You are here: Home: PCU Think Tank 1 | 2006: Section 2

Select Excerpts from the Discussion

CD 1, Tracks 10-11

DR LOVE: Could you review your work at Johns Hopkins on the prognosis of men with PSA-only relapse?

DR FREEDLAND: The purpose of our study (Freedland 2005) was to examine risk factors for prostate cancer death with a patient population that was largely similar to that in Pound’s study (Pound 1999). We studied more than 5,000 cases with nearly 1,000 recurrences, and 379 men had PSA doubling time data. Of those 379, 66 died of prostate cancer and 15 died of nonprostate cancer causes. So about three quarters of the deaths in our study on an actuarial basis were due to prostate cancer. It’s a young, healthy Johns Hopkins cohort in which we’re not seeing a lot of competing mortalities, so we obtained a long natural history of prostate cancer. Now we are up to 10 years of follow-up.

Overall, we confirmed the long natural history of prostate cancer we saw in the Pound series (Pound 1999), which was, on average, 13 years from recurrence to death. In our series it was more like 16 years.

When we assessed risk factors for prostate cancer death, we found exactly the same risk factors that predict metastatic disease and death from prostate cancer: PSA doubling time, early versus late recurrences and Gleason score.

For doubling time, the hazard ratio of 0.86 indicates that a one-month change in doubling time reduces the risk of death by 14 percent.

Absolutely capturing the real doubling time involves a lot of logistical issues and difficulties such that, at the end of the day, we wanted to regard it as a categorical variable. So we considered various cut points and eventually identified four categories. A doubling time greater than 15 months presented the lowest risk, and a doubling time less than three months presented a risk of dying from prostate cancer 27 times higher — these patients are at extremely high risk. With a PSA doubling time of three to nine months, patients are nine times more likely to die, and even with a PSA doubling time of nine to 15 months, patients are also at increased risk.

What we’ve shown is that men with rapid doubling times, high Gleason scores and early recurrence are at increased risk of prostate cancer death. We’ve generated tables that can predict that. Men who experience late recurrence generally have excellent survival, and time to recurrence, in the absence of knowing the doubling time, is an important prognostic feature.

CD 1, Track 21

DR LOVE: Dave, do you refer patients with hormone-refractory PSA-only disease to a medical oncologist to consider nonprotocol chemotherapy?

DR CRAWFORD: I find the urologists are all pumped up about this, but the medical oncologists are hesitant to treat them with chemotherapy off protocol.

DR PETRYLAK: I believe that after carefully discussing with the patient the risks and benefits of chemotherapy, in some cases, it’s totally appropriate to treat patients in that setting. We really don’t know what the proper sequence is between secondary hormonal manipulations and chemotherapy, and now immune therapy is coming on board, but I do believe it’s appropriate to consider it.

DR FREEDLAND: There was a study published in the Journal of Clinical Oncology that showed chemotherapy is active for hormone-naïve disease (Goodin 2005; Hussain 2005).

However, chemotherapy relies on dividing, growing cells, and hormonal therapy puts most of the cells into a G0 arrest, so there may not be a benefit in combining the two at that early stage of the game. So off protocol, I’m hesitant to do it.

DR GOMELLA: The medical oncologists I work with at Jefferson basically try multiple alternative therapies before they actually commit somebody to chemotherapy.

When we send them a patient with hormone-refractory disease, their favorite thing to use is high-dose bicalutamide. Our medical oncologists are not directly jumping to a docetaxel-based therapy before they burn a lot of other bridges in between.

DR KLOTZ: Are we talking about hormone-naïve or hormone-refractory disease? I don’t think anyone is advocating chemotherapy alone, without hormone therapy, for hormone-naïve disease.

DR LOVE: No, we are talking about adding it in.

DR SCHER: This is part of the open question. We don’t know how to optimally combine hormones and chemotherapy. As pointed out, if the cells are not dividing, we don’t know that they’re sensitive to specific types of agents.

DR KLOTZ: Off protocol, no one is advocating chemotherapy without hormone therapy for anything, except maybe in the neoadjuvant setting.

CD 2, Tracks 2, 4-6

DR LOVE: Paul, can you review the data on combined androgen blockade (CAB)?

DR SCHELLHAMMER: In 2000, the Prostate Cancer Trialists’ Collaborative Group did a meta-analysis and found a 2.9 percent five-year survival benefit in patients receiving CAB with flutamide and nilutamide (Prostate Cancer Trialists’ Collaborative Group 2000). In a trial with more than 800 patients with metastatic disease that compared two different CAB constructs — bicalutamide versus flutamide — there was a trend to benefit in survival for patients on the bicalutamide arm.

Currently, there is a trial in Japan comparing an LHRH agonist with or without bicalutamide. The preliminary data show an advantage in PSA progression and progression to hormone-refractory disease for adding bicalutamide (Akaza 2004).

DR LOVE: What about the issue of cost?

DR SCHELLHAMMER: An analysis by Dave Penson of CAB therapy with bicalutamide shows an overall quality-adjusted and overall per-year-of-life gain that is reasonable with bicalutamide for MAB. It’s within the construct of what we accept from renal dialysis in patients with renal failure (Ramsey 2005).

DR CRAWFORD: I believe CAB is the gold standard. It has plenty of data from large randomized Phase III trials to support it. None of the studies were negative, and I believe there is a reason to give it to patients with metastatic disease.

DR KEANE: The older studies dealt with patients with more advanced disease. If you examine the Dijkman and Janknegt studies, patients whose PSAs nadired down to less than 0.5 had a seven-month statistically significant survival advantage in heavy-volume metastatic disease (Dijkman 1995, 1997; Janknegt 1993).

We probably underestimated the effect of CAB and mistreated most of the patients. If you have a patient you plan to treat with an LHRH agonist, I would recommend CAB.

DR CHODAK: Not to be critical here, but the message that came out of Hopkins in response to CAB and the Peto analysis changed the attitude of many people. They feel the benefit is not worth the expense, and that is unfortunate because the benefit is greater than it is with docetaxel, and everyone is excited about docetaxel.

Now the truth is, we don’t have a lot of patients who qualify based on metastatic disease. However, we do have a group of patients who are doubling in less than three months and dying within five to six years. They must have metastatic disease to be doing that.

Maximum blockade gives you the best chance for survival when you have metastatic disease, and treating earlier metastatic disease is better. We have a parameter that is identifying people who are acting like they have metastases, but we can’t prove it, other than this PSA doubling time. Shouldn’t we be aggressively using maximum blockade in that setting?

DR SCHER: I believe those patients need systemic therapy, and the standard for those patients is maximum blockade.

DR ZIPPE: I believe CAB got a bad name because we were throwing it at everyone, including patients whom the urologists knew had 10 years to live. I believe now that we can stratify patients and show the five-year mortality rates, CAB will be well accepted for a subset of patients.

DR FREEDLAND: I believe part of the confusion comes when you evaluate the data on cyproterone acetate, which actually worsened survival. Some of the early meta-analyses that came out in The Lancet in the late 1990s actually showed no statistical benefit. That is the era when I was coming up in urology. I remember as a medical student saying, “Okay, combined androgen blockade does not add anything.”

Then some of the reanalyses, taking the cyproterone acetate data, which is a steroidal anti-androgen — rather than the nonsteroidal bicalutamide — out of the equation, you start to see modest benefit, and I believe that clouds the picture. I am not sure the entire community of urologists is necessarily aware of the data if they just received the original Lancet paper from 1997 and read, “Overall, there is no benefit.”

CD 2, Tracks 8-9

DR LOVE: Laurie, would you describe the role of intermittent androgen deprivation?

DR KLOTZ: The role of intermittent therapy has been clouded by the tremendous enthusiasm for early androgen deprivation therapy in patients with biochemical failure. I believe the question of early versus delayed therapy is still unanswered.

Studer reported at the AUA on a European trial of early versus delayed therapy in 900 patients (Studer 2005). They found no difference in time to progression, and the only patients who benefited from early therapy were those who had a PSA doubling time of less than a year.

There is a paradoxical dual effect of androgens on prostate epithelium. The direct effect is differentiation and, at some concentration, cell cycle arrest, whereas the indirect effect on the stroma, mediated by a number of growth factors, is proliferation. Thus you have two phenomena, differentiation and arrest induction and proliferation, so the idea that somehow testosterone is just a stimulator of prostate epithelial growth is not correct.

A trial initiated by Nick Berkoffsky 15 years ago shows us that re-exposure to testosterone may have an antiproliferative or differentiation-inducing effect. This trial of approximately 100 patients who received eight months of therapy demonstrated that the PSA lagged behind the testosterone; there’s a window here that is predictive in terms of time to androgen-independent progression (Opfermann 2006).

The median time off therapy — and these were biochemical failures — was eight months, after an eight-month induction, and they found the off-treatment interval gets shorter with serial cycles.

They also observed that the baseline PSA was a major predictor of the off-treatment interval. Interestingly, there was consistently about a 95 percent reduction in the PSA regardless of the baseline.

The quality of life tended to return to baseline in the off-treatment interval in all domains. Six months after discontinuation of therapy, the vast majority of patients had at least 50 percent recovery of testosterone, although about half still were less than normal. Few patients experienced prolonged castrate levels.

The issue of the equivalence in terms of time to androgen independence has been a major point of controversy, and currently about five international trials are examining this.

Dave Crawford is leading a trial of intermittent therapy in metastatic disease, and we are leading one in Canada in biochemical failure after radiation therapy. None of these are mature, except for the Portuguese trial that has just been reported (Calais da Silva 2006).

In the Portuguese trial, 630 patients with locally advanced or metastatic disease were randomly assigned to intermittent versus continuous therapy with an LHRH analog plus cyproterone acetate, following three months of induction therapy. The Europeans tend to use three months of induction therapy, whereas in the United States and Canada it tends to be more like eight months to a year.

Although a lot of studies have reported preliminary results showing no difference in mortality, the event rate is low. However, in this trial the majority of patients have died, and prostate cancer was thought to be the cause in approximately two thirds of the deaths in both arms.

Despite stratification by PSA at randomization or metastatic status, there was no difference between therapies and absolutely no difference in survival.

I believe this is excellent support for the concept that the time to androgen-independent progression and prostate cancer survival is not significantly different between the two approaches.

CD 2, Track 12

DR ZIPPE: Laurie, I’m trying to determine where intermittent therapy might fit into the way we practice now. I used intermittent therapy with patients at lower risk. If you subset your data, did you find that the survival advantage was the same? Was intermittent therapy better for the patients with low-risk disease or worse for those with high-risk disease?

DR KLOTZ: No, the baseline parameters did not predict for a differential response with intermittent versus continuous therapy. Minor differences did arise, but they did not achieve any significance.

DR KEANE: The type of LHRH agonist utilized may also have an effect. A paper from last year showed that you can have up to seven months during which the LHRH — for instance, leuprolide — remains in the system, versus a depo preparation such as Viadur® or Vantas®. One of those is gone in six weeks. So part of the length of time that you are seeing an effect may be a function of the LHRH agonist that you are using and the delay in washout of those agents.

Secondly, it is interesting that we all multiply our PSAs by two when patients are on a 5-alpha reductase inhibitor. The division of the difference between the responses was almost exactly two. You had 15 months versus 30 months in terms of duration of treatment. That may have been a function of the PSA. We don’t know, but it is certainly interesting to see that it’s exactly half.

Select publications

Home