Metastasis and mortality after treatment for localized prostate cancer


A restrospective review of data from Memorial Sloan-Kettering Cancer Center has provided us with some interesting new data on the progression of prostate cancer over time.

In a paper initially available on line and now published in the Journal of Clinical Oncology, Zelefsky et al. have reported data on a total of 2,380 patients initially diagnosed with localized prostate cancer and subsequently treated with either an open radical prostatectomy (RP) by one of two experienced surgeons or with intensity modulated radiotherapy (IMRT) at a dose level of not less than 81 Gy. Unfortunately, the dates of initial treatment of the patients and the lengths of follow-up are not given in the abstract or in the editorial comment on this paper by Evans available on the UroToday web site. We would guess, however, that most of these patients would have received their initial treatment at some time between about 1996 and 2005.

The study was designed to evaluate the effect of first-line treatment with RP or IMRT on risk for distant metastases and risk for prostate cancer-specific death in patients with localized prostate cancer treated at a single specialized cancer center. All patients included in this retrospective analysis had in itially been diagnosed with clinical stage T1c-T3b disease.

Patient-specific base data from this study can be summarized as follows:

  • 1,318 patients were initially treated with RP.
  • 1,062 patients were initially treated with IMRT.
  • 79/1,318 patients (6 percent) initially treated with RP received immediate adjuvant or salvage radiation therapy.
  • 597/1,062 patients (56 percent) initially terated with IMRT also receive 3 to 6 months of neoadjuvant androgen deprivation therapy (ADT).
  • 107/141 patients with biochemical recurrence after RP went on to receive true salvage radiation therapy with or without ADT or ADT and chemotherapy
  • 48 men were given ADT or chemotherapy without external beam radiation.
  • 88/207 men with a biochemical recurrence after IMRT went on to receive salvage ADT.
  • Salvage therapy for patients initially treated with RP was delivered at  median of 13 months after biochemical failure.
  • Salvage therapy for patients initially treated with IMRT was delivered at a median of 69 months after biochemical failure.

The results of the trial showed the following:

  • The men treated with IMRT (as might be expected) were older than those treated with RP and also had higher PSA levels, clinical stages, and biopsy Gleason scores.
  • The probabilities of 5-year biochemical recurrence-free survival predicted by the Kattan nomogram was 84 percent for RP patients and 80 percent for IMRT patients.
  • 21/1,318 patients in the RP group (1.6 percent) and 48/1,062 patients in the IMRT  group (4.5 percent) developed distant metastases.
  • The 8-year probability of freedom from metastatic progression was 97 percent for RP patients and 93 percent for IMRT patients.
  • The adjusted absolute difference in 8-year metastasis-free survival rates was 1.9 percent for men with low-risk disease, 3.3 percent for men with intermediate-risk disease, and 7.8 percent for men with high-risk disease (favoring RP in all cases).
  • 8/1,318 patients in the RP group (0.6 percent) and 22/1062 patients in the IMRT group (2.1 percent) died.
  • The corresponding 8-year probabilities of prostate cancer-specific survival were 98.6 percent for RP patients and 95.3 percent for IMRT patients.
  • The 8-year Kaplan-Meier probability of prostate cancer-specific death was 3.8 percent among the RP patients as compared to 9.5 percent among the IMRT patients.
  • After adjustment for case mix, surgery was associated with a reduced risk of metastasis (hazard ratio [HR] =, 0.35) and a similarly reduced risk of prostate cancer–specific mortality (HR =0.32).

Zelefsy and his colleagues reach the following conclusions:

  • Metastatic progression is relatively infrequent in men with low-risk prostate cancer that is treated with RP or IMRT.
  • RP patients with higher-risk disease had a lower risk of metastatic progression and prostate cancer-specific death than IMRT patients.
  • Differences in the use and timing of salvage therapy may significantly impact these results.

7 Responses

  1. I had an RP in August 2005 after diagnosis with a PSA of 7.6 and a Gleason score of 3 + 4 = 7. A MRI showed it was localized. My surgeon did not get pathology of the dissected prostate, so I don’t know if it is aggressive cancer.

    After surgery my PSA was 0.05. It rose slowly to 0.4 by June 2009. I then had 32 sessions of external beam radiation as salvage treatment in July/August 2009. In December 2009 my PSA was down to 0.16. I have been told to get my next PSA in June 2010.

    What is my likely life expectancy now that I am 61 years 9 months?

    And do you think I have aggressive cancer?

  2. Dear Mr. Hooper:

    I think that you should join the social network associated with this site and ask for people’s input on that network, where there will be many people willing to share their knowledge and experiences with you.

  3. Good advice Mike! This is the first time I have heard of no pathology result on a RP gland. I wonder how common is this?

    Mr. Hooper, by all means bring your discussion to the social network.

  4. Mike, I also believe that these cases have to be more recent. Was IMRT available in 1996? And that dose at least 81 Gy is much higher than dispensed in the mid-90s

  5. Ralph:

    Early IMRT systems were available by 1996, if memory servces, and if anyone had such a thing it would likely have been at Harvard. Don’t get me wrong, I suspect the vast majority of these patients would have been treated later rather than earlier.

  6. These are very interesting data. We just made the decision for my husband (46 years old, diagnosed with T3, Gleason 10, PSA 19.5) not to have a radical prostatectomy but neoadjuvent hormone therapy and IMRT, followed by 6 courses of (prophylactic) Taxotere chemotherapy. Two out of 3 groups of doctors said he was not a surgical candidate. Only Duke University sided with radical prostatectomy, but could not back it up with data, saying “The pendulum is swinging back towards RP.” Now I am wondering if we have made the right decision or not. The abstract does not give enough detail on the high-risk patients; did they mostly do monotherapies (i.e., radiotherapy that was not combined with long term hormonal therapy)? This would be interesting since the Bolla study clearly shows that RT + hormonal therapy is more efficacious than just RT. What are your thoughts? Is this some groundbreaking study that “proves” RP is the treatment of choice also for high-risk patients?

  7. Dear Kirsten:

    First and foremost, no, this is absolutely not ‘some groundbreaking study that “proves” RP is the treatment of choice also for high-risk patients.’ It is a retrospective analysis of data from a single institution (normally considered to be “level 4” evidence.

    It is always difficult to know the best way to treat someone with your husband’s high-risk clinical characteristics. Some institutions would argue that surgery combined with some form of adjuvant or neoadjuvant therapy would be the best option; others would argue in favor of an option similar to the one you had already opted for. This is no “right” answer to this dilemma, and we just do not know, for an individual patient, what might be best.

    I fully appreciate how frustrating this is for the individual patient. I wish we were able to tell you and your husband with certainty that one option was better than any other, but as you have already discovered from talking to the doctors directly, there simply aren’t sufficient data to be compelling one way or the other.

Leave a comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.