Time to prove a treatment-related prostate cancer survival benefit

For most of the past 20 years it has generally been considered that it took 10 years to show that a new form of therapy was effective and safe in the treatment of localized prostate cancer. We may need to seriously re-assess that time period.

One can probably quibble about the exact number (because all long-term, epidemiology-based survival data are based on statistical models and assumptions, and not on “real” data), but it is apparent that here in the USA the relative 15-year survival for all men diagnosed with prostate cancer is now somewhere between 90 and 95 percent. This relative survival rate is a comparison of the observed survival of all patients diagnosed with prostate cancer compared to the expected survival of men of comparable age and date of birth who are not diagnosed with prostate cancer. The absolute 15-year survival rate for all men diagnosed with prostate cancer is very different. (For a discussion of the different types of survival data commonly used by cancer statisticians, click here.)

It is worth comparing this to one of the very few, accurate sets of “real”, long-term, absolute survival data for prostate cancer, which can be extracted from the Scandinavian Prostate Cancer Group Study Number 4 (SPCG-4) trial of radical prostatectomy vs. watchful waiting in “early” prostate cancer.

The most recent update to the results of the SPCG-4 trial, into which the first patient was enrolled in 1989, were reported (by Bill-Axelson et al.) in The New England Journal of Medicine on March 6, 2014. That report was based on a cut-off date of December 31, 2012, with a total follow-up period of 23.2 years, and a median follow-up of 13 years. According to  Bill-Axelson et al.:

  • 447/695 patients (64.3 percent) had died.
    • 200/347 patients (57.6 percent) in the radical prostatectomy group.
    • 247/348 patients (71.0 percent) in the watchful waiting group.
  • 162/695 patients died of prostate cancer.
    • 63/347 patients in the radical prostatectomy group.
    • 99/348 patients in the watchful-waiting group.

In other words, the absolute overall survival rates among men diagnosed with prostate cancer in this trial at a median 13 years of follow-up were:

  • 248/695 = 35.7 percent among all patients enrolled in the trial.
  • 147/347 = 42.4 percent among the men treated by radical prostatectomy.
  • 101/348 = 29.0 percent among the men treated by watchful waiting.

The differences between these numbers are statistically significant (as were the mortality results reported by Bill Axelson et al.). But, none of these numbers are anywhere close to 90 or 95 percent.

Back at the time that the SPCG-4 trial was initiated, there was no PSA test, and the only way that most of the patients in that trial were diagnosed was because they had (a) a positive DRE, (b) cancer identified in a sample after a TURP for treatment of a urinary tract problem, or (c) other symptoms suggesting a prostate problem. In other words, most of those men did not have what is now known as clinical stage T1c disease. Today, most men being diagnosed with prostate cancer in America are being diagnosed with T1c disease (and with a Gleason score of only 3 + 3 = 6).

In this context it is worth considering the absolute survival data of patients in the Sunnybrook active surveillance cohort (most recently reported by Klotz et al. at the annual meeting of the American Urological Association meeting in Orlando this year). According to Klotz and his colleagues, among the carefully selected and monitored men in that series, the absolute overall survival rate was:

  • 693/840 = 82.5 percent (among all men enrolled in the Sunnybrook cohort since 1992)

Unfortunately, Klotz did not give a median follow-up period for the men in this cohort, but we do know that it has to be at least 8 to 10 years and that a significant number of these patients have now been followed for > 15 or > 20 years.

Thus, if a man diagnosed with localized prostate cancer of any type in America today is only 5 to 10 percent more likely to die of any cause than a man who has no prostate cancer in the ensuing 15 years, this has implications for how long we need to follow patients before we can be sure that a specific treatment is effective.

The “New” Prostate Cancer InfoLink is of the opinion that, as of now, if a new form of treatment was introduced for localized prostate cancer tomorrow, it would probably take 15 years to actually demonstrate that such a treatment was as effective as currently available treatments for men with intermediate-risk prostate cancer. It might take longer than that to prove that any treatment was more effective than expectant management in men with low-risk disease.

This has all sorts of ramifications for the introduction of new forms of treatment for localized prostate cancer. Key questions that should be looked at in the evaluation and uptake of any new form of therapy should now be highly focused on things like:

  • Is the new therapy at least as effective as (i.e., non-inferior to) some established standard?
  • Is the new therapy associated with demonstrably fewer short- and long-term side effects?
  • Can the new therapy allow patients to better maintain their pre-treatment quality of life (as compared to currently available therapies)?

We are all highly aware of the pitfalls of the currently available treatments. They are better than they were, but they aren’t good enough, and they are still being chronically over-used in low-risk men and men with limited life expectancies. New treatments for prostate cancer need to be held to a higher standard than has been customary in the past. We need to be asking how such treatment really benefit patients compared to what is already available. And we need real data (not just opinion) to clearly demonstrate any such benefits.


4 Responses

  1. Ten years studies are typically only good for the cases that are higher risk or metastatic. Of little use for low-risk disease and most intermediate cases. This was a complaint I had out of the gate with my diagnosis at age 44. Luckily my personal 8-year study is going well.

  2. For the calculation of the absolute overall survival rate for men treated with watchful waiting, you use an a denominator of 234, but it should be 348. Dividing 101 by 348 gives approximately 29.0 percent, as stated.

  3. Ooops … Typo duly corrected. Thank you.

  4. Well said! The implications of this commentary are huge for all the relatively immature treatments including proton beam therapy, SGRT, HIFU and XRT plus brachytherapy. Articles promoting any of these therapies based on PSA failure rates are grossly inadequate and mean very little other than an opportunity to promote these various therapies. Patients would be better served if doctors involved with those therapies were more honest in acknowledging it will take many more years to know if these therapies offer patients comparable or better results.

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: