Long-term survival of high-risk patients < 60 years of age treated by radical prostatectomy

A multi-institutional, international study has provided us with data on the 10-, 15-, and 20-year survival of a cohort of 600 relatively young, high-risk prostate cancer patients, all of whom were treated with first-line radical prostatectomy.

The paper by Dell-Oglio et al., in the journal Urologic Oncology, looked at the long-term outcomes for these 600 men, with a particular focus on their prostate cancer-specific and other-cause mortality rates. The patients were all initially treated between 1987 and 2012 (apparently at relatively well-known academic centers in Europe). To meet the criteria for study entry, the patients all had to be ≤ 59 years of age and “high risk” at diagnosis — meaning that the patients had to have at least one of the following: a PSA level of >20 ng/ml, or a clinical stage of T3 or higher, or a biopsy Gleason score of 8 to 10.

Here are the core study findings:

  • Prostate cancer-specific mortality rates were
    • 11.6 percent at 10 years
    • 15.5 percent at 15 years
    • 18.4 percent at 20 years
  • Other-cause mortality rates were
    • 5.5 percent at 10 years
    • 13.5 percent at 15 years
    • 19.3 percent at 20 years
  • The 5-year probabilities of prostate cancer-specific mortality were
    • 6.4 percent among men who had survived for 5 years post-surgery
    • 4.6 percent among men who had survived for 10 years post-surgery
    • 4.4 percent among men who had survived for 15 years post-surgery
  •  The 5-year probabilities of other-cause mortality were
    • 2.7 percent among men who had survived for 5 years post-surgery
    • 9.6 percent among men who had survived for 10 years post-surgery
    • 8.2 percent among men who had survived for 15 years post-surgery
  • The primary determinants for risk for prostate cancer-specific mortality (all P ≤ 0.03) were
    • Year of surgery
    • Pathological stage and Gleason score at time of surgery
    • Surgical margin status
    • Lymph node invasion
  • None of the above five factors were associated with risk for other-cause mortality (all P ≥ 0.09).

The overall mortality rates for this cohort of high-risk patients can be calculated as follows:

  • Overall mortality at 10 years is 11.6 + 5.5 = 17.1 percent.
  • Overall mortality at 15 years is 15.5 + 13.5 = 29.0 percent.
  • Overall mortality at 20 years is 18.4 + 19.3 = 37.7 percent.

In other words, > 60 percent of this cohort of high-risk patients initially treated by radical prostatectomy were still alive at 20 years of follow-up. Whether this outcome would be evident in a large cohort of patents treated in the community setting is an unanswered question at this time.

The authors conclude that prostate cancer is the leading cause of death among relatively young, high-risk patients treated by radical prostatectomy in the first 10 years post-surgery, but that other causes become more important after that initial 10 years. They go on to argue that, consequently, radical prostatectomy should at least be considered as the appropriate initial treatment for younger men with high-risk prostate cancer, and that strict follow-up should be mandatory for such patients over the first 10 years after surgery.

It would at least be interesting to see data from a similar cohort of high-risk men treated with (for example) a combination of external beam radiation with a brachytherapy boost — if a sufficiently large cohort of men treated in this way and followed for 20 years could be identified.

14 Responses

  1. These results are quite encouraging, especially considering that many of the patients were treated when nerve sparing and other improvements were likely not universally available for the early years, and that some important modern supportive technologies were not available. Indeed, some of the patients may not even have had initial PSA evaluations, as that test, developed in the US, was not even approved for disease monitoring by the FDA until 1988, the year after the first patients were treated in this series.

    Regarding a radiation cohort for men treated with external beam with brachytherapy boost, I am not aware of any such cohort that comes near to 600 patients, but the Dattoli cohort, which also included ADT, now has comparable median follow-up and adequate risk stratification. The latest published paper is from 2010 and details results for “321 consecutive intermediate and high-risk disease patients [who] were treated between 1/92 and 2/97 by one author (M. Dattoli) and stratified by NCCN guidelines. 157 had intermediate-risk; 164 had high-risk disease. All were treated using the combination EBRT/brachytherapy ± hormones.”

    “Nonfailing patients followup was median 10.5 years” at that point in 2010; so, as treatment in this series spanned the period from January 1992 to February 1997, follow-up in 2016 would by next month be at least 19 years for all patients and likely beyond 20 years for the vast majority. A difference between the subject paper and the Dattoli study was that the endpoint above was prostate cancer-specific mortality while the Dattoli end point was recurrence; obviously, the survivor number for the Dattoli study, had it been published, would be even better than the non-recurrence number, and obviously the mortality statistic, at least as of the 2010 publication, was available. Success was 74% (non-recurrence; so 26% recurrence and prostate cancer-specific mortality obviously less than 26%) for the high-risk patients at that median follow-up for the overall group, but, as the graphs indicate virtually no recurrences even years before this point, that figure or a figure just a little lower would likely hold for 15- and 20-year cut points. Another difference between the cohorts was that the above surgery study focused on younger men. (I have the complete paper but have not consulted it for this comment.) I hope the Dattoli team updates their results.

  2. Just a quick reference … food for thought. It might be worthwhile comparing this article with this one.

  3. What are survival rates at 5, 10, 15, and 20 years for those with no treatment? The study is otherwise discouraging to show prostate cancer mortality greater than all other mortality combined.

  4. I am very cheap so I was unwilling to pay for access to the full text of the underlying abstract. That’s why we have a Sitemaster (thank goodness). :-)

    What I would find helpful about this study is to know what post-surgery treatments the patients had, and if they had some, how were the researchers able to accurately associate the surgery with cancer-sepcific mortality? For example, if some of the patients had adjuvant or salvage radiation to the prostate bed, and/or long-term ADT, how were the researchers able to isolate the “cause” of long term cancer-specific survival to the surgery? Similarly, if the researchers kept track of what additional treatments the patients had after the surgery, the results may shed some light on what post-surgery treatments may be more effective than others over the long run, statistically speaking at least. Are those facts disclosed by the abstract (in which event it may be worth $31.50 to purchase it!)?

    If indeed there were no further treatments for any of the patients after surgery, the results are very impressive (and I should have skipped the post surgery radiation, BPLND, and ADT). Thank you.


  5. Dear Jerome:

    Please note that this article only deals with mortality rates for men with high-risk prostate cancer. The mortality rates at 5, 10, 15, and 20 years for men with high-risk disease who have no treatment until they exhibit symptomatic or metastatic disease are significantly worse than those given above.

    it is hardly surprising that, among younger men diagnosed with high-risk prostate cancer, the prostate cancer-specific mortality rates are higher than the other cause mortality rates while those men are still < 75 years of age.

  6. I am in this group and cautiously concerned. While doing well at approaching year 10, the data here show that years 10 to 20 have significantly increased mortality due to prostate cancer. I was age 44 with stage pT3b in 2006.

    This is a good study that I believe is hypothetically correct. As far as other treatments with radiation, we are probably 5 to 10 years away from seeing similar data. And one step further, about the same timeline away for those that chose combination therapies over monotherapies after diagnosis with high-risk disease. All would be data I wished I had had available when my decisions were made.

  7. Dear Richard:

    (1) The text of the abstract is freely available … the link is given above. You just need to scroll down the page a bit when you go to the link.

    (2) It is the text of the full paper that would cost you $31.50 … and I haven’t seen that either. However, …

    (3) I have absolutely no doubt whatsoever that many of these patients received one or more additional therapies, and that at least some of that information is given in the full text of the paper. Some of the patients (particularly those who exhibited biochemical progression and/or metastasis over time) almost certainly received several further treatments; others may never have needed any at all. However, …

    (4) You seem to be missing the key point of the paper, which is simply that radical surgery is a reasonable first-line option as treatment for high-risk, prostate cancer. This is the first time that I have ever seen 20-year mortality rates among a large cohort of relatively young, high-risk patients all treated with the same first-line option, and they are significantly lower that I would have expected (i.e., the survival rates are significantly higher). That’s exactly why it would be interesting to see comparable data on a similar, large cohort of younger men treated with first-line radiation therapy (almost certainly plus at least a few months of ADT). The paper was never designed to tell anyone the benefits of the second- and third-line therapies, and the numbers of patients would be too small to provide any meaningful information because this study was not designed as a trial to compare such information. It tells us one thing and one thing only … mortality rates at 10, 15, and 20 years after first-line surgery in a defined set of younger, high-risk patients.

  8. Tony,

    I think 5-10 years is optimistic for seeing comparable data for any kind of radiation. As far as I’m aware, no one is tracking mortality on a large enough sample of young high-risk men who received what is now considered curative doses of radiation. As federal patient data requirements are implemented, we may eventually be able to look at this sort of thing, but then will the findings still be relevant to the treatments then available? This is why we have to use surrogate measures (e.g., freedom from failure) in lieu of mortality data.

  9. Allen:

    If you look at the surgery study, you will see that what they did was combine data from younger patients from several series.

    I would have thought the same thing could be done within 5 to 10 years from men who received radiotherapy at a dose level of 74 Gy or higher. Remember, we are not looking for some type of “perfect study” to accumulate this type of information. We are only trying to find out if 20-year prostate cancer-specific and other-cause survival or mortality rates after first-line radiation are comparable to these surgical ones.

  10. Maybe, but I doubt it. I don’t think 74 Gy is enough either — at least 78 Gy or brachy boost therapy. As we now know, those few extra grey make a big difference. Men in radiation studies are typically older (median age is typically around 70), and very few of the studies reported prostate cancer-specific mortality (PCSM). And the high-risk component is usually very small. There was a long-term study at the University of Michigan/Schiffler that looked at EBRT vs brachytherapy boost, and they did track PCSM in high-risk men as long as 10 years. In that study, PCSM was under 7% for brachytherapy boost at 10 years, with the Kaplan-Meier curve seemingly plateaued.

    <a href=https://prostatecancerinfolink.net/2015/11/07/how-long-is-long-enough-length-of-follow-up-on-clinical-trials-for-primary-treatments/We recently looked at this kind of plateauing, at least in biochemical failure rates, that seems to occur comparatively earlier in many of the long-term higher-dose radiation studies. Recurrence rates for radical prostatectomy (at Johns Hopkins) didn’t reach a plateau until about 15 years. Presumably, most of those recurrences were among high-risk patients.

  11. Thank you, Allen, I got the boost plus 3 years ADT, and am clearly interested in reading about prostate cancer-specific statistics. This is good to know, as I am a high-risk patient. You might remember that. The two radiation runs were completed by July 2009. No biochemical failure yet. The main problem now is radiation cystitis. That turned up last October, as acute urine retention with macroscopic hemeturia. To be brief, I was sent home from the Uppsala emergency room three times, once without a catheter, twice with. Even the first was irresponsible: my bladder was emptied quickly and I was told to leave. Retention came back. I could not tolerate the catheter and, after return 3, I refused to leave. I wanted a bed, observation, and if possible treatment. I was worried about a burst bladder. Suppose I could not return to the hospital in time. I got my way after nearly 2 weeks of stalling, with a hunger strike. I got a neat cystoscopy and a CT scan of my urinary tract 2 days later. This would have taken at most three days in hospital, before neoliberal cuts were started. I fear that few in Europe can now count on a high level of public care, especially if you are over 65, or unemployed, or female. This was verified for some cardiovascular issues in Uppsala last September. Reduction of beds by reduction of investment, leading to waiting lists and waits in hospital. The aim is to cheapen services with an eye to privatisation. Why should anyone take this sitting down?

  12. Hi George and Allen,

    I wrote about the Dattoli team’s long-term results using combo brachytherapy plus EBRT with high risk patients in my first post. The complete study is available free online via this link to the abstract. While prostate cancer-specific mortality is not addressed in the Dattoli study, as I recall, it is obviously less than the total for all high-risk patients minus the non-recurrence figure as the non-recurring patients were surviving prostate cancer.

    A number of other studies involving radiation were eligible for the database of studies compiled by the Prostate Cancer Results Study Group. Some of them involved fairly lengthy follow-up of high-risk patients.

  13. Thank you, Jim. I will read all this soon.

  14. Jim,

    I don’t see any hope of drawing an apples-to-apples comparison between this paper and the Dattoli paper you cite. Dattoli simply does not provide any mortality data — one has the impression he simply ceased to follow his patients once they developed PSA failure. By looking at Dattoli’s 10-year numbers I was able to make a crude estimate that by 10 years of follow-up, 36% of his entire group had died without developing PSA failure. But there is no clue as to the fate of those who did have PSA failure.

    Likewise, I think the Prostate Cancer Results Study Group bases their reports on biochemical progression-free survival, with no information about mortality.

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 )

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: