Androgen-independent prostate cancer

When cancer advances despite primary hormone therapy

We often hear the term prostate cancer and assume it is one disease. Practically speaking, it is. On a molecular level, however, scientists are revealing a far more complex picture. Cancer has an innate ability to adapt to its surroundings. As it progresses, cancer cells tend to change, morphing to a point where the differences between tumor cells can be dramatic. That’s why some researchers believe late-stage prostate cancer is more accurately described as a mix of cancer cell types.

Each year, an estimated 25,000 men will find out their prostate cancer has changed enough to become resistant to standard androgen-deprivation therapy, also called hormone therapy. At this point, the cancer is classified as androgen-independent prostate cancer (AIPC) or hormone-refractory prostate cancer, meaning that the cancer is still able to thrive despite hormone treatment.* The first sign of AIPC is typically a rising PSA level, a shift that can be extremely distressing for patients, but not entirely unexpected. A majority of patients with AIPC will have already watched their PSA levels rise at two previous critical junctures: before their initial cancer diagnosis and after supposedly curative local treatment. When PSA levels rise again, despite primary hormone therapy, it represents a third critical decision-making point for patients.

Hormone therapy, a mainstay of prostate cancer treatment for more than 60 years, has been shown to shrink tumors, reduce pain associated with bone metastases, and extend survival. Hormone therapy, which relies on drugs called luteinizing hormone–releasing hormone (LHRH) agonists, deprives the body of testosterone, the male hormone (or androgen) that fuels prostate cancer cell growth.* The drugs prevent the secretion of LHRH, which is the brain’s chemical signal to start testosterone production (see Figure 1).† Treatment causes testosterone levels to drop by up to 95%. Some doctors add a second drug, called an anti-androgen, to block the effects of any remaining testosterone on prostate cancer cells. When an LHRH agonist and an anti-androgen are given together, it’s called a combined androgen blockade.

In some patients, hormone therapy may slow disease progression for more than a decade; in others, it may keep cancer in check only for a few months. Eventually, prostate cancer cells begin to resist the treatment. Why do responses to hormone therapy vary so much? How does the cancer become resistant? Is it resistant to all types of hormone therapy? What options do patients have in terms of treatment? To learn more about androgen-independent prostate cancer, the Harvard editors invited three experts from Harvard Medical School to address these questions and offer insight into this phase of the disease and ways to manage it. The panelists were

• Glenn Bubley, M.D., a medical oncologist and director of Genitourinary Oncology at Beth Israel Deaconess Medical Center. He has been a principal investigator for studies looking at cancer cell biology and the effectiveness of therapies in prostate cancer patients. He is an associate professor of medicine at Harvard Medical School.
• William Oh, M.D., a medical oncologist and clinical director of the Lank Center for Genitourinary Oncology at Dana-Farber Cancer Institute. He has served as a lead investigator for numerous studies on therapies for prostate cancer, including investigational chemotherapy regimens. He is an associate professor of medicine at Harvard Medical School.
• Mary-Ellen Taplin, M.D., a medical oncologist at the Lank Center for Genitourinary Oncology at Dana-Farber Cancer Institute. Her research focuses on molecular changes in prostate cancer cells as well as novel, second-line hormonal and chemotherapy regimens. She is an assistant professor of medicine at Harvard Medical School.

Marc B. Garnick, M.D., moderated the session.

Understanding the nature of AIPC

How do you define AIPC?

OH: With AIPC we’re talking about men who have been on hormone therapy or who have had a bilateral orchiectomy and whose testosterone levels have dropped to less than 50 ng/dl,* and who subsequently show evidence of disease progression. This can be in the form of rising PSA levels or bone or soft tissue metastases.

BUBLEY: Typically, when we talk about hormone therapy, we’re referring to the primary hormone therapies, which are the LHRH agonists leuprolide (Lupron) or goserelin (Zoladex). We confirm that the patient has been compliant in taking his medications, and we test his serum testosterone level to rule out the rare chance that the medication hasn’t dropped the testosterone far enough. If the PSA level starts rising in a low-testosterone environment — what we call a castrate environment — we know the patient has entered a new phase of prostate disease: AIPC.

Has the clinical presentation of AIPC been changing?

TAPLIN: It is evolving. About three-quarters of the patients I see have rising PSA levels after initial primary treatment without evidence of metastatic disease, or what we call biochemical failures, which is an increasing proportion. We are also picking up very small metastases due to improvements in imaging techniques. So we’re spotting signs of progression much earlier than in the past.

OH: Over the last decade, as PSA levels have been more closely monitored, therapies have started earlier. Ten years ago, we would typically wait for patients to have radiographic evidence of metastases, such as a positive bone scan, or symptoms of disease before starting hormone therapy, regardless of PSA doubling time. Because men are now frequently starting hormone therapy based on rising PSA levels, some are developing hormone-refractory disease, or AIPC, before they develop visible metastases.*

How is AIPC different from prostate cancer that has not yet been treated with hormone therapy?

OH: Androgen-independent prostate cancer is very heterogeneous, and we don’t fully understand the biology of it, but it does tend to present in different ways. It commonly spreads to the bones, but it may also spread to other organs or the lymph nodes. PSA levels tend to rise faster in cases of AIPC than they do in patients with earlier stages of the disease. This is likely due to increased tumor burden. But it may also reflect the type of cancer cell we’re dealing with. That’s a big focus of current research.*

Hormone therapy represses cancer by lowering androgen levels, particularly testosterone. How does the cancer eventually become resistant to this therapy?

BUBLEY: One of the evolving concepts in androgen-independent prostate cancer is that it’s actually not androgen independent. There’s a belief that residual androgens, even in minute amounts, may be enough to trigger the androgen receptors, parts of cells that bind to androgens to fuel cell growth.* Though LHRH agonists do a good job of limiting testosterone production by the testicles, the adrenal glands, located near the kidneys, continue to produce weaker androgens. And in the case of AIPC, the androgen receptors appear to become increasingly adept at using these adrenal androgens, along with other mechanisms and growth pathways, to maintain sufficient levels of androgen or androgen-like hormones to keep the tumor alive and promote cell growth.

TAPLIN: We believe the androgen receptor remains a player in prostate disease progression, and it may do so in a variety of ways. About a third of tumors have androgen receptor hypersensitivity, which means they can use very small levels of androgens more efficiently. About a quarter of patients have androgen receptor point mutations, which allow the androgen receptors in the cancer cells to bind with and use molecules such as other hormones, anti-androgens, or other medications to grow. As Dr. Bubley mentioned, we think that these tumor cells may latch on to weaker androgens from the adrenal glands or elsewhere, and then convert them into more potent androgens. In addition, prostate cancer cells are able to make their own androgens rather than relying on testicular or adrenal androgens. There are probably many other pathways, including genes that may work independently or with androgen receptors to fuel tumor growth, particularly in the late stages of very advanced cancers.

Can you predict who will develop AIPC?

OH: Patients who are cured with initial local therapy, such as radical prostatectomy, and don’t develop a recurrence, won’t develop AIPC. We can predict at the time of the original diagnosis which patients are likely to relapse using their Gleason scores, rate of PSA increase, and other clinical information. Patients who relapse are more likely to go on hormone therapy and subsequently to develop AIPC.

TAPLIN: PSA doubling time is a very good indicator of how quickly a cancer is progressing and a patient’s chances of survival. In patients whose PSA level doubles within three to six months after local therapy, the cancer tends to progress more quickly. So, we intervene earlier in these cases with hormone therapy.*

Considering that hormone therapy has some significant side effects and eventually leads to androgen independence, how do you decide when to prescribe it in the hormonally naïve patient?

BUBLEY: In patients who have PSA doubling times of less than six months, I tend to intervene pretty quickly because we know that these patients have cancers that are advancing at a faster rate. For patients with doubling times greater than a year, I tend to delay hormone therapy. In fact, I have some slow doublers who have PSA levels in the hundreds and do not have any signs of metastatic disease. I’ve followed them for years and have not treated them with hormone therapy.

TAPLIN: I think PSA doubling times between six months and a year present a gray area. In these cases, it really depends on the patient’s goals and acceptance of side effects. But in general, if a patient has a PSA doubling time that’s longer than a year, we take a watchful waiting approach and follow him with periodic bone and CT scans.

OH: One retrospective analysis found that for patients with rapid PSA doubling times, starting androgen-deprivation therapy before the PSA reached 10 ng/ml was associated with a better outcome than waiting until the PSA rose beyond 10 ng/ml. For patients whose doubling times were slower, early intervention with hormone therapy didn’t make as much difference in overall survival. I think it just points out that we don’t have any solid data about exactly when to start hormone therapy.

Once a patient begins hormone therapy, is there any way to tell how long it will work before resistance to the treatment occurs?

BUBLEY: Patients with a PSA nadir, or low point, less than 0.2 ng/ml are going to have a much longer response to treatment, whether using continuous hormone therapy or intermittent therapy, which involves taking breaks from the medication.* The patients who are more likely to develop androgen-independent prostate cancer, or develop it sooner, have a less robust PSA response, and that means a PSA that fails to drop below 4.0 ng/ml. And it doesn’t matter what PSA level they start at — 100 ng/ml or 10 ng/ml. The more important thing is how low the PSA goes. People with low nadirs are going to have a longer response to primary hormone therapy.

What about the rate of PSA decrease after hormone therapy is initiated?

OH: We’re actually looking at this right now. We’ve found, paradoxically, that patients who have the shortest time to their nadir PSA value after hormone therapy may actually have a worse prognosis. I’m speculating, but the patients who have a more rapid response in the beginning may have cells that are ultimately more resistant. It was an unexpected finding. The study was prompted by an observation in a patient of mine whose PSA dropped from somewhere in the 400 to 500 ng/ml range to 0.2 ng/ml in three months; six to eight months is typical. So we’re trying to study this phenomenon in other patients in our database.

Making treatment decisions

If a patient is already taking hormone therapy and his PSA starts to rise, what’s the next step?

BUBLEY: When a patient on Lupron starts to have rising PSA levels, I usually will take some baseline measurements and restage the disease. I typically start with a bone scan and CT scan, and sometimes I’ll add a chest x-ray. As I mentioned before, I’ll check serum testosterone levels to confirm the cancer is progressing in a castrate environment. If PSA levels are rising in this setting, there is a need to consider intervention because the timeline, in terms of development of metastases, is much faster.

How do you counsel patients when they first get a diagnosis of AIPC?

BUBLEY: If they had a low PSA nadir with a primary hormone therapy, then I can tell them with a bit of certainty that they’re likely to respond to a number of newer, and even older, anti-androgen and anti-adrenal therapies. This is why I often use the terms Lupron resistant or primary hormone-therapy resistant rather than androgen independent. These patients, in most cases, haven’t tried and are not necessarily resistant to a variety of second-line hormone therapies.* In many cases, although a cancer becomes Lupron resistant, it’s often not resistant to ketoconazole, diethylstilbestrol (DES), estrogen (Premarin), or high-dose bicalutamide (Casodex). All three of us, I’m sure, have had patients who’ve had a year and a half of response to ketoconazole, and another year of response to estrogen before we’ve had to think about chemotherapy. Some of my patients have had very long responses, with months turning into years. Beyond that, chemotherapy has been shown to improve survival. And then the field is replete with a number of investigational therapies.

TAPLIN: We approach it the same way at Dana-Farber. We educate patients about secondary hormonal options followed by chemotherapy options. And we try to have clinical trials available every step of the way. We have one or two clinical trials of second-line drugs under way now. And we have ongoing trials that focus on various types of chemotherapy, both first-line chemotherapy and what is termed salvage chemotherapy. For patients who have had chemotherapy, we’re looking at salvage hormone therapy, which involves the use of newer agents that are in clinical trials to try to lower testosterone further.

OH: I would just echo this point: there are many therapeutic options available, and some patients have fantastic long-term responses to additional hormone treatments, even after chemotherapy. Today there are relatively long-term survivors of androgen-independent prostate cancer, which is really a change from five or 10 years ago.

Do you continue primary hormone therapy after a patient has been diagnosed with AIPC?

BUBLEY: The clinical trials mandate that we continue it. We can’t get them on any of the clinical trials if they’re not castrate surgically or compliant with a therapy like Lupron or Zoladex. I know there are conflicting opinions about whether or not it makes any difference to continue LHRH agonists. But if the patient has been taking the drug for a number of years, he may never recover testosterone secretion, so it’s likely a moot point.

OH: There were two studies performed many years ago that formed the basis for this general practice. One of them showed that there was really no benefit to continuing an LHRH agonist in the setting of hormone-refractory disease progression. The other showed a slight survival benefit to continuing it. There are some patients who may not be a candidate for a trial and want to take a break from medication. We can check their serum testosterone levels and consider restarting the LHRH agonist if their testosterone levels do rise.

In the same vein, what do you do with patients taking anti-androgens as part of a combined hormone blockade?

BUBLEY: We always stop the anti-androgen.

OH: You have to stop the anti-androgen because patients can have what’s called an anti-androgen withdrawal response.*

When do you consider chemotherapy?

OH: Chemotherapy has been proven to provide a modest survival benefit for patients with metastatic androgen-independent prostate cancer. And typically I will initiate it for such patients. I think it’s clearly beneficial for patients who are symptomatic. But even if they’re not, we know that there’s a survival benefit.

What do you consider to be the top five chemotherapeutic agents for AIPC?

OH: My list would include docetaxel (Taxotere), paclitaxel (Taxol), mitoxantrone (Novantrone), carboplatin (Paraplatin), and vinorelbine (Navelbine).*

BUBLEY: Mine would be similar. I think we would all start with docetaxel. I’m not sure I’ve seen a lot of people respond to paclitaxel who haven’t responded to docetaxel, but I’m sure there’s a small subset who do. And Dr. Oh showed that a subset of patients who are resistant to docetaxel respond to carboplatin.

OH: Quite a number of patients have responded well to docetaxel but eventually experienced disease progression. That’s when adding a second-line chemotherapy like carboplatin may play a role. Studies show it has modest activity, but it can actually be quite dramatic in some patients. So, we’re trying to identify the subsets of patients who have the best response.

The agents mentioned so far are commercially available. What about investigational therapies?

OH: Some of the more promising investigational chemotherapy drugs for AIPC are probably, like carboplatin, in the platinum-compound family. Satraplatin is not FDA-approved, but it has been shown to be effective at limiting progression, as well as reducing pain. Though survival is the traditional endpoint for approval of chemotherapy drugs in the United States, many investigators who saw the results of the SPARC (Satraplatin and Prednisone Against Refractory Cancers) trial think that the drug is promising and that the platins may have a role.

BUBLEY: The field is also moving toward combining therapies. For example, the effect of docetaxel, although real, is modest. Can we make the effect more dramatic? Studies of docetaxel plus other drugs are looking at that question.

OH: People are optimistic about combining chemotherapy with angiogenesis inhibitors like bevacizumab (Avastin), which target tumor blood vessels. The optimism stems from some of the promising phase II data from the multicenter CALGB (Cancer and Leukemia Group B) trial. But it also comes from our knowledge of the kind of broad activity that the combination has had against other diseases, including colorectal, lung, and possibly breast cancer.

Do you treat a patient with AIPC and a rising PSA differently than one with symptomatic, metastatic disease?

OH: For those who present only with rising PSA levels, it is more of a gray area. Chemotherapy, which is the only FDA-approved treatment for androgen-independent prostate cancer, is really indicated for those with metastatic disease. That doesn’t stop us from treating patients who only have PSA progression, because we have a wealth of evidence that second-line hormone therapy may help.

TAPLIN: Another example of when treatment might differ is if there are bone metastases. Bone-building medications, such as zoledronic acid (Zometa), are specifically approved for men with metastatic disease. So, we might start zoledronic acid for patients with metastases, whereas for those with PSA progression only, we wouldn’t.

Doctors can use a genetic test called Oncotype DX to predict how patients with breast cancer might respond to a particular treatment. Is there a similar genetic test for prostate cancer patients?

OH: A lot of companies and organizations are interested in creating a similar test for prostate cancer that could accurately predict treatment outcomes in subtypes of patients. Ideally, it would help determine which treatment a patient should receive. For instance, is a patient likely to benefit most from hormone therapy, chemotherapy, or radiation after local treatment? We’ve never really understood why hormone therapy may work for many years, even a decade or longer, in some patients, and in other patients, it may work for only six months. A great deal of research is being done to better understand how genetics may help answer questions like this.

One study we were involved with found three inherited differences in genes related to androgen metabolism that accurately predicted how long hormone therapy would work. What that suggests is that the kind of machinery we’re born with probably has an impact on how well certain therapies will work. And I think we will begin seeing more genetic and molecular tests that can make these predictions.

Doctors often say that the median survival time for a patient diagnosed with AIPC is two to three years, but many men live much longer. How do you explain that?

OH: I think there are two populations with androgen-independent prostate cancers. There are patients who develop metastases quickly and progress quickly, and there are patients who have slower-growing cancers. As we discussed, many of our patients are diagnosed with cancer before the advent of metastases because we’re monitoring their PSA levels so closely. And we know that often those patients will live five years or longer after diagnosis of AIPC. The survival statistics give an average that includes both patients who progress quickly and those who progress more slowly. This makes it difficult to predict how long an individual patient will survive. Beyond averages, we don’t have a lot of good data.

What are your thoughts on the Provenge vaccine* and the FDA’s delay in approving it?

TAPLIN: Patients are definitely justified in wanting it to be available, but it hasn’t been definitively proven to work.

BUBLEY: We’d all like to see treatments like sipuleucel-T (Provenge) approved so we’d have more treatments to offer our patients. But we haven’t seen data proving that it is effective. That said, I’m enthusiastic that there may be a role for vaccines in the treatment of prostate cancer. I tend to think vaccines would be most effective in the early stage of micrometastatic disease, when the immune system has the best chance of controlling a cancer, and in combination with primary treatments such as surgery.

What about other novel therapeutic approaches? What can we expect in the future?

TAPLIN: We believe the most effective way to battle prostate cancer will likely involve a multitargeted approach. New medications are being developed to target androgens from the adrenal glands to drop testosterone levels to less than 1 ng/dl, and there are new androgen receptor antagonists in clinical trials. As the effectiveness of those treatments is evaluated, our definition of what we consider a castrate level of testosterone may change. A variety of inhibitors are also being developed to block androgen receptor function. And various other approaches, including gene therapy, are in development. In the future, we foresee combining these therapies with current approaches to maximize our ability to control tumor growth.

It has to be difficult for a patient to hear that he has AIPC. What emotional issues do patients and their families face?

TAPLIN: Being diagnosed with any kind of cancer is usually very stressful for the patient and for family and friends. Patients have to deal with the psychological aspects of the cancer diagnosis and concerns about life expectancy. They have to deal with the side effects of the hormone therapy, including sexual side effects and the newly appreciated cardiovascular risks. They have to deal with doctor’s visits. And there’s a lot of work that patients do to understand their situation and the choices that are put before them, including learning about clinical trials and deciding whether to enroll in one.

OH: I think men with prostate cancer are a unique population. Many of them are used to being providers for their family. There are often issues with communication, both with their physician and with their partner and their family. I think we probably do undertreat these issues and underutilize resources, partly because the patients themselves may not be very forthcoming about how devastating all of these issues really are. And while they look to us for treating the cancer, I think it is important to address what are sometimes considered ancillary needs, but needs that are at the core of what the patient is going through emotionally.

BUBLEY: And it’s not just the patient who’s suffering from prostate cancer. His wife or partner is suffering from prostate cancer, and his children are suffering from prostate cancer. The best we can do is the old-fashioned closed-door talk. We find out what their needs are and try to be aware of all of the resources that are available.

What kinds of support services are helpful for patients diagnosed with AIPC?

BUBLEY: Often medical centers have specialized cancer support programs. Patients who don’t have access to this type of program should talk to their physician about what resources are available in their community. A referral to a psychiatrist, psychologist, or social worker can be very helpful, and there should be no stigma attached to seeing one. This is a very difficult time. Taking an antidepressant may be in order. I also think family therapy is a good idea.

OH: Some men may be hesitant to accept antidepressants or a referral to a psychiatrist. That is partly why I think peer groups, spouses, and families are important. In my practice, it’s often the wife or significant other who tells me that there’s a problem when the patient is saying, “I’m perfectly fine.” So I think it’s important for physicians, particularly oncologists and primary care doctors, to engage patients in this conversation and help them find resources at community hospitals. What’s key is that the patient has to take steps, too. Many of them won’t address emotions as a primary issue. They’re trying to deal with the cancer on a physical level. But what’s happening psychologically is a critical part of caring for the whole patient.

Any other advice for patients?

TAPLIN: We always suggest that patients check with their doctors about taking part in a clinical trial.* There are an increasing number of new therapies, and researchers continue to explore potential new treatment targets, particularly as we begin to understand more about the biology of prostate cancer.

BUBLEY: It’s a difficult time for patients, but it’s not as bleak as it used to be. Our understanding that the disease is not resistant to all hormonal therapies is changing our outlook and expectations. I counsel patients to take part in clinical trials because it gives them more treatment options. If they get an investigational drug and it doesn’t work, they can always stop taking it and continue on a standard therapy.

OH: There are patients who feel that without solid evidence of effectiveness, they would not use second-line hormonal therapies. But there are some patients who respond to them for years and have an excellent quality of life. Although randomized trials have not been done to prove a survival benefit, we do see improved survival. The studies just haven’t been done to prove it. In that regard, I like to give patients a sense of hope that there are multiple interventions to consider, and that there are more in development.