New Analysis Confirms Suspected Bias of Large U.S. Prostate Screening Study

A new paper appearing in Annals of Internal Medicine used simulation modeling to try to correct for a known bias in a notable U.S. trial of prostate cancer screening, called PLCO. That study along with a large European one played an important role in the creation of screening guidelines from the American Cancer Society and others.

While the major European trial, called ERSPC, suggested routine PSA screening reduced the risk of prostate cancer death by about 30 percent in men who were very compliant with screening and treatment, the U.S. study failed to show a benefit. The U.S. trial is widely-believed to have been contaminated; many men in the control arm, who were assumed to be unscreened, were in fact screened thanks to the wide availability of PSA testing at the time, diminishing the differences between the studies two arms.

We asked Chief Medical Officer Otis W. Brawley, M.D. if the new data changes the current understanding of PSA screening.

“This new analysis supports the widespread understanding that PSA screening under optimal conditions has larger mortality benefits than were shown by a large U.S. trial. It also supports a move towards watchful waiting that all experts hope will make screening as useful as possible.

“The new analysis uses a measure known as ‘mean lead time’ to try to account for suspected biases in the U.S.-based PLCO trial. Some consider this modeling controversial and unproven. Nonetheless, it may be useful in this situation.

“The findings from the new analysis are consistent with the assessments of most experts, who generally agree that prostate cancer screening is associated with a reduction in prostate cancer deaths, and that the U.S. trial undercounted this benefit.

“Screening recommendations from the American Cancer Society and others have acknowledged the limitation of the U.S. trial, so this new analysis does not significantly change the body of evidence on which our recommendations are based.


Otis W. Brawley, M.D., MACP

“A reduction in the risk of death is only half the equation when weighing whether to recommend screening. Guidelines from the ACS and others must weigh those benefits against some fairly serious side effects associated with screening and its subsequent treatment. The question is whether the benefit of widespread routine screening outweighs the risks of harm.

“The ACS was one of the first of many organizations that now recommend informed or shared decision making regarding prostate cancer screening, based on the fact that the potential benefits of screening can be offset by harms associated with the large number men who would be treated.

“Most organizations including the ACS now agree men in their 50s should be informed by their clinician regarding the potential benefits and risks of screening. This analysis will not change that recommendation. 

“Prostate cancer screening has been an area in need of clarity. This study validates the findings from the two major studies that prostate cancer screening does save lives.

“Those two studies also showed that many men diagnosed with prostate cancer through screening are candidates for observation rather than immediate, aggressive therapy.

“This approach, called watchful waiting, has lessened the harms associated with detecting prostate cancer through screening and has begun to change the risk benefit ratio.

“It is hoped that continuing the trend toward carefully selecting which men get screened for prostate cancer, developing better methods for distinguishing low and high risk prostate cancers, and carefully monitoring those who are found to have low risk cancer has the potential to optimize PSA screening to leverage its lifesaving potential while reducing harms of unnecessary treatment.”

For more information, see Can Prostate Cancer Be Found Early? on

About David Sampson

I am the director of medical and scientific communications for the American Cancer Society national home office.
This entry was posted in Guidelines, Prostate, Research, Screening and tagged , , , , . Bookmark the permalink.

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s