Yesterday I saw a column in the Wall Street Journal with the headline, “Beginning Prostate-Cancer Screening At Age 40 Holds Benefits, New Data Show.”
The article, written by a reporter whose columns I read regularly and whom I respect for her knowledge and insights, states, “But now there’s new evidence that starting (prostate cancer screening) at 40 could not only catch the worst cancers but could also spare men from unnecessary treatments later in life.”
The reporter notes, “A single PSA score isn’t all that useful, and that what really matters is the rate at which PSA scores change over time.” She goes on to write, “And using ‘PSA velocity’ can help identify men who have slow-growing cancers that don’t need any treatment.”
So now we have further confusion in the already confusing messaging surrounding prostate cancer screening.
Unfortunately, what we don’t have is the evidence to support making a blanket recommendation to men under age 50 that they would benefit from getting early PSA testing.
Prostate cancer is an interesting disease. What many people don’t know is that it actually can begin developing in men early in life, and if you live long enough you are almost certain to have evidence of prostate cancer in your prostate gland when you die.
Many men have prostate cancer. Fortunately, a much smaller percentage of them die from the disease.
What we have known for many years—at least I must say that I was aware of it way back when I was practicing internal medicine in Baltimore—was that a single PSA test was helpful, but serial testing over several years was more informative.
The PSA tests in those days were imperfect, and the results would differ from test to test and lab to lab, but doctors were aware that it was not just the level of the PSA test that was important, but the rate of change as well.
For some reason, that knowledge does not appear to have made a significant impact on how we have been interpreting PSA tests. There has been a substantial emphasis on a single value on a single test. If the PSA is over 4.0, then doctors and patients become concerned, and if it is less than 4 they feel safe.
But much recent research has shown that you can definitely have prostate cancer even if your PSA is below 4.
The real question is what is the “right” PSA level where a doctor should recommend or not recommend a further workup for prostate cancer, including possible “blind” biopsies of the prostate when no evidence of a mass is seen on ultrasound. And, how many cancers will be diagnosed at these lower test thresholds that in fact would never have caused any harm?
The headline quoted above was based on a research report that appeared in a recent issue of the Journal of the National Cancer Institute.
The authors of that report took a look back at men who participated in a study that began in Baltimore in 1958. This study, called the Baltimore Longitudinal Study of Aging (BLSA) was designed to follow people for many years and monitor their health and mental well being to see how changes occurred in their bodies over time.
The research report looked at 1201 men who had serial PSA measurements beginning in 1991, or where blood samples had been frozen and were available for PSA testing (PSA testing was not available in the study prior to September 1991).
These men were then classified, based on what happened to them, as having no evidence of prostate cancer, alive with prostate cancer or having died from a cause other than prostate cancer, or as having died from prostate cancer.
What the researchers found was that 10-15 years prior to the diagnosis of prostate cancer, the PSA levels of the men who died from prostate cancer had been rising at a faster rate than men who didn’t have prostate cancer or who died from other causes.
The term that is used for measuring the rate of change in PSA values from year to year or test to test is “velocity.” So, if a man had a greater change in the PSA test year after year than other men, his PSA velocity was high. If the rate of change was slower, he had a slower velocity. And, of course, if there was no change, there was no “velocity” to measure.
If the PSA velocity was 1, then the risk of death from prostate cancer was increased about 4 times. For example, if the PSA test one year was 4, and the next year 5, the PSA velocity was 1.
When you look at the diagrams in the report, it is clear that men who died with prostate cancer had a much more rapid climb in their PSA levels than men who were alive with prostate cancer, and both of these groups had a more rapid annual increase in PSA than men who did not have any evidence of prostate cancer.
But there was no single velocity value that could absolutely distinguish between men who had lethal prostate cancer from those who did not.
The net result is that if you measure PSA change in younger men, you might pick up some who have a significant change in their PSA, but whose PSA test remains below 4—which is generally considered the upper limit of normal.
For example, a young man could have a PSA of 1. Two years later, the PSA could be 3. That would be a velocity of 1 (3-1=2 divided by 2=1).
In this case, although the PSA test would still be “normal,” the velocity would suggest the possibility that prostate cancer might be present.
So why not start screening everyone at age 40 based on this report?
As the authors note, this was an unselected group of men. That means it was a retrospective look at some of the men in the BLAS, but not all. When the BLAS was designed in 1958, PSA testing wasn’t even on the radar screen.
Second, there is no “right number” for PSA velocity that this study says will reasonably distinguish between benign and malignant conditions, or for that matter between aggressive and non-aggressive prostate cancer.
However, it is also appropriate to note that the authors point out that when the prostate gland is small (as men age, their prostate usually increases in size) even small changes in PSA velocity likely have significance. Benign enlargement of the prostate also causes increases in PSA, but this is usually a problem for older men.
The authors do discuss the possibility of prostate screening with PSA tests beginning earlier in life, but they also go on to say, “We cannot be certain that if a life-threatening prostate cancer had been identified by PSA velocity earlier in the natural history of the disease, treatment at that point in time would have changed the outcome.”
And that is the crux of the discussion.
Before we go out and make a blanket recommendation that every man begin testing at age 40—even if that testing is periodic every couple of years and not annual—we need to have a better handle on the risks and benefits of that testing process.
How many unnecessary biopsies will be performed at what personal cost? How significant are the harms? How many lives will be saved? Will society be benefited in a significant way by recommending this type of test?
This study—as important as it is in raising once again the awareness that changing PSA values over time are an important component of PSA testing—does not provide the answers to these questions.
The editorial that accompanied the JNCI report points out how complex it is to make a diagnosis, and how important it is to understand the analytic tools that help the doctor reach that diagnosis.
But the author of the editorial also hones in on the key issue with respect to prostate cancer.
He writes, “PSA screening for prostate cancer has been a hope ever since introduction of the test in the mid-1980s. To date, no solid evidence exists to confirm the realization of this hope, but at least two large studies to address the efficacy of PSA screening are in progress.”
The editorialist goes on to point out that the report is “by no means definitive regarding PSA velocity, nor is it without its weaknesses, some of which (the authors) address in their discussion.”
He concludes, “In the meantime, screening for prostate cancer by using PSA is not a practice grounded in evidence in the literature and should not be undertaken on the basis of the results of this or previous studies not based on randomized comparison of mortality and morbidity.”
The bottom line is that although this study is interesting and the findings worthy of consideration as we care for our patients, it is clearly not a study upon which we should be changing our current recommendations for prostate cancer screening.
Medicine has been bedeviled for years by the fact that much of what we have recommended as treatment has not been grounded in sound science.
The result has been that we have a lot of “because I say it so, it is so” as part of medical treatment. The problem is that experience has taught us that some of those treatments have been found not to be effective once subjected to scientific scrutiny.
Going forward, many doctors, scientists and professors believe we have an obligation to provide our patients with care based on good science that leads to good care.
I am not saying that this report was not good science. For the most part, it is.
But it is far from sufficient to warrant a significant change in what we recommend to our patients when they ask us what to do regarding prostate cancer screening.
We are a long way from the headline that suggests the benefits of early testing for prostate cancer have been proven. And we are a long way from saying that treatment of a particular man’s prostate cancer can be directed solely by determining the PSA velocity.
Simply stated, we just don’t have the evidence we need to make such recommendations.
+++++++
Personal note:
Some of my friends and colleagues have noted that the frequency of my entries has decreased recently.
Some of that has to do with a dearth of significant news to report. But the greater factor has been a very busy travel and meeting schedule over the past couple of months.
I apologize for the decrease in the number of postings, and will try to do better. But, I suspect it will be January before I am able to get back on track. Until then, I will do my best to keep entries current.
Thanks for reading, and thanks for your understanding.