Understanding your clinician’s mindset about active surveillance

As one of our regular commentators constantly points out, active surveillance (AS) is not a treatment for prostate cancer. It is a means by which to manage risk for prostate cancer progression in men with low- and very low-risk disease or higher-risk disease but a limited life expectancy. Another way to describe this is that AS is a means by which to defer unnecessary treatment until it becomes necessary for men with low-risk and early stage disease.

As many of our readers will be aware, the “classic” candidate for AS is a man of 65 years or older who has a life expectancy of up to 15 years and has been diagnosed with Gleason 6, clinical stage T1c prostate cancer after findings showing a PSA of < 10 ng/ml, a negative digital rectal exam, and one or two positive biopsy cores with a limited percentage of tissue positive for cancer.

The premise behind AS is that a high percentage of these patients (of the order of 70 percent) will never demonstrate evidence of clinically progressive disease during their lifetime, and so treatment should be avoided if reasonably possible. This also implies that about 30 percent may well need treatment, and so all such patients should be carefully monitored in order to identify and act appropriately for those patients who show signals of risk for more aggressive disease (ideally confirmed by a higher Gleason score on repeat biopsy).

A recent commentary in “Beyond the Abstract” on the UroToday web site, however, tells us that at least some clinicians and researchers do not necessarily see things in quite this light. In this commentary, Pontes includes the statement, “The inadequacy of patient selection is one explanation for active surveillance failures.” By “failures” he is referring to men who need therapy after a period on AS.

Now we would argue that these patients are not “failures” at all. Indeed, precisely the opposite, these are men in whom AS has clearly identified a need for treatment after a period of careful monitoring. They are therefore successes! They are men in whom therapy has been deferred until there was evidence of need for invasive treatment (which was then either offered and rejected or offered and given). In the series of patients originally studied by Oliviera et al. (and on which Pontes is commenting), 93 patients met clinical criteria for AS (out of a total cohort of 406 patients) but were treated by radical prostatectomy. Of these 93 patients, 25.3 percent showed pathologic evidence of higher Gleason grade and 8.3 percent of higher stage disease on pathologic examination of the prostate post-surgery. (These are the patients Pontes is referring to as “failures” of AS.)

Pontes goes on to argue that “one third of patients that fulfill the criteria for insignificant cancer may actually harbor cancer with aggressive characteristics, implying that the current criteria of selecting patients for surveillance should be cautiously adopted. A reliable model to predict insignificant prostate cancer is needed, and until that happens, the prediction of clinically insignificant prostate cancer will remain a major problem in urologic practice.”

He appears to be missing the point of AS entirely. If and when we have “a reliable model to predict [clinically] insignificant prostate cancer,” we won’t need AS at all! AS is a technique that allows us, in the interim, to manage risk in those patients who we hope have clinically insignificant prostate cancer.

So what is a “failure” of AS? The “failures” are those patients in whom, despite careful selection and monitoring, there is progressive disease that does not allow for successful treatment of localized disease with curative intent at the time of evidence of increased risk. In Klotz’s series of patients, at 10 years of follow-up, only 17/450 patients or 3.8 percent have demonstrated prostate cancer-specific mortality (definitive evidence of a “failure” of AS). Also in this series, 117 patients received definitive treatment for localized disease after a period on AS; half of these 117 patients — 58/450 patients (13 percent) —  subsequently demonstrated a rising PSA. These patients can also be defined as “failures” of AS (although that is arguable if they never ever demonstrated subsequent symptoms of progressive disease). It is not entirely clear how many of these 117 patients had chosen to have definitive treatment because of evidence of increased risk and how many elected to have definitive therapy because they “couldn’t deal with” knowing that they were living with prostate cancer — even though there was no good evidence of increasing risk.

Klotz’s series includes many men who were well under 65 years of age at the time they started AS. It is also worth noting that Klotz’s series reflects some of the earliest carefully monitored experience of AS. The protocol that he and his colleagues use today for entry into their AS series and the methods for monitoring of their patients has been adjusted over time.

The critical point here is that if the mindset of the treating clinician is that AS is a poor substitute for immediate therapy (as opposed to the optimal means currently available to defer or avoid unnecessary treatment), then that clinician may not be fully committed to seeing AS as a management technique that is in the best interests of the patient. It is true that we are still refining the criteria for selection of candidates for AS, but we have already, and clearly, shown the value of this management technique in thousands of men. Furthermore, in perhaps many patients on AS, pathological evidence of progressive disease at the time of death may still not be associated with any clinical evidence of a significant problem. If one dies of something else, with extracapsular extension of Gleason 7 prostate cancer that never had any symptoms and that one never knew about, then (at least arguably) treatment of that prostate cancer was still never necessary, and AS was a successful management technique!

2 Responses

  1. Thanks once again for this selection and your thoughtful, well-reasoned exposition regarding active surveillance (AS).

    I’ve read the source article by José Pontes Jr., MD, et al., and, while not right-on-the-money regarding AS, the article struck me as informative about the mindset of many more conventionally oriented surgeons as they reluctantly grapple with the impressive success of AS.

    Dr. Pontes uses an unfortunately misleading frame-of-reference in the following statement, I think, regarding the Klotz results at the 10-year point: “It was interesting to see that half of these “drop out patients” showed biochemical recurrence when treated with radiation or surgery. This recurrence rate is above that observed in contemporary radical prostatectomy series meaning that these patients were inadequately selected for surveillance and perhaps deserved to be treated.” Let’s think about that. In a prostatectomy series, there is typically a mix of patients who truly need treatment (the “drop outs” from an AS series plus those with higher risk clinical characteristics), and these patients are mixed in with other patients, many of whom would probably not need treatment if under an AS approach. Therefore, the 30% “drop out” number in the author’s article, based on the Klotz series, needs to be spread over a larger base — not just the patients who would qualify for AS. Let’s give that a try.

    There were 406 patients in Dr. Pontes RP series, and 93 would have qualified for AS using some criteria fairly close to what is now being recommended. The post-RP biopsy results showed higher risk results in “one third” of these 93 patients, so we can use 31 as our number for patients at higher risk that was hidden pre-surgery. These 31, as I’m looking at it, would be equivalent to the “drop outs” from the Klotz series that Dr. Pontes mentions (essentially, another affirmation of that number of approximately 30% who will need treatment after time in AS). Well, if we spread those 31 cases over the entire group of 406 patients, the result is 7.62% who might be considered “failures” under AS, though Sitemaster has some apt comment about that. However, even taking 7.62% as it is, that number, a not-needing-treatment rate of about 92.4%, stacks up well against outcome research from prostatectomy series for low-risk patients. I’m looking at a graphic for non-recurrence after RP versus follow-up time informally published by the Prostate Cancer Results Study Group, and only two of the series meeting their criteria showed success that good (one about 93% at about median follow-up of 4.5 years, and one at about 92% at about 9 years). Six other RP series had non-recurrence rates below the 90% level (follow-ups ranging from about 5 years to about 10 years). I hope someone notes that in an official comment on the Pontes article. Have I missed something here? It looks pretty clear cut.

    Another significant issue is the criteria the Pontes group used in generating their equivalent of a group eligible for AS: “Today, surveillance is usually offered for patients with Gleason score <7, stage T1c, PSA <10 ng/mL and with less than 25% of biopsy cores positive for tumor." While there is still no "official guideline", Dr. Peter Carroll, the noted surgeon from UCSF, called an AS conference of 200 experts in 2007, and they came up with a consensus statement (notably omiting age): Gleason <7 (same as Pontes); PSA level <10 (same as Pontes); Digital Rectal Exam indicating no nodule (equivalent to Pontes stage T1c, tighter than the limit of Stage 2a used in a number of AS programs); % of biopsy cores with cancer <34% (similar to but more liberal than Pontes at <25%); PSA velocity of <2 (not included in Pontes); and PSA density of <0.15 (not included in Pontes). It follows that the additional two criteria would snag some patients who otherwise would be eligible for AS, at the least triggering further staging. But the point is that the Pontes criteria are not as sophisticated as the current consensus statement on AS eligibility, at least per the consensus statement from one expert group. That's not to say the Pontes group's definition was not reasonable, but rather, it was arguably not quite the current state of the art. Moreover, some practices are using additional tools to screen potential AS patients, including color Doppler ultrasound, endorectal MRI with spectroscopy, finasteride or dutasteride challenge, and PCA3 results.

    Another point, echoing Sitemaster, is that the Pontes' results do not prove that having the 31 men on active surveillance for a while results in "undertreatment," as the article asserts. It seems clear that treatment would be delayed for a time, but results from major AS programs show that most patients ultimately determined to need treatment are picked up by years 2 or 3. Perhaps more importantly, accumulating follow-up evidence on those treated after being on AS is showing that their cancer control and side effect outcome results are highly comparable to those of patients treated immediately.

    Here's a final point, recalling Dr. Willet Whitmore's famous caution, about the Pontes team's final point, which was: "A reliable model to predict insignificant prostate cancer is needed, and until that happens, the prediction of clinically insignificant prostate cancer will remain a major problem in urologic practice." I can't help noting that [the lack of a] a reliable model for picking out which patients need treatment, and which patients can be successfully treated with surgery, has not stopped many surgeons from operating on prostate cancer patients!

  2. An excellent piece and response from Jim.

    One aspect of the AS studies — and the anecdotal evidence from my site — is that many men in the studies have subsequent negative biopsy results. In private mail I am told that this applies to about 25% of the men.

    Now I know that a negative biopsy result doesn’t prove anything because it only means that it is not positive, BUT … if the man had this result from his first biopsy he would not have been diagnosed as having prostate cancer and he would have joined the hundreds of thousands (maybe millions) of men who undertake AS without being diagnosed and who ultimately die from other causes, still completely ignorant that they carry a small package of defective cells.

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 )

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: