Hazard for prostate cancer-specific mortality after radical prostatectomy


A new analysis of information from the Surveillance, Epidemiology and End Results (SEER) database suggests that the “hazard” for prostate cancer-specific mortality after radical prostatectomy is very low but that it does increase in a continuous manner for at least 15 years post-treatment.

In looking through the data in this paper by Shikanov and Eggener, it is important to understand how “hazard” is defined as a epidemiological concept, which is as follows:

  • Hazard is the rate of a specific event at a specific time that is conditional on the survival of the patient until that time.

So, for example, we can talk about the hazard of prostate cancer-specific mortality at (say) 10 years after a radical prostatectomy in a group of patients diagnosed with stage T1c disease or the hazard of a finding of evident metastasis at 5 years in a group of men diagnosed with clinical stage T2 or T3 disease initially treated with external beam radiation therapy.

The authors identified a total of 127,236 men of up to 75 years of age for whom relevant information was available in the SEER database, all of whom were treated by radical prostatectomy between 1988 and 2003:

  • 36,684 men (30.4 percent) had pathologically organ-confined disease post-surgery.
  • 41,806 men (32.9 percent) had pathologically non-organ-confined disease post-surgery.
  • 46,746 men (36.7 percent) was formally “unstaged” post-surgery because data on lymph nodes was not available, but tumor stage could be derived from other data available in the records.
  • 100,816 men (79.2 percent) had a Gleason score of 7 or less.
  • 26,420 men (20.8 percent) had a Gleason score of 8 to 10.

They were also able to classify 119,987 of these patients into one of four groups, as follows:

  • Group 1 comprised 71,106 men with a Gleason score of 7 or less who had pathologically organ-confined disease after their surgery.
  • Group 2 comprised 23,063 men with a Gleason score of 7 or less who had pathologically non-organ-confined disease after their surgery.
  • Group 3 comprised 13,660 men with a Gleason score of 8 to 10 who had pathologically organ-confined disease after their surgery.
  • Group 4 comprised 12,158 men with a Gleason score of 8 to 10 who had pathologically non-organ-confined disease after their surgery.

Careful analysis of the outcomes of the data from these men over time allowed the authors to show that:

  • Mean patient age was 62 years (range, 32 to 75 years)
  • Median follow-up for the entire cohort was 7.2 years (range, 0 to 19 years).
  • The overall prostate cancer-specific survival rates were
    • 98 percent at 5 years
    • 95 percent at 10 years
    • 90 percent at 15 years
  • For the four groups of patients identified above, the overall prostate cancer-specific survival rates at 15 years post-surgery were
    • 95 percent for Group 1
    • 89 percent for Group 2
    • 87 percent for Group 3
    • 73 percent for Group 4
  • The overall annual prostate cancer-specific mortality hazard rates were
    • 0.4 percent at 5 years post-surgery
    • 0.7 percent at 10 years post-surgery
    • 1.0 percent at 15 years post-surgery
  • For the four groups of patients identified above, the annual prostate cancer-specific mortality hazard rates were
    • 0.2 percent at 5 years and 0.5 percent at 15 years for Group 1
    • 0.5 percent at 5 years and 1.2 percent at 15 years for Group 2
    • 0.7 percent at 5 years and 1.6 percent at 15 years for Group 3
    • 1.5 percent at 5 years and 3.7 percent at 15 years for Group 4
  • The following characteristics at time of surgery were each independently associated with a worse prostate cancer-specific outcome.
    • Higher patholologic stage (hazard ratio [HR] = 2.3)
    • Higher Gleason score (HR = 3.1)
    • Age of > 65 years (HR = 1.5)
    • Black race (HR = 1.2)

Shikanov and Eggener note that the cancer-specific mortality “hazard” for many other localized malignancies (e.g., lung, breast, and colorectal cancers) actually decreases over time after a high early period of risk for failure of localized intervention. They state that the increasing hazard for prostate cancer-specific mortality after surgery is probably “due to the prolonged natural history of cancer recurrence and the often prolonged period of response after androgen deprivation.”

It should be pointed out that a series of assumptions and limitations may affect the accuracy of the estimates in this study. However, it does appear to be the report that has quantified the long-term annual hazard of prostate cancer-specific mortality post-surgery.

6 Responses

  1. Bummer … My late Dad was a Group 1 patient who succumbed to prostate cancer 14 years after a radical prostatectomy. RIP. His fate no doubt biases my own anxieties about the severity of prostate cancer, and patient prospects post-surgery.

  2. Dave:

    Alas … Individual experience and mass statistics are rarely a good match! Sorry about your Dad, and your personal anxiety is therefore utterly understandable.

  3. Does this study take into account the treatment one would get for a rising PSA after radical prostatectomy?

  4. Chris:

    This study would take account of any treatment a man had received after his surgery. However, you do need to appreciate that most men who received a radical prostatectomy back in 1988 weren’t getting the sort of immediate PSA follow-up that is customary today and were, in any case, probably significantly more advanced at the time of their diagnosis since almost no one was getting PSA tests prior to a diagnosis at that time.

  5. An interesting study. What would make it more interesting is to get comparative figures for the other therapy options, including active surveillance, to demonstrate that surgery does produce a better long-term outcome. That would be of considerable value to a man considering his options, in my opinion.

  6. Terry:

    The problem is that one couldn’t replicate this study in anyone except a set of patients who had had surgery as first-line treatment because one wouldn’t be able to know either his pathologic Gleason score or his pathologic stage. These require a post-surgical pathology report!

    :O)

    Mike

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

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

%d bloggers like this: