15-year survival and related data after first-line brachytherapy


Stone, Stock, and colleagues at the Icahn School of Medicine at Mount Sinai, New York, have been treating prostate cancer patients with brachytherapy since 1990 and (to whatever extent was possible) maintaining meticulous records throughout that time period. They have now published data on 10- and 15-year survival based on a cohort of nearly 1,700 patients.

All patients were evaluated every 6 months post-treatment. The authors defined biochemical failure according to the Phoenix criteria (a PSA level that had increased by 2 ng/ml above the PSA nadir). Patients were considered to have died of their prostate cancer only when the patient had evidence of clinical recurrence. All other patients were considered to have died of an alternate cause — even if they had a PSA-only recurrence.

Here are the key factors that they report:

  • Their cohort comprises 1,669 men, all treated between 1990 and 2007.
  • All patients were initially diagnosed with clinical stage T1, T2, or T3 localized prostate cancer.
  • The average (mean) follow-up is 10 years.
  • The average (median) age of the patients was 66 years.
  • Patients were treated in one of four possible ways, as follows
    • Radioactive seed implantation alone (n = 683/1,669, 40.9 percent), for men with low-risk disease (according to NCCN criteria)
    • Radioactive seed implantation + androgen deprivation therapy (ADT) (n = 417/1,669, 24.9 percent) or radioactive seed implantation + external beam radiation therapy (EBRT) (n = 88/1,669, 5.3 percent), for men with intermediate-risk disease (according to NCCN criteria)
    • Radioactive seed implantation + ADT + EBRT (n = 481/1,669, 28.9 percent), for men with high-risk disease(according to NCCN criteria)
  • 898/1,669 patients (53.8 percent) were treated with ADT for a median of 6 months.
  • At the time of analysis,
    • 37/1,669 men (2.2 percent) had died of prostate cancer.
    • 201/1,669 men (12 percent) had died of other causes.
  • The biochemical recurrence-free survival rates for the entire patient cohort were
    • 89.3 percent at a median of 10 years
    • 67.0 percent at a median of 15 years
  • The prostate cancer-specific survival rates for the entire patient cohort were
    •   98.1 percent at 10 years of follow-up
    • 94.1 percent at 15 years of follow-up
  • The prostate cancer-specific survival rates when broken down by risk group were
    • 96.3 percent for patients with low-risk disease
    • 97.5 percent for patients with intermediate-risk disease
    • 85.2 percent for patients with high-risk disease
  • The all-cause survival rates for the entire cohort were
    • 86.8 percent at 10 years of follow-up
    • 57.0 percent at 15 years of follow-up
  • The use of ADT did not have a positive impact on all-cause or on prostate cancer-specific survival.
    • In men with high-risk disease, increased time on ADT (> 6 months vs. ≤ 6 months) was associated with a reduction in all-cause survival (hazard ratio [HR] = 1.175, p <0.001).
    • Across the entire patient cohort, the use of ADT was associated with a reduction in all-cause survival.
      • From 89.8 to 84.2 percent at 10 years of follow-up
      • From 60.3 to 54.9 percent at 15 years of follow-up
    • Across the entire cohort, increased time on neoadjuvant ADT (> 6 months vs. ≤ 6 months) was associated with a reduction in all-cause survival.
      • In younger men (age < 66 years, p = 0.017)
      • And in older men (age ≥ 66 years, i = 0.05).
    • In men who were treated with ADT
      • The 10-year all-cause survival rate was 90.9 percent in men treated for ≤ 6 months.
      • The 10-year all-cause survival rate was 82.6 percent in men treated for > 6 months.
      • The 15-year all-cause survival rate was was 60.8 percent in men treated for ≤ 6 months.
      • The 15-year all-cause survival rate was was 56.9 percent in men treated for > 6 months.

Stone and Stock conclude that

  • Prostate brachytherapy has a favorable 15-year cause-specific survival rate, especially in patients with high-risk disease.
  • All-cause survival is lower in patients with preexisting diabetes, atrial fibrillation, and emphysema.
  • The use of ADT for > 6 months has a negative effect on all-cause survival, even in younger patients.
  • The use of ADT has no apparent beneficial effect on all-cause survival.

It is clear from this paper that brachytherapy, when carried out by highly skilled practitioners, can be seen to be associated with a high level of long-term prostate cancer-specific survival at 10 and 15 years post-treatment (comparable to the outcomes of patients treated surgically). On the other hand:

  • It provides us with no significant data on the 10- and 15-year rates of long-term side effects and complications of brachytherapy.
  • It appears to show that, with the possible exception of men with high-risk prostate cancer, there is no benefit to neoadjuvant or adjuvant ADT, and there is a very clear down-side to the use of ADT for any longer that 6 months.
  • It offers us no evidence at all that brachytherapy is associated with a long-term survival benefit (compared to simple monitoring of some type) in older patients who have low- or intermediate-risk disease.

Thus, although Stone, Stock, and their collaborators over the years are to be congratulated on compiling all the data presented in this paper, it leaves the newly diagnosed patient with only two pieces of new knowledge:

  • That adjuvant or neoadjuvant ADT is probably not a particularly good idea when associated with brachytherapy unless there are some other really compelling reasons to suggest this, and
  • That > 6 months of ADT should be avoided unless it is absolutely essential.

It is also worth noting that only 238/1,669 patients (14.3 percent of the cohort) had actually died at the time of this analysis. There is a lesson here that is also worth taking into account, which is that until we have (perhaps) 20- and 25-year follow-up data on a series of patients like this, we may be seriously lacking in knowledge about the impact of any form of treatment on prostate cancer-specific survival in men diagnosed since the widespread availability of the PSA test.

4 Responses

  1. To some of your points:

    They published some data that includes 12-year toxicity last year (see Section 4.3 of the paper), with some further detail about temporal patterns.

    To some extent, ADT seems to be a substitute for dose. The authors reported this effect in another paper. This has also been observed for HDR brachytherapy and SBRT. Biologically equivalent dose (BED) escalation is also the reason that Dr. Stone thinks it’s seldom necessary to add EBRT for his intermediate-risk patients.

    On your call for having 20 to 25 years of data before we can properly evaluate the life-extending effects of any therapy. … Well, yes that’s true, but we will never have that. That’s because the treatments, the doctors administering the treatments, and the new patient characteristics do not stay the same over time. For example, the 15-year bRFS rates for low-dose-rate brachytherapy at the University of Washington in Seattle for patients treated between 1998 and 1992 were 86%, 80% and 62% for low-, intermediate-, and high-risk patients, respectively. At the same institution, for patients treated between 1995 and 2006 the 12-year bRFS rates were 96%, 98% and 73% for low-, intermediate-, and high-risk patients, respectively. What accounted for that significant improvement in results? Probably a host of causes, including clinician experience, improved methods, dose escalation, and the characteristics of patients who were treated there. The same thing can be said for all treatments, including surgery, SBRT, IMRT, and protons. By the time we get the data to properly evaluate any treatment, the treatment will have changed so much that the data is no longer relevant.

  2. Allen:

    Thanks for bringing people’s attention to the earlier paper that does (to some extent) address long-term side effects associated with brachytherapy … even though it doesn’t give a lot of detail.

    With respect to 20 to 25 years of follow-up, we are actually starting to see such data regularly now from surgical series (with the longest two that I know of being actually 30 years; see here and here). My point here is not that the series can all be compared with exactitude, and clearly all technologies improve over time. Rather, my point is that until we have similar data from a series of patients managed over the truly long-term on active surveillance or similar forms of monitoring, we really have very little certainty that the long-term survival “benefits” of any form of immediate intervention are actually that beneficial (most especially for low-risk patients, but also for many intermediate-risk patients too).

  3. I very much agree with you that any therapy, particularly for low-risk men, has to have better cause-specific survival (CSS) than active surveillance (AS) with very long follow up. Klotz reports 86% CSS with 20 years of follow-up on his AS cohort. That compares to 83% CSS after 30 years in the Swedish study you cited. Men treated surgically had a 76% CSS after 30 years, according to the Lewinshtein abstract you cited.

    It all seems very comparable so far — so why get treated right away (which is your point, I think)?

    The problem with long-term data will haunt us here too, I fear. AS will improve as mpMRIs or CDUS become a normal part of the AS protocol, and new biochemical tests improve patient selection. But results after radical propstatectomyhave improved too over the years, as Johns Hopkins has observed. It’s a moving target.

  4. Allen:

    You are completely correct, but given Klotz’s data, the margin for improvement in prostate cancer-specific survival is already very small indeed! There is only so far for the target to be moveable, unless we all start to live until 110+ years of age, which seems devastatingly unlikely in the next 50 years.

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: