March 25, 2009

The recent publication of two articles on prostate-cancer screening in the March 2009 issue of the New England Journal of Medicine has spurred renewed controversy over the place of the PSA test   in promoting men’s health.  This blood test checks for minute quantities of an enzyme called prostate-specific antigen or PSA. Higher than normal amounts of PSA in the blood is a possible indicator of prostate cancer or other prostate conditions, such as benign prostatic hyperplasia (BPH), and often triggers further medical tests.

The articles present the first reports from two large, randomized trials studying the relationship between PSA-based screening and prostate-cancer mortality: The European Randomized Study of Screening for Prostate Cancer  and the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial.  Both studies were initiated to answer a persistent question: Does regular PSA testing of men showing no signs of prostate cancer reduce the rate of death from prostate cancer?

According to the European study, which involved over 162,000 men between the ages of 50 and 74 in seven countries, PSA-based screening reduced the rate of death from prostate cancer by 20 % but was associated with a high risk of overdiagnosis and overtreatment.

The American PLCO trial found the rate of death from prostate cancer was very low for both the 38,343 men in the group that received annual PSA-based screening and the 38,350 men in the control group who received “usual care.” The conclusion: “Screening was associated with no reduction in prostate-cancer mortality.”

In the end, both studies raise more questions than they answer. All that can be concluded is that the studies will stimulate vigorous debate within the medical profession and the prostate health advocacy community.

Indeed, this debate is already raging. But our policy on early detection using the PSA test remains the same.

Within the framework of the recent debate, we take the following positions:

•  A man has the right to know whether he might be at risk from prostate cancer, a disease that kills an estimated 4,300 Canadian men annually.
• The PSA test is safe, no more risky than any other blood test. So PSA testing is not harmful in and of itself.
• Currently, the PSA test is one of the best methods of early detection available. Refinements of the simple PSA test, such as measurements of PSA velocity, PSA doubling time, and percentage of free to total PSA, have improved the use we are able to make of its results. So getting a higher than normal PSA reading need not necessarily lead to follow-up procedures or treatments that pose risk, but should spark a serious discussion between a man and his doctor.
• “Overdiagnosis” is a term that is potentially confusing; one either has prostate cancer or one does not. The question should be how best to respond to individual men who are diagnosed with prostate cancer rather than whether or not men should have access to information that might lead to a diagnosis.
• The question of whether to offer asymptomatic men the PSA test should be divorced from the issue of the overtreatment of prostate cancer.  Not all prostate cancers need treatment. But we should use what tools we have to discover prostate cancer early, so it can be treated if necessary.
• PSA test results are used consistently in nomograms designed to help physicians and patients decide which prostate cancer therapies will result in the greatest benefit. Although we need new and more accurate ways of distinguishing between aggressive prostate cancer and more indolent forms of the disease, today, PSA levels are used to help make this determination.  A baseline PSA reading at age 40 can prove useful to men as they age.

More About the Two Studies on Prostate-Cancer Screening

Neither the European Randomized Study of Screening for Prostate Cancer (ERSPC) nor the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (PLCO) is complete. So Dr. Michael J. Barry, principal investigator of the Prostate Patient Outcomes Research Team (PORT-II) and Harvard Medical School professor wonders why results were published in 2009. “Both decisions to publish now can be criticized as premature,” he writes, especially considering that the studies offer conflicting evidence about the possibility that regular PSA-based screening reduces prostate-cancer mortality, “leaving clinicians and patients to deal with the ambiguity.” ¹

It is no wonder that responses to this research have been confusing.  Health and science editors in the UK tend to focus on the ERSPC study. Headlines such as these are the norm: “Prostate Cancer Screening Could Cut Deaths by 20%”  (Guardian) and “Prostate screening to be reviewed: Routine prostate cancer screening could cut death rates from the disease by 20%, a major study suggests) (BBC). In the States, with the focus on the American study, reports are less optimistic. “Prostate Test Found to Save Few Lives,” reads a March 18th headline in New York Times, and CNN tell us that “Prostate screenings don't reduce cancer deaths.” Canadian coverage is somewhere in the middle. “Landmark Study Stokes PSA Testing Debate” announces the Globe and Mail. And the Calgary Herald, which quoted president Bob Shiell, had a very balanced headline: “New studies on prostate tests revive debate: Cancer survival rates varied in U. S., European research.”

So how do the studies compare? One of the most accurate and easily understood comparisons is a video of a roundtable discussion between oncologist Philip Kantoff and primary care physician Mary McNaughton-Collins. The debate is moderated by Dr. Thomas Lee, an associate editor of the New England Journal of Medicine, and both a video and transcript are made available on the NEJM website.  (Click here for access to a video of this discussion.)

The main points of comparison are outlined below and demonstrate the difficult task medical researchers have when trying to gather evidence of cause and effect in an area as complex as human health.

• The PLCO trial used a higher PSA level to trigger further medical investigation than did the ERSPC trial  (4 ng/mL as opposed to 3 ng/mL, usually, but this varied by European country). So this difference may have increased the number of men diagnosed with and treated for prostate cancer, as well as the number who underwent unnecessary biopsies, in the ERSPC trial.
• Each study recruited subjects at different times and is assessing mortality rates after longer or shorter median follow-up periods. The ERSPC study reported that rates of death from prostate cancer in the screened versus the control group only began to diverge after 7 to 8 years and continued to diverge further over time. To complicate things, although the PLCO trial had a longer median follow-up time, it included fewer men, so its calculations concerning mortality rates are based on 174 prostate cancer deaths as compared with the 540 such deaths among men in the ERSPC trial.   Perhaps the follow-up time has not been long enough to exhibit the survival benefits of early detection. Remember that, according to Dr. Peter Albertsen’s nomogram, approximately 80 % of men diagnosed with low-risk localized prostate cancer are alive after 15 years, even without treatment. Or perhaps the PLCO study did not include sufficient numbers to show an effect on mortality rates. (The ERSPC study reported that 1,410 men would need to be screened and 48 additional cases of prostate cancer would need to be treated to prevent one death from prostate cancer.)
• Each study suffers from methodological challenges. The ERSPC trial did not have a uniform study design because it had to adhere to the policies of several European nations. In the PLCO trial, a major consideration is whether the prevalence of PSA testing in the United States “contaminated” the results. The control group, the group of men who were not screened with annual PSA testing, underwent “usual care” and that, sometimes, meant that they had the PSA test. In fact, in the sixth year of the study, 52 % of the men in the control group had undergone a PSA test as opposed to 85 % of those in the group that was screened. Consider these percentages in light of the fact that about twice as many cases of prostate cancer were diagnosed in the screened arm of the ERSPC study as in the control group while, in the PLCO trial, about 9 % of those in the screened arm were diagnosed as compared with about 7.75 % in the control group.  Dr. Kantoff, in a roundtable discussion broadcast by the New England Journal of Medicine, concludes that these results speak to “less contamination in the ERSPC study than in the PLCO study.”2
• Both studies considered the “risks” or “harms” associated with prostate-cancer screening to be overdiagnosis (the detection of prostate cancer in men in whom it would not have been detected in their lifetimes except by screening) and overtreatment (the aggressive treatment of prostate cancer that is not life threatening). Neither offers a comprehensive assessment of the “benefits” of PSA-based screening compared to its “harms.”

So, until such time as medical research provides a more accurate means of alerting a man to the possibility that he has prostate cancer, we urge all men to remain proactive about their health and to access the tools we have available for early detection.

1. Michael J. Barry, “Screening for Prostate Cancer---The Controversy That Refuses to Die,” New England Journal of Medicine 360, no.13 (2009): 1352.

2.  Perspective Roundtable: Screening for Prostate Cancer, New England Journal of Medicine Online