PSA monitoring after prostate cancer surgery: how long is long enough?


A paper currently “in press” in the Journal of Urology suggests that annual PSA testing “may be safely discontinued after 10 years for men with a prostatectomy Gleason score 6 or less and/or limited life expectancy.”

Loeb et al. conducted a retrospective analysis of data from 10,609 men who received a radical prostatectomy as primary treatment for prostate cancer between 1978 and 2009 at Johns Hopkins Medical Center. Of these 10,000+ patients,  1,684 (15.9 percent) had biochemical recurrence, which was defined as a PSA level of 0.2 ng/ml or greater, at some time following their surgery. A further 1,583 patients (14.9 percent) had an undetectable PSA level at 10 years after their surgery. (An “undetectable” PSA level is not defined in the paper but is assumed to imply a PSA level < 0.1 ng/ml.)

The authors were primarily interested in studying predictors of late biochemical recurrence (i.e., recurrences occurring > 10 years after surgery).

Here are the key findings:

  • This was a cohort of relatively young, relatively low-risk, primarily white patients.
    • The average (mean) age was 58.1 years (range, 33 to 81 years).
    • Only 789 (7.5 percent) were African American.
    • The average (median) preoperative PSA level was 5.7 ng/ml.
    • Clinical stage was T1c or less in 6,618/10,609 (62.8 percent).
    • Biopsy Gleason scores were 6 or less in 8,021/10,609 (76.0 percent).
    • 6,696/10,609 (63.3 percent) had organ-confined disease at the time of surgery.
  • Surgical indicators of non-organ-confined disease at time of surgery included
    • 3,099 cases of extracapsular extension (29.3 percent of patients)
    • 1,512 cases of positive surgical margins (14.3 percent)
    • 449 cases of seminal vesicle invasion (4.2 percent)
    • 333 cases of lymph node metastasis (3.2 percent) men.
  • Among the 1,684 men who had a biochemical recurrence
    • 77.0 percent recurred within 5 years
    • 16.6 percent recurred between 5 and 10 years
    • 4.9 percent recurred between 10 and 15 years
    • 1.5 percent recurred after 15 years
  • Late recurrence (after 10 years or more) was associated with
    • More favorable pathological features
    • A higher likelihood of metastasis-free survival
    • A higher probability of prostate cancer-specific survival
  • Among the 1,583 patients with an undetectable PSA level at 10 years after surgery
    • 99 men (6.3 percent) recurred at some later point in time
    • The actuarial probability of biochemical recurrence and metastasis at 20 years varied by stage and by grade.
    • No metastases were observed in patients with pathological Gleason scores 6 or less.
    • 1/1,583 patients (0.06 percent) died of prostate cancer within 20 years.

The authors conclude that, in this cohort of patients (all treated at a single, high-volume, academic medical center), men with an undetectable PSA level for 10 or more years after their surgery have a low risk of subsequent biochemical recurrence, with correspondingly lower rates of metastasis and death.

Loeb and her colleagues further recommend that:

  • Men with an undetectable PSA at 10 years after their surgery should be counseled that their risk of subsequent prostate cancer-related morbidity and mortality is low.
  • Annual PSA testing may be safely discontinued after 10 years for men with an undetectable PSA level and a pathological Gleason score of 6 or less and/or limited life expectancy.

22 Responses

  1. So is it safe to say that for anyone who had a G7 or higher that they should continue monitoring for the rest of their lives (or to a point when they can identify another risk of mortality within a certain time period)?

    I have already resolved that my testing won’t be stopped…it was really a no brainer…

  2. I think that the best answer to that question is that it depends on the precise situation. As an example, there is a significant difference between two otherwise identical patients initially diagnosed with Gleason 3 + 4 and Gleason 5 + 5 disease. Clearly the latter is always at slightly higher risk for progression after 10 years because he is at slightly higher risk for cancer that metastasized to the bone marrow, went into stasis, and that could start to regrow at any point in time. At some point in the future, we may have enough data to expand guidance, but at present we don’t appear to have such data.

  3. So at 30 to 80 bucks a test and a history of prostate cancer why would anyone argue against playing it safe once a year? Most of us will have at least annual blood draws and adding one more vial isn’t exactly inconvenient. I would love to never worry about this stuff again, but at stage pT3b with a 4 + 3 = 7 Gleason sum it does not make any sense to stop testing until I am at least in my seventies. Right now we are testing twice a year (4 years into my journey).

    I am, however, reminded by my buddy’s father’s case. He never had a PSA test when he was diagnosed in 1981. His relapse was 29 years later with a Gleason 10 invasion of his bladder. (No PSA detected but the dissected tissue was positive for PAP.)

    For most with G6 it is probably safe to stop at 10 years, but we are never out of the woods completely.

  4. So Tony … That’s exactly why the article above is really only relevant to men with an initial diagnosis of Gleason 6 disease and lower. And after all, one is tempted to wonder how many of those 10,000+ patients might have done just as well without any active treatment at all (i.e., on active surveillance or watchful waiting).

    Your perspective is very typical of — and probably correct for — a man recently diagnosed in his 40s with high-risk prostate cancer. However, if (by the time you are in your 70s) we have data showing that risk of further progression after (say) 15 years for men initially diagnosed with Gleason 4 + 3 prostate cancer and pT3 disease who have stable disease after surgery + radiation + ADT is 1 in 1,500, then maybe you will decide you can stop your PSA tests too. After all, by then you will have been cancer-free for about 25+ years, not just 15!

    :O)

  5. I am confused about the numbers here.

    First, “Of these 10,000+ patients, 1,684 (15.9 percent) had biochemical recurrence, which was defined as a PSA level of 0.2 ng/ml or greater, at some time following their surgery. A further 1,583 patients (14.9 percent) had an undetectable PSA level at 10 years after their surgery.”

    Was it a grand total of 1,684 had BCR and of the 1,684, 1,583 had an undetectable PSA at 10 years or was it a grand total of (1,684+1,583) had BCR during the entire study period?

    If the latter is true then how did 77% of the 1,684 have BCR before 5 years?

    I don’t understand where the 1,583 comes from if only 1,684 out of the 10,000+ had BCR over the entire study period. Are they saying out of 10,000+ only 1,583 were BCR free at 10 years?

    Sincerely, Confused

  6. Chris:

    There are two quite different groups of patients here within the total of 10,000+.

    The first group of patients is a group (of 1,684 men) that clearly had a biochemical recurrence at some time (i.e., any time) after their surgery, defined by a PSA that rose to 0.2 ng/ml or higher.

    The second is a group (of 1,583 men) whose PSA remained undetectable for at least 10 years.

    A small percentage of the men in the first group must (presumably) also be included in the second group, i.e., those whose PSA was undetectable for 10 years and then started to rise. The full text of the paper states that 107/1,684 men (6.4%) had a biochemical recurrence 10 or more years after their radical prostatectomy. As indicated above, 99/1,583 men in the second group had a biochemical recurrence after 10 years. The difference between the 107 men in the first group and the 99 men in the second group likely reflects the precise data available on the 8 men who appear to be “missing” from the second group. The fact that the two numbers are so similar at 107 and 99 would strongly endorse the idea that about 100 of the 10,000+ patients were eligible for inclusion in both the first and the second groups of patients, as suggested.

  7. So out of 10,000 men only 1,583 had an undetectable PSA at 10 years? That seems like a very very high rate of recurrence.

  8. No. You need to distinguish carefully between men who had a recurrence (i.e., a PSA of 0.2 ng/ml or higher), men with an undetectable PSA at 10 years (which I am assuming means a PSA of < 0.1 ng/ml, but which is not defined precisely anywhere in the paper), and then all those men whose PSA might be slightly higher than 0.1 but is still stable and < 0.2 ng/ml (apparently something of the order or 6,890 out of the total of 10,060 men in this study cohort).

    If Loeb et al. actually used a figure of < 0.01 or some other number to define undetectable, then the number of men with an "undetectable" PSA at up to 10 years is still 1,583, but the range from "detectable" to "recurrence" increases from 0.1 to 0.19 to 0.01 to 0.19. I can't clarify this further since I don't know how Loeb et al. defined undetectable.

  9. OK. I think I get it now.

    1,684 men met the clinical definition of recurrence with a PSA above 0.2.

    All the rest were below that and 1,583 were actually below the detection limit of whatever lab they were using at 10 years. Still seems kind of high to me that such a large amount of people had detectable PSA after 10 years even though they were below 0.2.

  10. Chris:

    I have e-mailed Dr. Loeb to see if she can do two things for me:

    (a) Clarify that an undetectable PSA actually did mean a PSA of < 0.1 ng/ml.
    (b) Tell me how many men in the total cohort of 10,000+ were actually followed for a full 10 years.

    You need to appreciate that any patient in this cohort who had surgery after 1999 could not possibly have been followed for 10 years, so this means that the number of men who actually did not have either an undetectable PSA at 10 years or a PSA recurrence within 10 years might be only of the order of 3,000 men or so. Does that make sense to you?

  11. Mike,

    A question: When did widespread PSA screening become the norm? If those 3,000 men were all treated in or before 1999, is it accurate to generalize that they were more likely to have been diagnosed with later stage, or to say it another way, less likely to have been diagnosed with very low-risk and indolent prostate cacner?

    If the answer to my question is “yes”, then most of those 3,000 men were symptomatic in some way prior to treatment, yet approximately half of those men had no recurrence at 10 years? So … does that suggest that up to half of even “symptomatic” prostate cancers are actually likely to have not micro-metastasized, or be non-aggressive, or indolent?

    Obviously, I’m hypothesizing here. I know that you can’t provide immediate answers. But attempting to answer such questions would/should seriously reframe the issue of what constitutes appropriate standards for screening, “early” diagnosis, and standards of treatment for “cancers” that do not even rise to the grade or pervasiveness of those 1500 men who were radically treated more than a decade ago – perhaps unnecessarily.

  12. Tracy:

    Widespread use of the PSA test to evaluate patients prior to radical prostatectomy had become the norm at major prostate cancer centers like Johns Hopkins by at least 1992 and arguably earlier. Furthermore, in pursuit of his efforts to demonstrate the value of nerve-sparing surgery in the treatment of prostate cancer, Dr. Pat Walsh had been making every effort to seek out and treat younger men with the earliest stages of prostate cancer from 1982 onwards.

    If anything, for a series of patients that begins in 1978, this series was likely to have had a relatively high percentage of patients who met criteria for low- and very low-risk disease. In other words, the answer to your initial question is almost certainly “no.” My guess would be that the vast majority of these patients would not have been symptomatic at the time of treatment.

  13. I guess it does not matter much. To me a 6% recurrence rate after 10 years is still pretty high. I would not consider that low enough to stop testing.

  14. My father was disgnosed in 1999 and had his prostate removed … no sign of cancer in his lymph nodes and no further treatment was required. I do not know his Gleason score. In April his PSA had risen to 3.8, and in June to 7.9. He is scheduled for another appointment tomorrow. He has been exhibiting some extreme forgetfulness, and I am very concerned. He is 66 years old with a history of HTN and DM.

  15. Dear Sportwood:

    Please join our social network which is specifically designed for discussion of individual cases and issues affecting the management of such individuals. The fact that there was no sign of cancer at the time of your father’s original treatment does not mean that there may not have been a very small site of early spread of his cancer that has taken years to develop.

  16. I had a radical prostatectomy 8 years ago. Now my PSA is 0.24 ng/ml over my last 3 tests (i.e., steady at 0.24 over a 9-month period). How can one biopsy for Gleason with no prostate? What should I fear if anything at this point?

    R

  17. Dear Rick:

    (1) If your PSA is stable, there may be little to nothing to worry about. There could, for example, have been a small amount of non-cancerous prostate tissue left behind at the time of surgery that has grown back a little after 8 years. If your PSA starts to rise further, that is a whole other matter.

    (2) It is possible to biopsy appropriate areas of the so-called “prostate bed” (i.e., the area around where the prostate used to be) to look for any signs of prostate cancer in a man who has already had a radical prostatectomy. However, the difficult element is determining exactly which tissues to biopsy. Various types of scan — and particularly sophisticated types of MRI and PET scan — can be used to try to visualize areas of concern. However, your PSA level is still low at present and imaging may not show anything at this stage.

    (3) If your PSA starts to rise again, one of the things you may want to discuss with your doctor is salvage radiation therapy to the prostate bed and pelvis.

    (4) Your risk depends, to a significant extent, on your original PSA level at diagnosis and on the pathological findings after your original RP. If you had extracapsular extension or positive seminal vesicles, or positive lymph nodes, or even a tiny amount of positive surgical margin at the time of your original surgery, then your risk for progression is higher. The same is true if you had a high pathologic Gleason score like 8, 9, or 10.

  18. Hi guys. Need some clarification! I had an RP on 08/16/2012; post-op Gleason 3 + 4; positive surgical margin, but don’t know the extent or length (path report didn’t say). I’ve had PSA tests every 3 months since then, with the first two coming back undetectable, i.e., < 0.05.

    Two weeks prior to my last PSA test I had a colonoscopy performed. My results from the PSA test (different lab this time) came back detectable at 0.028 ng/ml. My urologist is talking secondary treatment already, with my PSA still well below the BCR limit. Did the instrument from my colonoscopy cause my PSA to be detectable by engaging residual prostate tissue which released PSA into my bloodstream? I'm having PSA anxiety guys. Please help! I have another PSA test scheduled for July 1 to see if it was a lab error or if secondary treatment is right around the corner! Hopefully not!

  19. Dear Mike:

    (1) It is extremely difficult to compare PSA data this low from two different laboratories, so if I was wearing your shoes I wouldn’t be reading too much into the 0.028 result until I had had the second test done, and I would point out that, in any case, 0.028 ng/ml is still about half of 0.05 ng/ml (the original “undetectable” value).

    (2) I think you have done the right thing to get another PSA test done. However, you need to be clear whether this is going to be done by the first lab or the second. (I hope it’s not from a third lab because that won’t help one scrap!)

    (3) I honestly don’t think the colonoscopy has any relevance at all here. The half-life of PSA in the blood stream simply isn’t long enough have been affected by a colonoscopy done 2 weeks before blood was drawn for your last PSA.

    (4) Even if your next PSA comes back somewhere between 0.025 and 0.030 ng/ml, I would certainly not consider any form of further treatment yet if I was wearing your shoes. I’d just look at the doctor and say, “Let’s get another in 3 months before we rush to any decisions.” If you think about it, this may just be the PSA level you have had ever since your surgery. If it is stable at something like 0.025 ng/ml, it is well below any level that would clearly signal biochemical recurrence.

  20. Sitemaster , why would the second lab report the results as detectable being at a lower level ?

  21. Because different labs may use slightly different tests and technical methods. It’s not as though there is just one PSA test; there are several different ways to measure PSA, although some are much more common than others.

  22. Sitemaster,

    Thanks for clearing things up. I do feel better … but will feel even better if my next PSA results come back lower or stable! I will post them when the results come in, after 07/01/2013.

    Thanks again!

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 1,154 other followers