therapy has a long precedent in breast cancer. It has been known for years that estrogens could be given in breast cancer to induce remissions, and then could be withdrawn at the time of progression to induce remission again. I hypothesize that what happens in prostate cancer is that the androgen receptor, to which all of these anti-androgens will bind to, becomes mutated after time. Unmutated, those receptors will recognize an agent like flutamide as an antagonist (a drug that blocks the effects of the male hormones), but a critically located mutation will cause those patients to have a response if that agent is removed. In this case, the patient is getting away from the normal receptor context and getting into a mutant receptor context. This is hypothesis and not necessarily fact.
I have made progress with some patients by, what I term is thinking like a mutant receptor. Some mutants actually respond to hormones in an extraordinarily sensitive manner and “promiscuous” manner. Therefore, something like DHEA, which is ordinarily a very weak androgen, can be recognized by a mutant as a very potent androgen. So DHEA becomes like DHT (the most powerful natural androgen), instead of something weak. This phenomenon has been extremely well demonstrated in the lab, but less well demonstrated in the clinic because of the difficulty in designing the experiments. Nevertheless, anti-androgen withdrawal is something that we need to be aware of, because if a patient starts to have progression, we should discontinue the anti-androgen and find out if withdrawal therapy is going to work.
Megace® is used at times for patients who have hot flashes, and at times for patients to boost their appetite. But in prostate cancer, Megace may interact with the androgen receptor, particularly mutants, and cause excessive cancer growth. And you can actually get responses by withdrawing Megace. I do not prescribe the use of Megace in prostate cancer patients (even for hot flashes), because I don’t know who has a mutant and who doesn’t. But I do know that if you withdraw the Megace, that you can get responses. DES (diethylstilbestrol) has been associated with withdrawal responses, as have retinoids, and even agents like TNP470, an anti-angiogenic agent.
Figure 1 shows the progression-free survival results from the SWOG 9426 study.6 Most of the patients showed progression quickly after the anti-androgen was stopped, but some of the patients would go out two, three, and even almost four years after having nothing but flutamide withdrawal or bicalutamide withdrawal. So we learned not to be too quick to act. It is important to find out what happens when the anti-androgen is stopped because some patients will go on for a significant period of time.
|Figure 1. Antiandrogen Withdrawal: Progression-Free Survival Results from SWOG 9426|
When we did our multi-variant analysis, we found that bicalutamide was less likely to be associated with withdrawal responses than the other anti-androgens. But it was also interesting that the longer a patient was on the anti-androgen, the more likely it was that he was going to have a withdrawal response. In other words, if a patient was only on the anti-androgen for two, three or four months, the probability of withdrawal response was very low. But if a patient was on the anti-androgen for more than 32 months, the probability that he would have a withdrawal response was really quite high. These conclusions proved to be valid in the multi-variant analysis (where we took all the other factors into account).
Paradoxically, we have learned that we can also use these anti-androgens and get responses even if the first one has failed. Since we are dealing with a relatively slow-growing disease in many patients, it is certainly worthwhile trying different anti-androgens in selected patients. After all, the toxicity of these agents is low, and some patients will have a response. I have found that this approach is most likely to be effective for patients without evidence of metastatic disease. Hence, if a patient has a PSA rise only, and he is progressing after initial hormonal therapy, switching anti-androgens may in fact elicit a response. The duration of that response is highly variable.
Ketoconazole is an interesting agent; it is an antifungal agent that was approved by the FDA some years ago, but one of its side-effects turned out be a lowering of testosterone levels. It is inconclusive whether this testosterone-lowering ability explains its efficacy in hormone-refractory disease, but nevertheless, we do know that we can lower a testosterone level from X to X-1 or X-2 with ketoconazole. And many patients will in fact have a response. In some patients, this response can be very gratifying.
After hormonal deprivation, we call these HRPC cancer cells “resistant”, but actually they are just more sensitive to testosterone. So even working with simply decreasing testosterone, we can sometimes get more responses, although they don’t last forever. Eric Small did a reasonable phase II trial7 that was published in 1997. Table 6 summarizes what actually happened to Dr. Small’s patients. The toxicity was reasonable. The ketoconazole dose used here was 1,200 mg a day (400 three times daily) plus hydrocortisone. A substantial number of patients will respond to ketoconazole, and sometimes for a long period of time. I have a patient now with advanced disease who has had been responding to ketoconazole for three years, so clearly, this is something that is worth considering.
Glucocorticoids not only include the cortisol-type compound steroids that are produced by adrenal glands, but also hydrocortisone, prednisone, and dexamethasone (Decadron®). In addition to this anti-angiogenic activity in and of themselves, glucocorticoids also suppress the pituitary’s ACTH adrenocorticotropic hormone, which leads to a suppression of adrenal androgens such as DHEA and androstenedione. One of the interesting things about these multiple potential mechanisms is that they don’t kill cancer cells directly even if they are applied directly on cancer cells. I think their value is either via an anti-angiogenic effect or their indirect hormonal suppression effects on adrenal androgens.
Glucocorticoids are used in combination with ketoconazole, or in combination with such taxanes as docetaxel (Taxotere®), or mitoxantrone. It is important to remember that these glucocorticoids are active agents in and of themselves. I have observed many patients who attribute some magical change in their PSA to whatever factor that they may be taking from a health food store. But in fact, they were taking prednisone or Decadron® along the way. As a very specific example, one of my patients with severe allergies has a PSA that is moving along fairly slowly. Every time he takes a Medrol® Dosepack for his allergies, his PSA goes down. Originally, he was trying to attribute the PSA effects to something else, but when I put together his medical history, it clearly showed that the Medrol® was the agent. Another example is a patient who had temporal arteritis, a disease where the body attacks the blood vessels. The patient began taking prednisone alone, and his PSA came down. So we have to be careful of these factors; they are sometimes more active than one might think.
Dexamethasone monotherapy can be effective. In a Japanese Phase II trial8 published in 2002, Morioko et al used 1.5 mg dexamethasone a day. This is not an extraordinarily high dose, but 59% of the patients had a PSA decline of greater than 50%. Now, Saika et al conducted a small trial with the same parameters, but it had only a 28% PSA decline >50%.9 Even so, the bottom line is that a substantial number of patients have benefited from this therapy. When I ran (and published) a prednisone trial with a daily dose of 20 mg of prednisone, a third of the participants experienced a PSA decline rate of >50%.10 These are active agents, but they often get mixed in or used to treat everything from allergy to temporal arthritis to rheumatoid arthritis, various spinal conditions, inflammation, etc. Glucocorticoids can add value, but they too have side effects and consequences that must be taken into account.
DES is a prototypical estrogen, but there are other estrogens as well. Table 7 summarizes the results of a multi-institutional trial with DES or PC-SPES, led by William Oh from Dana-Farber, with a population of hormone-refractory patients. In the DES arm of the study, the >50% PSA declines were about 24% (a little lower than what I would have predicted).The median duration of response was 3.8 months, and the median time to progression was 2.9 months. DES can feed back on the pituitary and lower testosterone. Clearly, there is value added for estrogens even in hormone-refractory disease, and again, some patients can do quite well.
5-alpha reductase inhibitors, such finasteride (Proscar®) and dutasteride (Avodart®), can be used as well. An experimental 5-alpha reductase inhibitors trial11 in HRPC is summarized in Table 8. The participants were primarily hormonally naïve men, but there were also some HRPC men as well. We were surprised to see that four out of 15 HRPC men (27%) had PSA declines of 50% or more. We thought we might see something in hormone-sensitive disease, but instead, we saw more activity in patients who were already castrate.
One of these patients had responded for about four years. What would explain these results? There are several hypotheses. One might think that it is the decrease in the dihydrotestosterone that these agents induce, but fairly substantial increases in estrogen levels in the blood were also noted. The testosterone has to be broken down one way or another. And the usual way that it gets broken down is through the 5-alpha reductases that go into DHT. But if that pathway is blocked, the other way that the testosterone can become degraded is through the estrogen pathway. Hence, you actually increase the estrogens by using these particular agents. This had never been reported before, and it may explain the effects we saw, but we are not sure. Nor am I sure why estrogens work in hormone-refractory disease. I just know it’s an empirical fact, and I don’t always understand everything. It keeps me fairly humble.
Thalidomide is another agent that is not well understood. Table 9 summarizes the results of Doug Figg’s trial at the National Cancer Institute.12 It has a little bit of activity, but not a lot of activity. PSA declines of greater than 50% were present in about 18% of the patients who received 200 mg/day. This agent certainly exerts some anti-angiogenic activities after it is metabolized to various metabolites. There are studies that show that thalidomide is not active, that it is the metabolite that is active. These anti-androgenic agents, of which thalidomide might be one of a class, may not show a lot of PSA decline. Instead, there may be just stability and failure to progress. Moreover, this drug definitely has some toxicities, like constipation, fatigue, tingling, numbness, and peripheral neurotoxicity. What’s interesting, though, is the fact that when, in a fairly good-sized trial, thalidomide was put together with Taxotere, a prolonged survival was achieved. However, the trial uncovered an unexpected side-effect: thrombosis. If thalidomide is used with another agent, particularly a chemotherapy agent, there is a much higher risk of side effects like pulmonary embolism and deep vein thrombosis. In the Figg trial, there were a lot of clots initially, so they actually had to give heparin injections to every participant.
To summarize what we have learned about these agents, it is not necessarily the PSA decline that counts, it may be the time to progression that counts, and perhaps the type of agent utilized will have different end points than other agents do. Not all agents kill cells directly. Some interfere with the activity of the blood vessel growth, and in that case it is not a decline in PSA that counts, but rather a time to progression that may be prolonged.
External Beam Radiation
Although a lot of hemi-body or wide-field radiation is widely used more internationally, we don’t use it nearly as much in the U.S. Instead, we use local-field radiation, where the radiation is focused on one particular area. Often that is a bone lesion that is painful. However, a randomized trial with strontium revealed some interesting things. Patients who get external beam radiation to bone on one occasion face a virtually 100% probability that they will get radiation to bone in the future. In other words, bone metastases represented a systemic disease that was being treated with a focal therapy, and more radiation will be needed at some point. In Figure 2, the dotted-line curves represent radiation alone, and the solid-line curves illustrate strontium treatment. Figure 1 is from the pivotal strontium trial13 that led to the FDA approval of strontium, because it reduced the need for future radiation therapy. One problem with strontium-89, however, is that a recent European trial showed that patients treated with strontium-89 lived for less time than did radiation treated patients.
|Figure 2. Time to Further Radioactivity.|
Radioisotopic Therapy of Metastatic Disease
Prostate cancer is a disease that has a unique pattern of metastases. This has been known and accepted by everyone since the late 19th century when a Dr. Paget, who did a lot of work with bone, formulated a hypothesis termed the “seed and soil” hypothesis. The seed was the cancer cell, and the soil was where it landed. Prostate cancer metastases are uniquely distributed. I’ve never seen a metastasis to the heart in all my years. When I look for pulmonary metastases, I see only a very few. However, when I look at bone, metastases are very, very common. This is a disease that goes to bone and is also remarkably osteoblastic. There is some relationship between bone and prostate cancer that is not fully understood, but which I think is fundamental and very important.
Experiments done at Stanford, when they were first using radiation therapy revealed that when metastasis occurred, the cancer did not go to the areas that had been previously irradiated. This implied that the radiation had the ability to change the soil upon which the seed was landing. There are a variety of FDA-approved radio-pharmaceuticals, and the following have been used from time to time:
• Samarium-153 EDTMP
I have covered radioisotopes in a separate Insights article (Vol 8, No 2, May, 2005), so they will not be covered here in detail. Suffice it to say that these are interesting agents that deserve more use and more clinical trials. Their use in combination with chemotherapy is of particular interest. Samarium-153 EDTMP (Quadramet®) is the agent I think to have the best ratio of effectiveness to side-effects.15 It is an excellent targeted therapy.
Bisphosphonates have two potential uses in prostate cancer. One is to prevent or treat osteoporosis, the other is to prevent skeletal related events such as pathologic fractures. Zolendronate (Zometa®) is the only one that the FDA has approved in HRPC. In a prospective randomized trial with a composite end point, called skeletal-related events (SRE), there was an improvement, but it was not as dramatic as we would have liked it to be. There is lots of progress still to be made. The study organized the results in terms of SRE’s. Radiation to bone was reduced from 33 in the placebo group to 26 in the Zometa® group. Fractures were reduced from 25 to 17, although it should be noted that these were not all pathologic, cancer-induced fractures. Some of these may have been osteoporotic fractures. Other reductions were spinal cord compression from 8 to 4; surgery to bone, from 4 to 2.
Zolendronate and other bisphosphonates do not reduce pain; rather they reduce the rate at which pain will increase. And bisphosphonates have potential serious side effects such as osteonecrosis of the jaw. Any patient on zolendronate must be really careful about getting his teeth pulled or having extensive dental work. Moreover, the dose has to be adjusted if the patient has a kidney problem.
I won’t extensively cover chemotherapy here, but I do want to mention that in May 2004 the FDA approved docetaxel (Taxotere®) as a result of a trial called the Tax 327 study.14 The FDA-approved regimen for docetaxel is 75 mg/m2 every 3 weeks with prednisone 5 mg bid. If one looks at the survival curves for HRPC patients with quite advanced disease, and then looks at the comparison between mitoxantrone, which is a previous standard chemotherapy, the improvement with docetaxel is statistically significant, although it is not huge. I use docetaxel as my standard of care for patients with hormone-refractory disease once they come to chemotherapy. But nevertheless, it’s not the world’s greatest drug.
Treatment After Chemotherapy Failure
What can be done after primary chemotherapy has failed? I’m not sure if I have all the right answers, but I have a few ideas. I am co-principal investigator on a trial that now includes over 500 participants. It is a multicenter, multinational, double-blind, randomized Phase III trial of a platinum-type compound called satraplatin. I think there is something to the platinum compounds after the failure of docetaxel. We’re learning and focusing in on this area, trying to learn how to use them best.
The vaccines are becoming interesting. We know that the benefit is not likely to be huge; but consider a very small trial out of Dendreon. The overall survival was 25.9 months versus 22, so it’s about a 3.9-month advantage. That’s not the biggest advantage now, but nevertheless, this is a promising proof-of-concept trial. After all, it is a breakthrough, albeit a modest one.
There are a wide variety of newer agents we’re looking at in the field. Among them are the following:
Vaccines and immune stimulants
Small molecular growth factor antagonists
And this is not an exhaustive list. Very provocative, very interesting trials are upcoming, and we want to see them move earlier in the course of disease. Proper clinical trial design is critical, absolutely critical. As you know, patients can survive a long time if they don’t have metastases without progression. It depends on a lot of things including the rate of PSA rise, nadir PSA, etc. The whole field is actually moving fairly fast. This is an exciting field with exciting prospects.
HRPC is a disease with a number of options. We generally start with simple and relatively non-toxic alternatives in the asymptomatic patient then as the symptoms progress, more complex and potentially more toxic compounds are utilized. Secondary hormonal manipulations such as anti-androgens, ketoconazole, estrogens are often a good place to start. Docetaxel has been shown to prolong survival in large clinical trials and is the best form of chemotherapy available today. Radiopharmaceuticals such as samarium-153 are probably underutilized. External beam is a mainstay for painful skeletal metastases. Zolendronate is the only bisphosphonate FDA approved in HRPC. A variety of experimental therapies are now underway and clinical trials should be sought by patients and clinicians alike.
1. Stewart AJ et al: The clinical significance of a PSA nadir > 0.2 to patients with a rising post-operative or post-radiation PSA treated with androgen deprivation. Abstract 4547, ASCO Annual Meeting, 2005.
2. Newling DW et al: Orchiectomy versus goserelin and flutamide in the treatment of newly diagnosed metastatic prostate cancer. Analysis of the criteria of evaluation used in the European Organization for Research and Treatment of Cancer–Genitourinary Group Study 30853. Cancer 72(12 Suppl):3793-8, Dec. 1993.
3. Oefelein MG et al: Survival of patients with hormone refractory prostate cancer in the prostate specific antigen era. J Urol 171(4):1525-8, Apr 2004.
4. Bianco FJ et al: Prognosis after androgen deprivation therapy in men with a rising PSA after prostatectomy. Abstract 4552, ASCO Annual Meeting, 2005.
5. Sartor O, Koocheckpour S: Stem cells and prostate cancer. Clin Prostate Cancer 3(1):11-2, Jun 2004.
6. Sartor O et al: Anti-androgen withdrawal in prostate cancer: results from SWOG 9426. Abstract 785, ASCO Annual Meeting, 2002.
7. Small E et al. Ketoconazole retains activity in advanced prostate cancer patients with progression despite flutamide withdrawal. J Urol 157(4):1204-1207, Apr 1997.
8. Morioka M et al. Prostate-Specific Antigen Levels and Prognosis in Patients with Hormone-Refractory Prostate Cancer Treated with Low-Dose Dexamethasone. Urol Int 68(1):10-15, 2002.
9. Saika T et al. Treatment of androgen-independent prostate cancer with dexamethasone: a prospective study in stage D2 patients. Int J Urol 8(6):290-4, Jun 2001.
10. Sartor O et al: Effect of prednisone on prostate-specific antigen in patients with hormone-refractory prostate cancer. Urology 52(2):252-6, Aug 1998.
11. Eisenberger M et al. Phase I and clinical pharmacology of a type I and II, 5-alpha-reductase inhibitor (LY320236) in prostate cancer: elevation of estradiol as possible mechanism of action. Urology 63(1):114-9, Jan 2004.
12. Figg WD et al. A randomized phase II trial of thalidomide, an angiogenesis inhibitor, in patients with androgen-independent prostate cancer. Clin Cancer Res 7(7):1888-93, Jul 2001.
13. Porter AT et al. Results of a randomized phase-III trial to evaluate the efficacy of strontium-89 adjuvant to local field external beam irradiation in the management of endocrine resistant metastatic prostate cancer. Int J Radiat Oncol Biol Phys 2;25(5):805-13, Apr 1993.
14. Tannock IF et al: Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med 351(15):1502-12, Oct 2004.
15. Sartor O, Reid RH, Hoskin PJ, et al: Samarium-153-Lexidronam complex for treatment of painful bone metastases in hormone-refractory prostate cancer. Urology 63(5):940-5, May 2004.