The scientific basis for PSA monitoring after first-line treatment for localized prostate cancer

A new article in the Canadian Medical Association Journal has reported on a careful analysis of the published literature underlying current guidelines on the use of repetitive PSA testing to monitor risk for progression in men receiving first-line treatment for localized prostate cancer. Many patients may be surprised just how few data there are to support this extremely common clinical practice.

Dinnes et al. set out to determine the extent to which recommendations for repetitive monitoring of PSA levels to detect recurrent prostate cancer actually met established scientific criteria that should inform rule-based strategies for disease monitoring. The full text of this paper is available on line for the interested reader.

The authors identified a total of seven sets of clinical guidelines and two “statements of best practice” from nine organizations (four in North America, four in Europe and one in Australia). They then reviewed these guidance documents with care against the standard Appraisal of Guidelines for Research and Evaluation Framework.

The bottom line to their findings was pretty straightforward:

  • There is “considerable inconsistency” in recommendations for use of PSA testing to monitor men after first-line therapy.
  • Recommendations on when to test are largely based on “standard” follow-up schedules that have little scientific basis.
  • Recommendations on when to take further therapeutic action are based on consensus statements or on retrospective case series.
  • Eight of the nine guidance documents recognize “the potential presence of measurement variability,” but make no attempt to address the impact of such variability on the interpretation of  results of sequential PSA tests.
  • Recommendations are often made with few or no supporting references.
  • Of 48 papers cited as sources for recommendations, 14/48 (29 percent) were review articles and 34/48 (71 percent) were primary studies.

The findings of this study, again, come as no great surprise to long-time watchers of the prostate cancer world. There is little evidence for and no consistency about a whole variety of issues that underlie the question of how best to determine real recurrence of prostate cancer after first line treatment.

Here are just some of the questions to which we do not, actually, have good, science-based — as opposed to opinion-based — answers (and they certainly aren’t the only ones):

  • What should a man’s PSA level be after whole gland surgery administered with curative intent today (e.g., < 0.1 or < 0.01 ng/ml)?
  • What should a man’s PSA level be after radiation therapy admininistered with curative intent today (i.e., what is a truly acceptable nadir PSA level)?
  • What is the correct way to determine recurrence after surgery (e.g., a rise in the PSA level to ≥ 0.2 ng/ml or a PSA rise to ≥ 0.1 ng/ml with a doubling time of < 15 months)?
  • Are either the Phoenix criteria or the ASTRO criteria really accurate ways to assess recurrence after radiation therapy?
  • What should trigger initiation of salvage treatment for a man with recurrence after surgery?
  • What should trigger initiation of salvage treatment for a man with recurrence after first-line radiation therapy?

Obviously, there are no “one size fits all” answers to these questions. A 48-year-old man with a recurrence 9 months after first-line surgery for localized disease characterized by a PSA of 2.7 ng/ml and 2/12 small positive biopsy cores, one of which contained  5 percent of Gleason 9 prostate cancer, is at a very different long-term risk level than a man of 81 years with a small rise in his PSA a decade after first-line radiation therapy for localized disease characterized by a PSA of 4.8 and 4/12 positive biopsy cores all containing Gleason 6 disease.

The truth, however, is that we still, really don’t have scientifically established guidelines for assessing and managing the recurrence of prostate cancer, and we don’t have this type of guidaance because we don’t have data to support it.

5 Responses

  1. PSA testing was first approved by the FDA to monitor patients for biological failure of “first-line” localized therapy in 1986. Early detection and treatment of biological failures have helped reduce the death rate as illustrated by the SEER data.

    Biological failure has no absolute definition. After surgery the PSA value should be undetectable, which may result in a 0.1 ng/ml result depending on test method and lab controls. An increase to 0.2 ng/ml suggests failure, without a prostate the PSA increase has one source and that is failure of the local therapy.

    Biological failure after radiotherapy is much more complex. Under external beam forms of radiotherapy, the prostate receives a dose plan which varies in total dose (68-86.4 Gy for IMRT; 35-38 Gy for CyberKnife), varies in dose per session (1.8 Gy for IMRT to 9.5 Gy for CyberKnife), varies in number of sessions (35-45 for IMRT; 4 or 5 for CyberKnife; 28-36 for PBRT), and ionizing energy source varies (photons for IMRT, brachytherapy, and Cyberknife; protons, less common; and neutrons, rare). Brachytherapy has variations in energy source and dose delivery rate between HDR and LDR. Add to this the variability of our biology and of cancer biology and the mix is massive, without an absolute PSA value suggesting failure. Radiotherapy is known to have PSA bounce(s) in about 25% of patients; a bounce is not absolute. If it is going to occur, a bounce(s) usually does so within the the first 24 months post-radiation. I think the ASTRO-defined biological failure as a 2.0 ng/ml increase over nadir seems like a big number but effective.

    Personally I have been tracking my PSA values (including velocity) for 17 years and credit this to early detection and early treatment. My father was not so lucky. he was diagnosed in 1980 (before PSA screening) at 69 years of age; the disease had spread to his bones and lymph nodes when detected. My view of the benefit of PSA screening is a little biased based on my experience.

    Family history and PSA markers are indicators that can be very valuable for early detection of prostate cancer and for early detection of biological failure.

    What is the economic value analyzing PSA data for biological failure? With advanced radiotherapy the recurrence rate is very low, 2-3% at 5 years for the CyberKnife and HDR brachytherapy, a relatively small market.

  2. Had surgery 3 years ago to remove prostate; had PSA test every 3 months, so far always < 0.1 ng/ml. I weigh 268 lb and am 53 years old; does weight factor into this?

  3. Hugh:

    If your prostate and any other tissues that might have carried prostate cancer have been successfully removed, your body weight is irrelevant to your subsequent PSA levels, which should always remain below 0.1 ng/ml. However …

    Your body weight is far from irrelevant to a bunch of other issues (like your risk for cardiovascular disease) which may be very important to your long-term life expectancy. It is particularly important for larger men that they exercise regularly and optimize their diet to stay healthy and fit. In other words, it is not just a matter of the body weight per se, but your overall fitness and health.

  4. I am not a doctor but to say there can be a bounce in PSA after successful radiation therapy to me sounds like when they say first-born children sometimes only take 5 or 6 months to be born. It sounds like made up “facts” to cover up the much more logical. The baby was conceived before the marriage and all the cancer cells were not killed.


  5. Dear Carl:

    You are entirely entitled to your opinion, but relatively small “bounces” in PSA values several months after completion of radiation therapy and achievement of nadir PSA levels are, in fact, extremely well documented, and there are several potential explanations for this. Such “bounces” in the PSA level may be followed by stabilization of the PSA again (effectively for life) or by a decline in the PSA level again after the “bounce”.

    Progression of cancer after radiation therapy may well begin with something that looks very like a “bounce.” However, progression is characterized by continued rise of the PSA over time, even if the first rise looks like a “bounce.” The difficulty comes in being able to decide who needs treatment (because it is real progression) and who does not need treatment (because it is a bounce). Some “bounces” really are just that, and exactly why they occur in specific individuals can be extremely hard to determine.

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

%d bloggers like this: