30 years follow-up of men with initially untreated, localized prostate cancer


A new article in European Urology offers interesting data on the natural history of localized prostate cancer, based on a cohort of 200+ Swedish patients followed for > 30 years. It is important to note immediately that none of these men was originally diagnosed in the PSA era. They all had some form of symptomatic disease at diagnosis.

There has been a real shortage of information about the long-term impact of managing men with low-risk, early stage prostate cancer by any form of observation/monitoring as opposed to early active treatment, even though it is well understood that a significant percentage of localized, early-stage prostate cancer do have an indolent course. We also know that, within 15 years from time of diagnosis with localized prostate cancer, most deaths of men so diagnosed are not prostate cancer-specific, but are a consequence of a wide variety of other age- and health-related factors.

Popiolek et al. have now published data from a series of 223 consecutive patients, all from a regionally-define area of central Sweden, initially diagnosed with localized prostate cancer, initially managed by observation rather than any form of active intervention, and treated with androgen deprivation therapy (ADT) when symptomatic tumor progression became evident.

Here are the core findings of their study after 32 years of follow-up:

  • 220/223 men in the study (98.6 percent) had died.
  • 90/223 men (41.4 percent) had local progression of their prostate cancer over time.
  • 41/223 men (18.4 percent) had progression to evident metastatic disease over time.
  • 38/223 men (17.0 percent) died of prostate cancer.
  • Prostate cancer-specific survival decreased between 15 and 20 years of follow-up but then stabilized again after 20 years.
  • 9/223 men were initially diagnosed with a Gleason score of 8 to 10.
    • All 9 of these men (100 percent) died within 10 years of follow-up.
    • 5/9 of these men (55.5 percent) died of prostate cancer.
  • Among men with well-differentiated, non-palpable tumors at diagnosis,
    • Overall survival declined slowly for the first 20 years.
    • Overall survival declined more rapidly between 20 and 25 years of follow-up.

As we might have expected, the authors conclude that:

Although localized [prostate cancer] most often has an indolent course, local progression and distant metastasis can develop over the long term, even among patients considered low risk at diagnosis.

They are also careful to note that it is hard to know how to take practical advantage of these data and apply them in the PSA era (although we do know that PSA testing clearly adds some 7-10 years of “lead time” to the time at which a man with localized disease might be diagnosed based on clinical criteria alone).

So here are a number of other potential take-aways from this study:

  • Men initially diagnosed today with Gleason 8-10, localized disease (only 4 percent of the above series) clearly need immediate treatment unless they have a very limited life expectancy for other reasons.
  • Men initially diagnosed today with D’Amico, low-risk prostate cancer have a high probability of 15 to 20 years of prostate cancer-specific survival even if some form of monitoring is the only initial form of management.
  • Men who are initially observed but progress to metastatic disease (18.4 percent of the patients in this study) have a high probability of prostate cancer-specific mortality (55.5 percent in this study).

Until we have comparable data from a 30-year series of men rigorously monitored on an active surveillance regimen (i.e., at least 20 years from now), it will be impossible to offer any really accurate comparison between the men in this study (all diagnosed in the early 1980s and therefore in the pre-PSA era) and the men diagnosed by researchers like Klotz (in Canada) and Carter (at Johns Hopkins) and managed by careful monitoring of their PSA levels and repeat biopsies.

However, what this study does show us, once again, is that:

  • Yes, there is a long-term risk of disease-specific mortality associated with simple monitoring of localized prostate cancer.
  • That risk is relatively small for men with localized disease and Gleason scores of 7 or less.

Note that even in the above study, only 33/214 men (15.4 percent) died of prostate cancer within a 30-year follow-up period. We have not seen the full text of this paper, but if most of these men had any other indicator for D’Amico high-risk disease (e.g., a PSA level > 20 ng/ml or a clinical stage of T2b or higher at diagnosis), then it becomes arguable that the natural history of prostate cancer in the PSA era may show that almost no patients initially diagnosed with low- or intermediate-risk disease are at significant risk for prostate cancer-specific mortality within 30 years of follow-up.

Having said this, The “New” Prostate Cancer InfoLink wishes to be very clear indeed that avoiding death from prostate cancer is not the only possible reason why it might be wise to have treatment for early stage, localized disease. Avoiding the potential future symptoms of progressive and/or metastatic disease are also viable reasons for early treatment, … but only if one has a full appreciation of the risk/benefit equation on which one’s decisions are being based.

17 Responses

  1. Wow! Just, Wow!

    All of the unnecessary damage done to men’s bodies.

    All of the healthcare resources — for individuals and the healthcare system — sunk into unnecessary treatment.

    All of the fear and anxiety.

    All the diminution in quality of life …

    Just, Wow!!!

  2. Study said (below). So in the pre PSA era, with non-palpable tumors, how were they diagnosed? Why would a biopsy even have been done?

    — Among men with well-differentiated, non-palpable tumors at diagnosis
    — Overall survival declined slowly for the first 20 years.
    — Overall survival declined more rapidly between 20 and 25 years of follow-up.

  3. Dear Doug:

    Before the PSA test was available, there were still all sorts of reasons why a biopsy might have been done, including: difficulty urinating (but no indication of an enlarged prostate on DRE); prostatic pain; frequent need to urinate but no signs of an infection; etc. In addition, many men may have been diagnosed when they were given a TURP to resolve symptoms of benign prostatic hyperplasia (because there were no good drugs to relieve the symptoms of BPH in the late 1970s and early 1980s).

  4. This is the last update of the “watchful waiting” study by Johanssen and his group that started in the late 80s to better define the natural history of the disease. The 223 men in the cohort had a mean age of 72. There was some 58% disease progression and those men who progressed were treated with orchiectomy or estrogen therapy.

    Hardly “no treatment” in my view when reporting the result of watchful waiting. Wow! Over half of these men were castrated and this is supportive of watchful waiting? What if the mean age of the cohort was 55 instead of 72? Would results be the same? I doubt they would be.

    What the study shows is that given enough time early disease can in a number of cases progress and kill … and that number unfortunately is not a small number.

  5. “9/223 men were initially diagnosed with a Gleason score of 8 to 10.

    All 9 of these men (100 percent) died within 10 years of follow-up.

    5/9 of these men (55.5 percent) died of prostate cancer.”

    It’s interesting that all of these 9 men died (within 10 years) and not just from prostate cancer. Was that just co-incidence or was there some factor related to having a high Gleason score involved?

  6. Dear Ralph:

    Where are you getting data to state that “There was some 58% disease progression and those men who progressed were treated with orchiectomy or estrogen therapy” and that “Over half of these men were castrated”? The 2004 paper in the Journal of the American Medical Association clearly states that “Patients with tumor progression were hormonally treated (either by orchiectomy or estrogens) if they had symptoms.”

    I think you may be confusing the data in the current paper with the data from a total of 642 patients studied by Johanssen et al. and reported in 1997 (in which the median age at diagnosis was indeed 72 years). In those 642 patients, “prostate cancer accounted for 201 (37%) of all 541 deaths” within 15 years, but many of those patients had been diagnosed with non-localized disease and metastatic disease. The 223 men in the current study are a subset of the total of 642 patients on whom data were reported in 1997, but a subset who were all diagnosed with localized disease.

    And no one is suggesting that disease will not progress if given a long enough time to do so. But the degree of that progression will depend on the status of the patient at diagnosis.

  7. “Over half of these men were castrated”

    I have a question. Is being castrated worse than being made impotent from prostatectomy?

  8. Well … That kinda depends on how you look at the situation.

    Androgen deprivation therapy (carried out be medical or surgical castration) is customarily only given to men what are a serious risk for or actually have metastatic prostate cancer (i.e., cancer that is potentially or actually incurable because it has spread outside the prostate and the patient has a rising PSA after first-line treatment or long-term observation). Given these circumstances, the patient’s primary goal is usually to extend his life and his quality of life for as long as possible, and the loss of sexual function is a minor issue by comparison.

    In contrast, a man who has erectile dysfunction after first-line treatment of some type (surgery, radiation therapy, etc.) is (a) not strictly speaking “impotent” and (b) may not have been made as aware as he should have been of the risks associated with first-line treatment for prostate cancer. He may therefore be suffering “buyer’s remorse” when he finds he is sexually non-functional after his treatment.

    Neither situation is what any man would want to have to go through … but until we are able to find forms of treatment for prostate cancer thar can be effective at treating the cancer but have no adverse effects on sexual function at all, I am not sure how it matters whether post-treatment sexual dysfunction is better or worse than true impotence after medical or surgical castration. At least the radical prostatectomy doesn’t leave one with interminable hot flashes and gymecomastia (sore breasts and nipples) too!

  9. “I am not sure how it matters whether post-treatment sexual dysfunction is better or worse than true impotence after medical or surgical castration.”

    If it’s no better then the men who were castrated were no worse off in sexual dysfunction than if they had had surgery or radiation treatment years earlier. And their sexual dysfunction wouldn’t have started until years later than if they had received surgery or radiation treatment.

    If someone wants to highlight the seriousness of castration then they need to compare surgery or radiation treatment with possible castration some time (maybe 10 years) in the future. Castration has serious consequences but then so has surgery or radiation treatment. A valid comparison takes into account not just that castration usually causes more damage than surgery or radiation but also that castration is much less likely to be needed and that its harm causes fewer years of dysfunction than early surgery or radiation treatment.

  10. That may well be true … but I have absolutely no idea how one would conduct such a study. It’s not as though you could successfully and prospectively randomize men to early treatment or to observation. … Furthermore, if you tried to do a retrospective analysis it would be biased by choices made 1, 5, 10, or 20 years earlier. It would also be biased by the facts that: (a) many men have almost no side effects after initial recovery from radical prostatectomy, and full recovery of their sexual function; (b) a significant percentage of men treated with medical or surgical castration are elderly, have only mild side effects, and are utterly unconcerned by their loss of sexual function because they were already largely sexually inactive.

  11. Dear Mike,
    I am not confusing the Johansson article about the 223 patients. The mean age of these men was 72 years of age at diagnosis, See:
    http://tinyurl.com/ak2vll4 (page 2714).

    The point here is that to try to define the natural history of the disease a better mixture of early disease in younger men is needed. Further, if treated with hormone suppression, this distinction need to be clearly reported. The fact that survival in untreated prostate cancer is reported when a high percentage of men were castrated by orchiectomy or estrogen when progressing is confusing to say the least. Castration is not a walk in the park. …

  12. Dear Ralph:

    No one is suggesting that any form of androgen deprivation therapy is a “walk in the park.” Equally, I hope, no one is suggesting that an unnecessary radical prostatectomy (or any other form of unnecessary invasive treatment) is a good idea (or a “walk in the park”) either — especially if it leaves an otherwise healthy man either impotent or incontinent or both.

    It is certainly clear that we need data more appropriate to younger men, and some of that data will presumably come from the ongoing ProtecT study in the UK. However, the average age for diagnosis of prostate cancer in the USA is still about 65 years, I believe. The data reported by Johanssen and his colleagues in 2004 and last week continue to suggest that some 80 percent of the men in their study did not actually need or benefit from treatment for clinically significant prostate cancer during their lifetimes (and did not need or have medical or surgical castration). I would again, also point out that these were men who all had symptomatic indications of a prostate problem at the time of diagnosis. It is regrettable that the Johanssen data do not specify exactly how many men received medical or surgical castration (and based on what specific criteria), but I would again point out that they are consistent in their statement that the patients had to have progressive, symptomatic disease before medical or surgical castration was applied.

    All that I am saying is that the Johanssen data (in my opinion) only continue to add to the growing data suggesting that careful monitoring appears to be an outstanding first-line management strategy for many men diagnosed with low-risk and perhaps also intermediate-risk prostate cancer, and especially those with a life expectancy of less than 20 years. That looks to me as though an awful lot of men are currently being unnecessarily over-treated for localized prostate cancer.

    It is also quite clear from the Johanssen data that high-risk prostate cancer needs aggressive, early treatment if we are to be able to maximize a patient’s life.

  13. Dear Mike,

    I do think that, given the general knowledge that prostate cancer tends to be a slow-growing malignancy, it makes sense to establish a careful monitoring protocol such as active surveillance (AS) for men diagnosed with early stages of the disease. For that to happen we need to use PSA testing prudently until something better is developed.

    The Johansson study (in my opinion) shows that even in older men, past their life expectancy years, prostate cancer can progress, reduce quality of life and kill if the patient is not killed by other comorbidity. I am all for AS because I know that presently many men are rapidly treated after diagnosis. Many unnecessarily. After saying that, there is as much under-treatment as there is over-treatment here, but no one is talking about under-treatment here and many men continue to be diagnosed with advanced stages of the disease.

    There is no established standard for AS. The “carefully monitoring” part is not here yet. The notion that AS is currently accepted by urology is far from a reality. Nevertheless, there is activity to promote it as such in detriment of ignorant males diagnosed with the disease. The USPSTF recommendation if widely followed by gatekeepers, will result in increased mortality here. Sweden as well as other EU countries are examples of reduced use of PSA testing and men diagnosed with advanced stages of the disease.

  14. Ralph:

    I absolutely concur that the situation is far from perfect!

  15. Mike,

    “We have not seen the full text of this paper, but if most of these men had any other indicator for D’Amico high-risk disease (e.g., a PSA level > 20 ng/ml or a clinical stage of T2b or higher at diagnosis), then it becomes arguable that the natural history of prostate cancer in the PSA era may show that almost no patients initially diagnosed with low- or intermediate-risk disease are at significant risk for prostate cancer-specific mortality within 30 years of follow-up”

    I would pay money to see this paper to see if your theroy is true. I was diagnosed at 42 and it would interesting to see this.

  16. Chris:

    See Table 1 on page 2714 of the full text of the paper by Johanssen et al. giving data from this study at 20 years of follow-up. Note that the staging system being used is the TNM system as from 1978, so it is hard to know exactly how accurately these data can be compared to the modern TNM staging system.

  17. Chris,

    When diagnosed at an early age you did what the authors of the study recommended:”These findings would support early radical treatment, notably among patients with an estimated life expectancy exceeding 15 years.”

    There are lessons to learn about this study. The assumption that progression is without consequence is negated by the study; The authors reported: “Postponement of death is not the only treatment objective because local progression may create substantial suffering. Indeed, many of our patients experienced symptomatic local growth without generalized disease (Table 1), requiring treatment with estrogens or orchidectomy.” This lesson continues to be widely ignored …

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 )

Google+ photo

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

Connecting to %s

Follow

Get every new post delivered to your Inbox.

Join 1,151 other followers