PSA Screening Should Be an Opt-In Test, Not Opt-Out

David Graham, MD


October 19, 2015

In 1991, Dr William Catalona and colleagues[1] published a report in the New England Journal of Medicine describing the use of prostate-specific antigen (PSA) as a screening test for prostate cancer. I think it's easy to say that healthcare in men has not been the same since.

To step back and define things a little better, I think it's important to understand the requirements of a good screening test. It should be easy to perform, be relatively inexpensive, and affect mortality specifically due to the disease being screened for. That the PSA blood test meets the first two criteria is fairly clear. It's the last criterion that has been debated for years.

Two trials were published in the same issue of the New England Journal of Medicine in 2009 and gave conflicting results for screening.[2,3] The trial that did report a benefit of screening also gave the caveat that a number of men were overtreated on the basis of screening.

Even as far back as 1954, Dr L.M. Franks[4] reported that 31% of men older than 50 years were found to have prostate cancer on autopsy. With the advent of PSA, we have become much more likely to find these cancers—which would never have any clinical impact on these men—and then subject patients to treatment that they may never have required. As a result, many men have experienced both short- and long-term toxicities that are often very problematic. In addition, men with a shortened life expectancy owing to other medical issues have continued to be screened. One report has even estimated that the number of these men could be nearly 800,000.[5]

Because of incomplete data, the US Preventive Services Task Force (USPSTF) first put out a recommendation in 2008 stating that PSA screening should be limited to men with at least a 10-year life expectancy. In 2012, the USPSTF expanded their recommendation to state that PSA screening should not be performed, regardless of age.[6] The expansion was based on recognition that the harms of overdetection and treatment outweighed the benefit gained.

There has been significant and often vehement debate regarding these recommendations. The American Urological Association initially argued strongly against the USPSTF recommendations but has since evolved its guidelines to state that PSA screening could be considered in men aged 55-69 years only after significant discussion and "shared decision-making."[7]

Given that most screening is performed by primary care physicians and they most commonly pay attention to the USPSTF, it should be expected that screening rates have fallen. Interestingly enough, a review of PSA screening after the 2008 USPSTF statement did not show an impact on screening of men older than 75 years.[8]

More recently, a review of screening practices since the 2012 USPSTF recommendations was published in the Journal of Clinical Oncology.[9] They reported that screening in men aged 60-74 years declined from 51.2% to 43.6%. The screening rate decreased from 33.2% to 24.8% in men aged 50-59 years, whereas men older than 75 years had screening rates drop from 43.9% to 37.1%.

The concerning data showed that men with an expected survival of 9 years or less were still being screened at a rate of up to 32%. In other words, about 1.4 million men who are older than 65 years, but are expected to live less than 9 years, are being screened with PSA testing.

So the next question has to be, "What do we do with those men who are screened, arguably inappropriately, and found to have prostate cancer?" The notion of active surveillance has been recognized for more than 20 years and certainly has a role in specific men.

Investigators at Johns Hopkins recently reported their experience with active surveillance since 1995 in the Journal of Clinical Oncology.[10] They report on 1298 men with either "very low-risk" (stage T1c, Gleason score ≤ 6, two or fewer positive cores, and a maximum of 50% involvement in any one core) or "low-risk" (stage T2a or less, PSA level less than 10 ng/nL, and Gleason score ≤ 6) disease. They were followed with semiannual PSA testing and rectal exams, and annual biopsies.

They found some very interesting things. Only 49 of the 1289 men died during the follow-up period, and only two of those deaths were from prostate cancer. Men were 19 times more likely to die of cardiovascular disease than prostate cancer. An advancement of grade on the annual biopsies led to a recommendation for intervention. This only occurred in 22% of men in the very low-risk group and 31% of men in the low-risk group at 15 years. Outcomes of those interventions were generally excellent, with only 8% showing biochemical failure.

If you are a primary care physician, PSA screening for prostate cancer should be thought of as an "opt-in" as opposed to an "opt-out" test.

If you are a primary care physician, PSA screening for prostate cancer should be thought of as an "opt-in" as opposed to an "opt-out" test. In other words, you need to establish a significant reason to screen and discuss with the patient what adverse outcomes the screening may lead to. If the patient has an estimated 9 years or less to live, screening should not be an option at all.

If you are faced with a man diagnosed with prostate cancer and he falls into the very low-risk or low-risk group, you should strongly consider active surveillance. You certainly would not be putting him at significant risk, and you would be probably saving him from very real toxicities.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: