Non-prostate cancer-specific mortality rates in a cohort of > 3,000 patients

The issue of how long a man who is newly diagnosed with localized prostate cancer is likely to live before he dies a natural death from other causes is a crucial element in the decision whether to undergo treatment (and how aggressive that treatment may need to be).

While we have a long way to go before we will be able to answer this critical question with accuracy in individual cases, a recently published paper by Daskivich et al. (in the Annals of Internal Medicine) does seem to offer us some additional data and supporting guidance about risk and life expectancy based on factors like the patient’s age, co-morbidities, and tumor factors. We had previously mentioned a Reuters report about this paper in a post dated May 24 this year.

Daskivich et al. used data from the nationally representative, prospective, population-based Prostate Cancer Outcomes Study (PCOS) cohort of 3,183 men with non-metastatic prostate cancer at the time of diagnosis who had been followed for 14 years to assess overall and prostate cancer-specific mortality rates and then correlate these data back to factors known at the time of diagnosis and initial decisions about treatment or observation.

All patient co-morbidities were self-reported by patients at the time of initial diagnosis (i.e., at baseline) and then scored by the researchers on the basis of a count of 12 major co-morbid conditions.

The authors were able to show the following:

  • The 14-year, cumulative non-prostate cancer-specific mortality rates were
    • 24 percent for men with no co-morbid conditions
    • 33 percent for men with 1 co-morbid condition
    • 46 percent for men with 2 co-morbid conditions
    • 57 percent for men with ≥ 3 co-morbid conditions

In other words, about half the men with ≥ 2 co-morbidities died of something other than prostate cancer.

In addition, Daskivitch et al. also show that:

  • Among the men with ≥ 3 co-morbid conditions, 10-year non-prostate cancer-specific mortality rates were
    • 26 percent for those aged ≤ 60 years at diagnosis
    • 40 percent for those aged between 61 and 74 years at diagnosis
    • 71 percent for those aged ≥ 75 years at diagnosis
  • Prostate cancer-specific mortality at 14 years of follow-up was
    • 3 percent in men with low-risk disease
    • 7 percent in men with intermediate-risk disease
    • 18 percent in men with high-risk disease
  • Prostate cancer-specific mortality rates did not vary by number of co-morbid conditions (10 to 11 percent in all groups).

The authors’ conclusion will come as no surprise to regular readers of this blog:

Older men with multiple major co-morbid conditions are at high risk for other-cause mortality within 10 years of diagnosis and should consider this information when deciding between conservative management and aggressive treatment for low- or intermediate-risk prostate cancer.

These prospectively collected data, based on nearly 15-year follow-up of a large cohort of patients with clinically localized prostate cancer do re-emphasize some very simple facts:

  • Men with low-risk disease are at very low-risk of prostate cancer-specific mortality within 15 years of diagnosis.
  • Even men with intermediate-risk disease are at low risk (<10 percent) of prostate cancer-specific mortality within 15 years of diagnosis
  • Men of 75 years and older with low- or intermediate-risk prostate cancer and ≥ 2 co-morbid conditions are going to be at almost no risk of prostate cancer-specific mortality (because their risk of dying from something else within 10 years is so high).

6 Responses

  1. How uniformly defined is the phrase “cause of death”? And how is its uniformity determined?

    Example 1: If a man undergoing a routine prostatectomy for T2a GS=7 prostate cancer develops a hospital-acquired infection and dies of it, what is the cause of death?

    Example 2: If a man has exhausted all treatment possibilities for his metastatic Stage IV prostate cancer and dies of an self-administered overdose of barbiturates, what is the cause of death?

    Example 3: If a man in hospice care awaiting death from terminal prostate cancer catches pneumonia and dies of fluid in the lungs rather than multiple organ failure, what is the cause of death?

    When there are multiple causes of death, how are the statistics adjusted to amplify the signal and diminish the noise?

  2. Dear Paul:

    If you are asking me about uniformity in this specific study, in which the researchers had full access to all patient charts and related information, I think we can say it was probably highly uniform and the causes of death that you refer to would and should have been Example 1: hospital-acquired infection; Example 2: suicide induced by barbiturate poisoning; and Example 3: sequelae of pneumonia. In each case, prostate cancer would probably be clearly indicated as a complicating factor that has potentially causative implications.

    However, the real world is not as well organized. You could ask the same question about almost every death that occurs from anything except the most clear-cut of causes. Most of us actually die with a bunch of differing issues that may or may not all be associated with a singular underlying factor. For example, I don’t think that “obesity” is an accepted cause of death, but my bet is that is the singular underlying factor leading to the death of all sorts of people whose cause of death is listed in other ways (heart disease, diabetes, respiratory failure, you name it).

  3. Dear PaulC:

    Your comments are along my thought process … if you are inside the 10-year not-cured state of prostate cancer, and pass away due to something else or related to the effects of treatment for prostate cancer, is one classified as passing from prostate cancer or the other factor.

    Prostate cancer or prostate cancer treatment a contributing factor, possibly. … Think Whitney Houston, according to TV as my source, lifestyle before and then the bath … hmmm, relax and slip under … hmmm.

    The pigeon hole effect of cause of death arises,, my mother , 93, passed in her sleep,, naturally,, but, and I do not know, what it actually was listed as,, old age ?? but in the States, as I understand , there would have been a medical condition listed.

    So if PC has been determined, and treated, and one goes to sleep,, will the person be pigeon holed PC or another cause,, true if the person had heart problems, diabetes,, MS, or lung problems, kidney problems,,,they all can contribute to the cause,,, the problem that I see, is that the PC numbers are low,,, unless the person goes the full nine yards and passes , then will his passing be put into the PC pigeon hole.

    A question that can be talked about till the cows come home and the coffee maker is drained,,, no easy answer..


  4. Even though “initial treatment” was one of the study “measurements”, I assume none of the men in the study had any treatment for prostate cancer after diagnosis, and they were followed either until they died or for 14 years, whichever came first. I don’t know the number of men in the study who were in the high-risk category, but if only 18% of them died after 14 years of follow-up with no treatment, then the conclusion to be drawn is that even men with high-risk disease have an 82% chance of surviving for at least 14 years after diagnosis of non-metastatic prostate cancer, even if they do nothing about the prostate cancer. That’s an incredible statistic evidencing the indolent character of the natural course of prostate cancer. Maybe the AUA should amend its guidelines again and recommend no PSA testing or treatment ever for prostate cancer, and it should focus its resources on cancers which may cause an immediate serious risk to health if left untreated.

    Richard Stanton

  5. Dear Richard:

    Unfortunately there is a serious flaw in your baseline premise about the data from the Prostate Cancer Outcomes Study. In fact, at least 1,100 of the men in this study received first-line surgery and at least another 491 received first-line external beam radiation therapy as treatment for localized disease. You also need to appreciate that this was not a prospective study. It was a sophisticated data analysis based on data from the SEER-Medicare database. For more details, please click here and click here.

    Thus, it is entirely inappropriate to think that no one in this study got first-line treatment.

  6. Sitemaster:

    Well, that’s why I used the word “incredible” because I couldn’t believe that was the case!

    Before I commented, I struggled with the concept that no one had received initial treatment in this study, especially because the abstract in the link you provided stated “initial treatment” was a “measurement”. But the abstract results that were published in the link did not distinguish between men who had received initial treatment and those who had not. That, together with your article’s emphasis on using the data to assist one in deciding whether to undergo treatment led me to (incorrectly) believe that the men mentioned in this data had not received initial treatment.Thank you for clarifying. I no longer view the results as “incredible”. And it is now apparent that treatment was appropriate and necessary to extend the lives of many men in the study to the 14 year end point.

    I am glad I made the comment and I am glad you corrected me. All is right with the world again.

Leave a Reply

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

You are commenting using your 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: