Active Surveillance for Prostate Cancer: Age of the Patient
We have a closely related thread now ongoing - "Effectively Preserving Curative Options- Active Surveillance for Prostate Cancer" - in which the role that the patient's age makes has come up in discussing the decision whether to pursue Active Surveillance (AS) for prostate cancer instead of immediate treatment. I hope that discussion continues, but because age is such an important factor in the thinking of the major Active Surveillance programs in the US and world, and because their approaches differ substantially, I thought it deserved a separate thread.
Here's the essence of what I know about it: while most AS programs strongly prefer older patients, at least one is comfortable with patients who are at least 55 years old, and at least one will take truly low-risk patients of any age, and that program is led by a doctor who is arguably the leading guru for AS. He also has a well thought out rationale for his position.
Most of the AS programs prefer that their patients be older, with the later 60s and older considered ideal. On the flip side, for older men and those with significant, likely life-shortening other health conditions, the major AS programs are willing to go with higher-risk case characteristics. This is all kind of logical - the risk of treatment versus benefit of treatment balance is tilted toward higher risk and lower benefit for older patients with low-risk cancer. Obviously, if you are 70 years old and have a prostate cancer that is very unlikely to become a serious threat or burden on your life for at least 30 years, but with a risk of side effects of treatment that can be substantially higher and more intrusive and burdensome for many men as they become elderly, the net likely value of getting treated declines. Similarly, if a man has, for example, diabetes that is difficult to control and serious heart problems, especially if he is elderly, a low-risk case of prostate cancer is not likely to be a significant threat to his life or well-being, and possible complications from prostate cancer surgery (such as blood clots) or other treatment may seriously aggravate other health conditions . There is strong agreement among the AS programs that AS is a highly attractive option for such men.
But the real issue, still controversial and without an abundance of information yet, is what to do about younger men who have case characteristics that all suggest very low risk prostate cancer. Now that many men are being screened with PSA and DRE exams at a fairly young age, it is increasingly likely that men will be diagnosed with cancers that would never have affected them in their lifetimes and would not have been noticed except for the screening. I'm expecting that many more younger men will be screened because the prestigious and influential American Urological Association just revised its guideline to recommend that all men be screened by age 40, with even earlier screening for men at high risk, such as by having a couple or more "first-degree" relatives (like fathers, brothers) who were diagnosed with prostate cancer, especially in their 50s or earlier. I'm expecting that an increasing number of men with very low risk prostate cancer will be diagnosed in their early fifties, forties or even thirties.
I got some insight into this issue when I attended the IMPaCT Conference ("Innovative Minds In Prostate Cancer Today"), held in Atlanta in September 2007. Attendees were invited from among the researchers whose research projects over the past ten years had been supported by funds from the Congressionally Directed Medical Research Program (US tax dollars) managed by the Department of Defense for the Government, from among survivor representatives (like me) who had helped evaluate proposals for at least two years, and from prominent prostate cancer scientists and physicians. One panel discussion included three representatives from major AS programs, Dr. Laurence Klotz from the U. of Toronto (Sunnybrook), Dr. Fritz Schröder from the Erasmus University Medical Center (Netherlands), and Dr. Christopher Warlick, recently from the Johns Hopkins program, but having moved to the University of Minnesota. After each commented on their AS program, I got to ask what minimum age they looked for in their patients, mentioning that I had read that Dr. Schröder's program was comfortable with men as young as age 55. The question stirred a lively discussion among the panelists.
I thought that the Johns Hopkins program under Dr. Carter focused on patients in their mid-60s or older, but Dr. Warlick said that 44% of their program participants were younger than age 65, with the age ranging from 46 to 82, with an average age (median: half older, half younger) of 66. My impression is that most of the major AS programs have preferences for age that are similar to the Johns Hopkins program, targeting men in their 60s, especially latter 60s, and older.
Dr. Schröder confirmed that his program had an age minimum of 55. He also said that about 500 patients were now in his AS program.
However, Dr. Laurence Klotz, from the University of Toronto, Sunnybrook, really caught the audience's attention when he stated that he was comfortable with prostate cancer patients of any age for AS, but adding that the Toronto program takes an even more detailed look at risk factors and does even more observant monitoring for younger men, at least at in the patient's early years in the program.
Of course I cannot speak for Dr. Klotz, but, having heard him on the panel, having listened to and studied one one of his talks, and having read some of his papers, I can give an informed estimate of what his program would do with two young example patients, both 37 years old, who would be considered to have apparently low-risk prostate cancer based on the usual key combination of PSA, stage and Gleason Score. Let's say both got early prostate screening exams because of a mild concern with family history of prostate cancer, and based on some concern because of the DREs, coupled with their PSAs which were elevated compared to norms for their ages, both were given biopsies. Both patients have a PSA of 2, have Gleason Scores of 6 (3+3) confirmed by expert pathologists (with no other concerns in the biopsies), and are Stage 2a (can be felt in the prostate, but only in half a lobe or less). Patient A has a PSA that increased from 1.8 in a baseline test, to 1.9 a year later, and 2.0 the next year. Only one of 12 cores was positive, with cancer in only 5% of the core. His free PSA reading was 20%. Patient B has a PSA that increased from .4 in a baseline test to 1.0 a year later, and to 2.0 a year later; 3 of his 12 cores were positive for cancer, with percentages of cancer of 25%, 40% and 45% in the three cores; his free PSA test value was 16%.
Careful patient selection is key to all AS programs, but especially for younger patients in the Toronto program as described by Dr. Klotz. It's clear to me that Patient A is very low risk in all key case characteristics, but that is not at all true of Patient B. Consider the PSA doubling time PSADT(velocity). Patient A has a favorable encouraging PSADT of 13.17 years - think about it: it's going to take his PSA 13.17 years to double! - (158 months), or .1 ng/mL of PSA per year. Assuming the rate stays constant, as is typical, starting from a PSA of 2.0 at age 37, he will be 50 before his PSA reaches 4.0, about 63 before it reaches 8.0, 76 before it reaches 16.0, and 89 before it reaches 32.0. (Twenty percent of patients in the Toronto program have PSADTs of 100 years or more, as I recall the figure! ) In sharp contrast, Patient B has a rapid doubling time of just 10.34 months! When he is still 37 it could have doubled to 4.0, climbed to 8.0 when he is 38, to 16 when he is still 38, to 32 when he is just 39 (60 years earlier than for Patient A), and so on. And let's not forget the biopsy cores: even though just 3 cores are positive for Patient B, none exceeding 50% cancer in any core, which are numbers that look decent for the average low-risk patient, that is not such a good picture for a very young patient. The free PSA percentages, used in the Johns Hopkins program, might also help project the courses for these patients. Patient A has a free PSA that is close to that for men with healthy prostates; in contrast, while Patient B, at 16%, is above the 15% threshold that triggers concern at Johns Hopkins, he is close to that threshold.
As I understand it, the Toronto program would probably be comfortable with Patient A as an Active Surveillance patient; in accordance with their policies, he would be given extra monitoring in the early years, compared to that given to older AS patients, to make sure his prostate cancer was behaving in a way that lined up with the early favorable clues. In sharp contrast, I believe the Toronto program would strongly recommend treatment in the near future to Patient B. In fact, unless an older patient like Patient B, with his fairly rapid PSADT, had a relatively short life expectancy, I suspect the Toronto program would recommend treatment to him, too.
Dr. Klotz makes the reasonable and generally accepted point that careful screening of men at relatively young ages, followed by efficiently done 12 core (or more biopsies), is going to pick up some small cancers that some of these men would not have noticed for their whole long lifetimes if they had not had screening. However, he also maintains the position that we now have the ability to actively surveil these cancers and provide timely, highly effective treatment if they become aggressive. The Toronto program is finding that the chance that prostate cancer will suddenly become very aggressive for truly low-risk men is extremely remote, as I understand it. The program is also finding that many men are enjoying the benefits of AS for long years. All that makes good sense to me, but I'm sure many of us hope his program will do a special report on how well younger men have fared.
While all the heads of the major AS programs are highly respected, well-known physicians and medical leaders, I've heard several expert doctors say that Dr. Klotz is arguably the leading expert in AS. For me that makes his position regarding the minimum age carry extra weight.
Jim
The Following User Says Thank You to IADT3since2000 For This Useful Post: PatrickMBrennan (02-21-2011)
Re: Active Surveillance for Prostate Cancer: Age of the Patient
Another excellent body of work, Jim. The way I summarize this into a short sentence, in my own words, is that the individual’s clinical metrics, along with their personal, multi-dimensional circumstances of age, health, etc., all weigh into the entry and exit for AS. Here’s the one thing that gave me pause, however:
Quote:
Originally Posted by IADT3since2000
Assuming the rate [PSADT] stays constant, as is typical, starting from a PSA of 2.0 at age 37, he will be 50 before his PSA reaches 4.0, about 63 before it reaches 8.0, 76 before it reaches 16.0, and 89 before it reaches 32.0....
In the spirit of advancing the dialogue, I would put forward a comment that the critical assumption that PSADT is constant is only a good approximation within limited bounds—early stage development of PC.
I reference the report “Observations on the doubling time of prostate cancer. The use of serial prostate-specific antigen in patients with untreated disease as a measure of increasing cancer volume.” (PubMed) This report, and others, conclude that the increase of serum PSA over time is exponential (log-linear) in the vast majority of cases.
This passes the common sense test with my own general layman’s understanding of cancer cell growth: one cell divides into two, which divide into four, which divide into eight, which divide into sixteen…after a while, the tumor cell population starts getting significant. I would add for completeness that there is a strong concurrence that the growth rate of PC cells is rapid compared to normal cell lifecycle, but slow compared to other cancers…but it is not linear.
An exponential curve (or one of constant % increase) has an early stage where a simplified linear estimate can be a close approximation, but it is followed by an exponential (geometric) phase of increase. Google the phrase “exponential growth” to easily find an example of linear and exponential growth lines graphed side-by-side. The first (likely) search result, from a widely-used (albeit non-scholarly) site beginning with "W", has a good example...the board rules prohibit me from saying the name of this site. (By the way, for technical completeness it should be noted that an exponential growth curve would appear linear when plotted on log-linear scale.)
A lot of people take two PSA data points, draw a line and extrapolate linearly...especially if they are young and only have one, and then two PSA scores to go by. Again, within a limited range, this can be a reasonable approximation. However, I would also note that lots of factors may impact that data validity, not the least of which is the close proximity of the two points (1 year, for example) when extrapolating out multiple decades. Also, even a relatively small 5% error in either or both data points (due to inter- or intra-assay variation, physiologic variation, or imprecise measurements at low PSA values) could have a big impact when extrapolating over decades.
So, perhaps the assumption that a 37-year old's PSA increase remains approximately linear until he is 89 or beyond might be a stretch.
Pragmatically, with this understanding of PSADT curves, what I would like to see more written about is the exit criteria—when is the AS PC patient advised to exit the AS strategy to pursue more aggressive treatment options—for AS programs that enlist men at any age (such as the program referenced). How well has the body of knowledge progressed to confidently assess when the exponential growth inflection point has passed, before the localized PC has become systemic (spread to lymph nodes, or beyond)? It seems to me that age remains a critical factor (although still only one in the set of multi-dimensional circumstances) at least until this “exit criteria” is more precisely understood and documented. You did mention, Jim, that the Toronto program (the “any age” program) “...does even more observant monitoring for younger men.” So age is a factor, but what does “more observant monitoring” mean?
I’ll conclude with something we can probably all vociferously agree upon…there are no “blanket statements” which apply.
Thanks, again, for your excellent work gathering meaningful information, Jim.
Re: Active Surveillance for Prostate Cancer: Age of the Patient
Thanks to both of you for an erudite summary of the Canadien and other leading A.S. programs from Jim and a very well presented critique of the perils of linear extrapolation for younger men by KCon. The error potential of small deviations between assumptions and experience in any exponential extrapoplation is analagous to the effect of minute differences in the trajectory of a long range missile or the path of a comet.
However both analyses failed to address the more subjective issues which to my mind are being bypassed completely by the advocates for A.S.programs:
1. The Canadian program is being run and funded by a country committed to a single payer government run system where managing health care expenditures is an equal if not paramount objective. The costs of health care practically drive the Canadian economy.
2. Using chronolgical rather than the less objective but more valid physiological age as basis for entry and measuring points, discriminates against older men with longer life expectancies than chronologically younger men in poorer overall health and physical condition.
I question Jim's statement that its possible to predict that a low risk case won't become clinically significant for thirty years. I wonder whether anyone can make that call. As far as I know Dr. Scardino of Sloan-Kettering still uses the rule of thumb that there's a 50% probability of P.C. metastasizing when PSA reaches or exceeds 10 ng/ml.
My fundamental skepticism of scientific advocates has me looking behind the published studies promoting AS for underlying agendas i.e. economic in the Canadian example or in J.H.'s case, consistent with their well known age based patient selection bias.
Deferring treatment is a completely reasonable course for the truly older(physiologically as well as chronologically) and otherwise health compromised patient where the risk obviously outweighs the potential benefit of aggressive treatment. However for the younger healthier population it simply delays the inevitable for a few years and possibly to a time when surgery or other primary treatment will no longer be as effective or uncomplicated.
Utilizing one's curative window is in my opinion too precious an opportunity to gamble on the odds derived from studies whose conclusions could be challenged or overturned tomorrow.
Re: Active Surveillance for Prostate Cancer: Age of the Patient
Quote:
Originally Posted by shs50
2. Using chronolgical rather than the less objective but more valid physiological age as basis for entry and measuring points, discriminates against older men with longer life expectancies than chronologically younger men in poorer overall health and physical condition.
I think that on this point we are all saying essentially the same thing...
shs50 calls it: "physiological age" versus "chronological age" (leaving off the clinical staging, but probably meaning to also include that)
I phrased it this way: "the individual’s clinical metrics, along with their personal, multi-dimensional circumstances of age, health, etc"
Jim gave multiple hypothetical scenarios of various men, different ages, different health situations and different clinical assessment situations
A lot of this discussion is focused on the entry criteria. If you don't mind, I'd like to redirect the discussion back to my question about the exit criteria.
Restating for clarity: Some men might enter into an AS program, and never have to do anything more aggressive as they might outlive any significant problems that the PC might cause for them. However, others might enter an AS program, and then sometime later in life during the "surveillance" they find things might get worse (the exponential growth phase, perhaps), and then they might have to exit the AS program to pursue a different option, a curative option that they might have "preserved" (and may in fact have improved in it's state-of-the-art from when they first were diagnosed with PC). What is the exit criteria? shs50 alludes to one possible criteria from JH of PSA>10. Anyone with any experience/knowledge in this space?
Again, not an option for me, but simply using this discussion thread to learn more.
Re: Active Surveillance for Prostate Cancer: Age of the Patient
Hi kcon,
One of the great things about this board is that the back-and-forth exchanges improve what each of us knows. I certainly concur in much of what you have in this post, but there is an important point to clarify: that increasing PSA pattern that I use an example, with a constant PSA doubling time (PSADT) of 13 months, actually is exponential. I'll explain below in green.
By the way, I actually ran down the same mental track you did after I had posted after 11 PM and turned in for the night. Before going to sleep, I had one of those "oh shoot, did I get it wrong" moments , as I thought about that constant, linearly increasing PSADT of 13 months, knowing that the pattern of growth for the cancer and associated PSA is exponential instead of linear. Then I realized I had it right and went to sleep. There was a time when I could have figured out the exact equation in a few minutes, but that time was many years ago. Jim
Quote:
Originally Posted by kcon
.... Here’s the one thing that gave me pause, however:
In the spirit of advancing the dialogue, I would put forward a comment that the critical assumption that PSADT is constant is only a good approximation within limited bounds—early stage development of PC.
The way PSADT works is that, for the same amount of time that passes (13 months in the example), the cancer grows and its associated PSA increases at a remarkably steady exponential rate. As a convenient reference point, doctors and researchers use the time it takes the PSA to double. With that understanding - constant time interval but exponentially increasing cancer and PSA between each date, the PSADT is actually helpful throughout our prostate cancer journeys.
I reference the report “Observations on the doubling time of prostate cancer. The use of serial prostate-specific antigen in patients with untreated disease as a measure of increasing cancer volume.” (PubMed) This report, and others, conclude that the increase of serum PSA over time is exponential (log-linear) in the vast majority of cases.
Thanks for citing this interesting report by prostate cancer and PSA pioneer Dr. Stamey and his pathologist McNeal from Stanford U., and the other lead author whom I have not heard of. It's an old report but with some excellent information.
This passes the common sense test with my own general layman’s understanding of cancer cell growth: one cell divides into two, which divide into four, which divide into eight, which divide into sixteen…after a while, the tumor cell population starts getting significant. I would add ... that the growth rate of PC cells is rapid compared to normal cell lifecycle, but slow compared to other cancers…but it is not linear.
If you check the example again, you can see that, while the PSADT is 13 months, the amount of PSA increase in each 13 month time period is growing swiftly. During the first period, it increases by just 1 unit, from a PSA of 1 to a PSA of 2. But in the 13 months between months 39 and 52, for example, the PSA goes from 8 to 16, an increase of 8 of course. The next increase is 16, the one after that 32, and so on - clearly exponential and definitely not linear.
An exponential curve (or one of constant % increase) has an early stage where a simplified linear estimate can be a close approximation, but it is followed by an exponential (geometric) phase of increase....So, perhaps the assumption that a 37-year old's PSA increase remains approximately linear until he is 89 or beyond might be a stretch.
We are in agreement that the PSADT pattern is exponential, and my understanding is that whatever the PSADT time interval is for doubling of the PSA, it stays remarkably constant for each of us - it's individual, sort of like a fingerprint. That is a generalization - a general rule, but my impression is that it applies widely. At the short end, some patients have a PSA that doubles in just two weeks; at the long end, some patients have doubling times of 100 years or longer. It is also clearly true that treatment, including medications, can change the PSADT, as apparently can some of the things we consume. The wonderful, awesome thing about that is that we can use PSADT as a key indicator of whether treatments and other tactics that we try are effective against the cancer. For example, the exciting UCLA team's research with pomegranate juice - their pilot study - resulted in an average increase in PSADT from around 15 months to over 50 months! I also believe that sometimes mutations in the cancer occur that also affect PSADT.
...You did mention, Jim, that the Toronto program (the “any age” program) “...does even more observant monitoring for younger men.” So age is a factor, but what does “more observant monitoring” mean?
I'm not sure, beyond what I indicated in the example. However, I strongly suspect that Dr. Klotz or other doctors managing patients in the program have talked about that; he's an enthusiastic speaker and writer. If anyone knows, please share that knowledge.