Docetaxel in patients with androgenrefractory disease

Microtubule complexes have been identified as particularly important targets in prostate cancer systemic therapy. A series of agents — the vincas, taxanes and estramustine — have activity at that level. The taxanes — docetaxel and paclitaxel — target those complexes as their primary mechanism of action and have been widely investigated.

Dr Petrylak at Columbia conducted a Phase I/II trial of docetaxel and estramustine. This study demonstrated an intriguing median survival of approximately 23 months, which caught people’s attention.

We all recognize that prostate cancer is not breast or lung cancer, so objective response rates have to be considered carefully; prostate cancer is a disease that primarily involves bone only, with a smaller subset of patients with measurable disease. Survival data, even from a hypothesis-generating study like Dr Petrylak’s, was important, so the Southwest Oncology Group launched SWOG trial 9916 to investigate whether an agent or combination exists that would provide a meaningful difference in survival (Petrylak 2004a and b).

A related trial was TAX-327, which compared weekly versus every three-week docetaxel versus the gold standard, mitoxantrone/prednisone (Eisenberger 2004; Tannock 2004). The survival data from both trials demonstrated a two to two-and-a-half month median improvement in survival, and the hazard ratios indicated a 20 to 24 percent reduction in prostate cancer death on the docetaxel arm.

That’s a modest advance, but we need to put this in perspective. Ten years ago, chemotherapy was felt to be essentially of no utility in prostate cancer, and now we have objective response rates enough to provide a survival advantage. We all understand that it is not the “end all, be all,” because it’s still modest, but it is an important first step, and it changes how we think about the management of advanced disease.

Use of docetaxel in patients with metastatic disease

Although I was obviously delighted to see the results of SWOG-S9916 and TAX- 327, in effect, it’s created many more questions than were answered. The majority of the patients enrolled in the two trials had androgen-independent metastatic disease, and many, but not all, were symptomatic. Until that data were available, chemotherapy in a noninvestigational setting was used to palliate patients; therefore, most patients, at least theoretically, were treated when they had disease-related symptoms.

The question now is: Does the patient who has asymptomatic metastatic disease need to be treated at that time, or later? That’s a critical question to which we don’t know the answer. In my practice, for asymptomatic patients with lowvolume disease, I have a discussion about what we know about the trials. As an academician, I have clinical research opportunities for some of these patients and certainly would steer them in that direction. When a patient is not interested in participating in a clinical trial, I review the data with them and try to arrive at a reasonable decision based on their individual perspectives.

Early versus deferred treatment of androgen deprivation therapy

Earlier versus deferred hormonal therapy is a major breaking point in the GU community — particularly among the zealous believers in early androgen deprivation and the more nihilistic among us. In my own practice, because we see a large number of patients with biochemical failure, I have alternative, immunomodulatory investigational options. Putting that aside, PSA doubling time is increasingly useful to predict which patients are more likely to develop systemic progression in the hormone-naïve setting.

I discuss the controversies of early androgen deprivation with patients and discuss why my colleagues are advocates of earlier therapy. When the patient asks me, ultimately, where I stand on the matter, I tell them that I respect the toxicity profile of androgen deprivation therapy. For a long time we have undersold the impact of androgen deprivation on quality of life.

I tend to advocate early androgen deprivation therapy for the motivated patient with a shortening PSA doubling time, which sometimes occurs after a relative period of stability. Now, is that correct? I don’t know the answer to that question, but in my practice, that’s the situation in which I talk to patients in a more proactive way about earlier hormonal therapy.

Use of PSA as an endpoint in clinical trials

With each passing year, the number of patients with locally advanced prostate cancer — who are perhaps destined to do poorly relatively early — continues to decline as we detect disease earlier. This impacts our ability to perform adjuvant studies of chemotherapy. Currently, the FDA would not accept PSA failure as a clinical endpoint, so we have to wait for clinical progression or death. The FDA is actively considering these issues, and at least one forum was held last fall at the FDA and another one is planned. Changes may be occurring in the agency’s attitude toward PSA as an endpoint, but as of today, it’s a dilemma. If we can only perform one study a decade, it will be a long time before we can answer the question about adjuvant chemotherapy in the treatment of prostate cancer. As a clinical trial endpoint, PSA remains problematic in some settings. In patients with biochemical failure only, using PSA failure as a parameter of response remains unproven; however, in the adjuvant setting, I think most of us who take care of these patients would clearly accept time to PSA failure as an endpoint in patients undergoing radical prostatectomy — albeit not the only endpoint.

Of course, reasonable assurances must be made to ensure that the PSA failures are real and not simply low levels of detectable PSA in patients who are not destined to progress. That’s the optimal use of PSA in how we manage patients today, and it would be problematic to not use PSA failure as at least an intermediate endpoint.

Clearly, in studies of hormonal therapy, PSA failure would not be a useful endpoint. Biologic or targeted therapies are also potentially problematic unless we understand what these drugs do to PSA expression at the cellular level. With chemotherapy, we increasingly have reason to believe it would have validity in the postprostatectomy setting.

Time to delay of PSA failure is probably a good surrogate to activity. That’s not to say you should end the trial based on that endpoint and not collect other data, but I believe it’s an endpoint that will have some value and allow us to begin testing agents in the adjuvant setting without having to expose patients to Phase III investigations. It would allow us to perform hypothesis-generation studies and select agents that make rational sense based on some of these early endpoints, and then move on to formal Phase III studies.

In the metastatic setting, Dr Crawford presented data at ASCO 2004 that were based on the preliminary analysis of the SWOG randomized trial S9916. These data suggested that a three-month change in PSA was, in fact, a surrogate for survival in the androgen-independent setting (Crawford 2004). Is it the same in the hormone-naïve environment? I don’t know, and that’s an important question.

Select publications

Dr Dreicer is a Professor of Medicine at the Cleveland Clinic Lerner College of Medicine, Director of Genitourinary Medical Oncology and Associate Director of Experimental Therapeutics in the Departments of Hematology/Oncology and the Glickman Urological Institute at the Cleveland Clinic Foundation in Cleveland, Ohio.