Criteria for active surveillance are gradually improving

Being diagnosed with low-risk prostate cancer is a good reason to consider management under active surveillance (as opposed to immediate, early intervention) … but active surveillance won’t be right for every low-risk patient!

Even if we ignore the fact that some patients’ reaction to a diagnosis  of prostate cancer of any amount, and however low the risk level, is, “Get that cancer outta me,” there are other reasons why active surveillance (maybe in the short term and certainly in the long term) is going to be inappropriate for many patients.

This fact has now been better characterized by data from > 13,000 men collected in the National Prostate Cancer Register of Sweden. Every one of these men was initially diagnosed between 2007 and 2011 with Gleason 6 disease and a clinical stage of T1c/T2. Precisely 4,500 of these men (roughly a third) had a radical prostatectomy, and for about half of those 4,500 men, the registry contains data on the extent of prostate cancer in the biopsy cores.

The new study by Vellekoop et al. provides us with the following, specific insights:

  • The database included 13,159 patients.
  • Among the 4,500 men who decided to have a prostatectomy, 2,235 (49.7 percent) had adverse postoperative pathology (based on their pathological stage and/or Gleason score).
  • Among men who decided to undergo radical prostatectomy and who met rigorous study inclusion criteria of six different active surveillance protocols,
    • Adverse pathology was present in 33 to 45 percent of patients.
    • The lowest risk for adverse pathology (33 percent) was seen in men who met the historic, very rigorous, Johns Hopkins criteria for active surveillance.
    • The highest risk for adverse pathology (45 percent) was seen in men who met the less rigorous criteria for active surveillance used by the University of Toronto.
  • Predictors of adverse pathology were
    • Older age
    • Higher PSA levels at diagnosis
    • PSA density > 0.15 ng/ml/cm3
    • Palpable disease on rectal examination
    • Extent of cancer > 4 mm on biopsy.
  • Greater prostate volume had an inverse relationship with (reduced the probability of) adverse pathology.

In the full text of the paper, the authors go into extensive detail about three specific recommendations:

  • Clinicians should include a PSA density of ≤ 0.15 ng/ml/cm3 as a key criterion for inclusion of patients on an active surveillance protocol.
  • Clinicians should also include the sum of the linear extent of cancer (in millimeters, in all biopsies) as a second key criterion for inclusion of patients on an active surveillance protocol.
  • A sufficiently extensive initial or follow-up biopsy is essential. (Although the authors do not provide a definitive description for such a protocol, it is apparent that they would consider any TRUS-guided biopsy with less than 10 or 12 cores to be insufficient.)

It is worth comparing these recommendations with recent suggestions from Johns Hopkins about how they intend to improve their own criteria for inclusion on active surveillance.

Now on the one hand, this study does not mean that men at potentially higher risk should be told that they absolutely are not eligible for active monitoring as a management strategy; nor, on the other hand, does it men that men who meet these criteria “have to” go on active surveillance. There is (and rightly) a strong element of patient choice associated with these decisions. What we do seem to be getting closer to is a set of protocols that can define patients who have a good probability of high quality, long-term care on active surveillance — and if that is more tghan half of all the patients who get diagnosed today with low-risk prostate cancer, that is still an awful lot of men who really do not need any form of immediate intervention.

Why is this particular paper of importance? … Because unlike the data from the various institution-specific series of patients at institutions like Johns Hopkins, these are national data from a much larger set of unselected patients, so they give us a better perspective on risk across the patient continuum.

11 Responses

  1. One very interesting item: “the larger the prostate, the lower the pathology risk”. I wonder if that is because a larger prostate generally produces a higher PSA score, thus making the guy appear to be at more risk that he actually is? I have personal experience with this. My PSA score had gradually crept up to about 3.8 (which triggered an unpleasant and unnecessary biopsy). Later, my prostate had enlarged to the point that I couldn’t pass urine and I ended up undergoing a rather horrific operation that greatly reduced (but did not remove) my prostate. My PSA now reads about 0.45.

    We have yet to see how the two new tests (OncotypeDx and Prolaris) will fit into protocols for determining candidates for AS. In a video link you posted earlier, one of the doctors mentioned these tests as the biggest factor he sees in the clinical management of prostate cancer in 2014.

  2. Dear Doug:

    If I am understanding the full paper correctly, the comment about the inverse risk associated with prostate volume has more to do with the amount of cancer in prostates in different sizes.

    So, for example, I think that the authors are saying that, on average, if two men have a single 5 mm biopsy core showing Gleason 6 prostate cancer, but one man has a 50 cc prostate and the other has a 100 cc prostate, then the man with the larger prostate is at lower risk for upgrading or upstaging. PSA level is a quite separate risk factor.

  3. Of course, one problem with the analysis is the belief that going from Gleason 3 + 3 on biopsy to 3 + 4 on final pathology will necessarily result in a man’s death from prostate cancer. The reason for this statement is that older studies of patients who were on watchful waiting with a Gleason 6 cancer had a very low chance of dying from their disease. This must mean that Gleason 7 disease is not always dangerous because we know that many of the men with Gleason 6 on biopsy would have a higher tumor grade following a prostatectomy.

  4. The paper is very interesting but among the predictors of adverse pathology were: older age. In my view this also shows also the limit of this kind of study because they look at adverse outcome at pathology rather than adverse outcome for the patient (such as metastasis or death). Nor do they report on quality of life in after active surveillance as opposed to surgery. So at the end of the day this does not really help to advise patients.

  5. I have always wondered why young “age” is never a criteria considered or at least mentioned when classifying men as low risk. I was 42 and considered low risk at diagnosis. I remained low risk after RRP. Is there an age that is considered too young for AS? Who knows how many men used to have prostate cancer in their 40s when a PSA of 0-4 was normal (as mine was) and then went for who knows how many years unknowingly on AS until their PSA rose above 4.

    Wish I could have put off treatment for a long time.

  6. Chris:

    The percentage of men who actually get diagnosed with prostate cancer when they are in their 40s is very small.

    Some of those men almost certainly don’t need to have treatmernt immediately, and could just be monitored for some years until treatment becomes necessary, but I don’t think anyone has a clue what percentage of men diagnosed in their 40s that would be true for (as yet).

    It is also, potentially, the case that a small subset of those men diagnosed in their 40s might have utterly indolent disease that never needs treatment … but I don’t think there would be too many of those over the next 35 to 40 years.

    As far as I know, no one has ever attempted to define an age at which patients with very low-risk disease are “too young for active surveillance”.

  7. Chris,

    Age is a consideration. One of the best AS programs is the one at Johns Hopkins. Although they might accept younger patients, their protocol (see below) has a life expectancy limitation for admission.

    Very Low Risk Prostate Cancer
    — Life expectancy less than 20 years (62 years)
    — Cancer not felt on digital rectal examination (stage T1c)
    — PSA density (PSA divided by prostate volume) < 0.15
    — Gleason score 6 or less with no Gleason pattern 4 or 5
    — No more than 2 cores with cancer, or cancer involving no more than 50% of any core on
    at least a 12-core biopsy

    Low Risk Prostate Cancer
    — Life expectancy less than 10-15 years (76 to 69 years)
    — Cancer not felt on digital rectal examination and/or small nodule (stage T1c or T2a)
    — PSA below 10 ng/ml
    — Gleason score 6 or less with no Gleason pattern 4 or 5 on at least a 12-core biopsy

    I used the Social Security actuarial table to find ages at the limiting life expectancy.

  8. SIZE OF PROSTATE AND PATHOLOGY RISK (Comments from Doug and Sitemaster)

    Thanks for highlighting this interesting study that takes a population approach rather than an institutional approach to active surveillance.

    I’m thinking about Doug’s comment of January 15, 2014 at 2:06 pm and Sitemaster’s response of January 15, 2014 at 3:57 pm. What Doug wrote resonates with me as larger prostates, due to benign growth that is mostly not cancer, will produce more PSA because there are more normal cells, and that higher PSA will tend to set off alarms, especially if not related to size. For instance, some research indicates that a healthy cubic centimeter of prostate cancer will produce about 0.066 units of PSA, so a very large prostate of 100 cc with only healthy cells would be expected to produce about 6.7 units of PSA with none due to cancer. Though there are different rules of thumb, they agree on the point that a greater volume of healthy cells will produce a higher amount of PSA.

    Regarding the analogy of the man with a single 5 mm biopsy core showing Gleason 6 prostate cancer, but one man has a 50 cc prostate and the other has a 100 cc prostate, if the thinking of the advocates of 5-ARI use (finasteride/Proscar, dutasteride/Avodart) for prevention as well as support during active surveillance and ADT is right (their thinking is not directly related to this topic but involves the same principle), then the man with the larger prostate would be at higher risk because the overworked probes would be more likely to miss existing cancer, whereas the same number of probes in a smaller target area would be more likely to hit existing cancer than if those probes each was responsible for sampling a larger area. In short, a larger area dilutes the sampling power of a probe when the same number of probes are used.

    I haven’t read what the authors have to say about this, but have accessed the complete paper and look forward to studying it. Thanks for the link! (I am sandwiched between church and the Denver Broncos versus New England Patriot’s game and need to eat that sandwich for lunch now.)

  9. Jim:

    All this hypothesizing is all very well … but the recent review of the Johns Hopkins active surveillance data (like the Vellekamp study above) showed that the thing that was important was the PSA density … not the size of the prostate, nor the absolute PSA level (if the PSA was less than 10 ng/ml and the patient had a Gleason score of 6 or less).


    In one sense, age is important when men are quite elderly, as age is the key element in life expectancy calculations, and treatment is clearly of questionable value when life expectancy is short.

    On the other hand, my impression as a fairly close observer of active surveillance research is that age is falling away as a criterion for younger men. At one point not that long ago it was an important criterion at some major centers. I had a chance to probe this issue at the September 2007 IMPaCT Conference, Session 30 (Plenary, Friday afternoon) entitled Clinical Research, which would have been better titled “Active Surveillance with Deferred Therapy,” hosted by Dr. Christopher Logothetis. Panel speakers were Robert Carey, Dr. Stephen Freedland (Johns Hopkins), Dr. Laurence Klotz (U. of Toronto, Sunnyvale, a leading program), Dr. Fritz Schröder (Erasmus Medical Center, Netherlands, a leading program), and Dr. Christopher Warlick (U. of Wisconsin, recently from Johns Hopkins, another leading program). To put this in context, the conference was held in the fourth year after the first paper on results under AS was published by the Toronto group, a short time span in terms of medical research and implementing research in practice. I was representing survivors (“consumer representatives”) at this conference (Innovative Minds in Prostate Cancer Today), a review of the first ten years of research under the Prostate Cancer Research Program sponsored by the Congressionally Directed Medical Research Program (CDMRP), as well as presentations by leaders in prostate cancer research.

    After the presentations I asked an audience question whether there was an age cut-off in active surveillance at each program represented on the panel. (Some of this may be available online, but probably not the Q&A.) Dr. Warlick, speaking about Dr. H. B. Carter’s AS program at John’s Hopkins, said that the program preferred older men but that 44% of participants were younger than age 65. Dr. Schröder had mentioned an age cut off of 55 in his talk; his institution was uncomfortable with AS in younger men. My impression was that their comments reflected the state of thinking at the time for most doctors advising patients about active surveillance: discomfort with younger men doing AS. That’s why my sense was that the room became quite quiet when Dr. Klotz stated he thought patients of any age were suitable for AS, provided they satisfied criteria! He added that younger patients would be monitored more closely. Then, as now, Dr. Klotz would likely be named by most as the leading clinician/researcher for AS, so his view had extra impact.

    Sometime also in 2007, Dr. Peter Carroll of UCSF (another leading AS program) convened an international conference of leading experts on AS, a session described on pages 66 and 67 of the Scholz/Blum book Invasion of the Prostate Snatchers (2010). The conferees issued a consensus set of six criteria for AS: GS < 7; < 34% biopsy cores with cancer; PSA < 10; PSA velocity < 2 per year; PSA density < 0.15 (echoed in the study featured above); and no palpable node on DRE. (Violating a single criterion was viewed as grounds for serious doubt about success for the patient under AS.) But, as Invasion puts it, “The conference attendees believed that if the basic requirements of Low-Risk disease are met [as listed], men at any age could safely undergo active surveillance.”

    One well-known doctor specializing in prostate cancer has a number of men in their forties on AS.

  11. It is critical for people to appreciate, in considering who is an “appropriate” patient for active surveillance, that there are three different and non-competitive potential objectives in the application of active surveillance:

    A. To avoid all treatment for all time in a man who may never have any clinical symptomatology or problem with his cancer during his lifetime.

    B. To defer treatment for as long as possible in a man who may need treatment for low-risk disease at some point in his lifetime so as to maximize his quality of life for as long as possible.

    C. To be able to defer immediate treatment while still ensuring that effective treatment is given in a timely manner to men who need it to void any significant risk of death from prostate cancer.

    Given these three very different but interconnected objectives, it is certainly reasonable to consider at least a period of active surveillance in a very high percentage of men at any age … so long as they understand what is going on. That is why Klotz has always felt that active surveillance is appropriate for patients of any age. Others started out looking at active surveillance by considering only objectives A and C.

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 )

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: