PSA screening may or may not reduce deaths due to prostate cancer. Harms include false-positive results, unnecessary biopsies, and overdiagnosis.

Fenton JJ, Weyrich MS, Durbin S, et al. Le dépistage basé sur l'antigène prostatique spécifique pour le cancer de la prostate: rapport sur les données probantes et revue systématique pour la U.S. Preventive Services Task Force. JAMA. 2018;319:1914-31.

Review question

What are the benefits and harms of screening men for prostate cancer using the prostate-specific antigen (PSA) test?

Background

Prostate cancer is one of the most commonly diagnosed cancers in men. Many prostate cancers are slow-growing and are not life-threatening. The PSA test is a blood test used to screen for prostate cancer. It can help detect prostate cancer at an early stage when it may be easier to treat. However, men with a positive PSA test may have a prostate condition other than cancer (false-positive test) or prostate cancer that is unlikely to progress and doesn’t need to be treated (overdiagnosis). Men with positive screening tests often have further, more invasive tests, such as prostate biopsy. These tests can have complications.

How the review was done

Researchers in the USA did a systematic review of studies available up to February 2018. They found 8 studies assessing PSA screening, including 3 randomized controlled trials (RCTs) with 647,906 men who were between 50 and 74 years of age.

The key features of the RCTs were:

• men didn’t have symptoms before screening; and
• screening was compared with no screening.

What the researchers found

Compared with no screening, PSA screening:

• reduced death due to prostate cancer in 1 trial but not in 2 other trials; and
• had no effect on death due to any cause.

No prostate cancer was found in 10% to 18% of men who had a positive PSA test. In men who had a biopsy after a positive PSA test, 61% to 76% did not have prostate cancer.

Interpretation

Why do the 3 trials show different results for reducing death due to prostate cancer?

Differences within the trials may have affected the results. For example, some trials did PSA tests every year and some every 2 to 4 years. Trials also used different levels of PSA to indicate the need for a biopsy. If a lower PSA cutpoint was used, the PSA test would be positive for more people than if a higher PSA cutpoint was used. Some of the trials had problems: Some men in the no-screening groups had a PSA test done when they shouldn’t have (contamination), and some men who were supposed to have a PSA test done did not (poor adherence).

Conclusions

In men without symptoms of prostate cancer, PSA screening may or may not reduce prostate cancer mortality. Harms of PSA screening include false-positive test results, unnecessary biopsies, and overdiagnosis.

PSA screening vs no screening in men without prostate cancer symptoms*

PSA = prostate-specific antigen.

*Based on results of randomized controlled trials.