Prostate Cancer Screening – What Does the Evidence Tell Us?

In May 2012, the United States Preventive Services Task Force (USPSTF) released a recommendation against prostate-specific antigen (PSA) screening for men. The justification for this recommendation was that large randomized trials demonstrated that PSA screening increased detection of prostate cancer, but there was "small to no reduction in prostate cancer-specific mortality." This decision was met with controversy and spurred debate regarding the utility of PSA screening. Let's take a few minutes to review the evidence and also evaluate new data that has become available since the USPSTF recommendation was released.

There are two large randomized controlled trials that were designed to evaluate whether or not PSA screening is beneficial. The Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial (PLCO) Trial was a United States based trial randomizing 76,685 men to PSA screening (once yearly for 6 years) or a control arm. Unfortunately, yearly rates of screening in the control group ranged from 40% to 52%, so ultimately this trial measured the benefit of yearly screening versus less frequent screening; therefore, the ability of this trial to evaluate the benefit of screening (versus no screening) was limited. This trial showed an increase in diagnosis of prostate cancer in the yearly screening arm, but no difference in prostate cancer-specific mortality.

The second major trial of PSA screening was the European Randomized Study of Screening for Prostate Cancer (ERSPC). This trial, published initially in 2009 with over 182,000 men, evaluated screening with PSA once every 4 years versus a control arm. In the initial report, the cumulative incidence of prostate cancer was 8.2% in the screening arm versus 4.8% in the control group. Screened patients were more likely than control patients to have organ-confined (clinical T1-2) disease, non-metastatic disease, and lower Gleason Score (≤ 7) at diagnosis. This translated into a decreased risk of prostate cancer mortality among the screened patients, with a rate ratio of 0.85. In more practical terms, to prevent one prostate cancer death, the number needed to screen (NNS) was 1,410, and the number needed to treat (NNT) was 48.

We know from previous studies such as the SPCG-4, which randomized radical prostatectomy versus watchful waiting for prostate cancer, that the survival curves required 8 years of follow up to demonstrate a treatment benefit. Therefore, it stands to reason that that when screening is randomized (rather than treatment), it takes at least 8 or more years to demonstrate a benefit, and the true magnitude of the benefit will not be known for some time after that.

Indeed, this is what was observed in the ERSPC trial, as the 11-year follow-up study showed the NNS and NNT were 1,055 and 781, respectively, and the 13-year follow-up showed a further reduction of NNS and NNT to 781 and 27 to prevent one prostate cancer death (Figure). It is unknown how much further these numbers will drop with additional years of follow-up, but it seems likely that the observed magnitude of screening benefit to reduce prostate cancer mortality will continue to improve as data from this study continue to mature.

Detractors of prostate cancer screening may correctly point out that both the PLCO and ERSPC trials failed to show an overall survival benefit to screening, and also that screening primarily increases detection of low-risk early prostate cancers which may never have become clinically apparent or harmful to the patient. Increased diagnosis of early cancers may lead to overtreatment which is costly to the patient and healthcare system, and may carry substantial morbidity. On the other hand, proponents of screening argue that screening does reduce prostate cancer mortality and that the true magnitude of the benefit remains unknown, because data from trials takes decades to fully mature.

Critics of the USPSTF recommendation against screening may further point out that for breast cancer, the USPSTF recommends screening for women under age 50 should be an "individual choice" (Grade C recommendation) based on a number needed to invite [to screen] of 1,904. This number is significantly greater than the NNS of 781-1,410 seen in the ERSPC trial, despite the fact that the prostate cancer screening received a Grade "D" recommendation against screening from the USPSTF.

The question of whether or not to perform prostate cancer screening is a classic example of knowledge versus wisdom in medical decision-making. However, we need more information to make the best decisions with our patients. Additional study needs to be done to examine the morbidity and effects of treatment on quality of life, as this will significantly impact the decision to screen and/or treat prostate cancer for many individuals. Ultimately, I think the best solution is a balanced approach. The utility of screening for any disease can be increased if screening is performed among a higher risk population. Thoughtful selection of patients who would benefit most from screening, as well as individualized decision making regarding who, how, and when to treat prostate cancer, will hopefully reduce the tremendous burden that this disease places on thousands of patients every year.