english.prescrire.org > Spotlight > Archives : 2009 > Routine screening for prostate cancer is not justified

Spotlight: Archives

Every month, the subjects in Prescrire’s Spotlight.

2009 : 1 | 30 | 60 | 90

Routine screening for prostate cancer is not justified

In 2009, the value of routine screening for prostate cancer has not been clearly proven, but its adverse effects have.

Four comparative trials shed some light on the long-term risk-benefit balance of screening for prostate cancer based on PSA levels, whether or not combined with a rectal examination.

When the level of the prostate-specific antigens (PSA) exceeds 4 ng/ml, the risk of prostate cancer is multiplied by around 1.5 to 3 per 1 ng/ml increase. However, the level of PSA also increases with age, the size of the prostate and in the case of inflammation of the prostate. In practice, in around 70% of cases where the PSA is above 4 ng/ml, no cancer is detected. But the number of false positives is a source of concern, resulting in repetition of PSA tests and in prostate biopsies, which can cause blood in the urine and in sperm, fever, pain and inflammation.

Around 70% of diagnoses err on the side of excess, as they detect cancers which would not have had any clinical consequences during the patients' lifetime. These patients are then exposed unnecessarily to the adverse effects of treatment (surgery or radiotherapy): urinary incontinence, erectile dysfunction, rectal inflammation from the radiation treatment.

Adding a rectal examination to measuring PSA levels increases the number of cancers detected by a quarter to a third, but also increases the number of unnecessary biopsies by around two thirds.

While awaiting the final results of two major trials whose interim results, published at the beginning of 2009, are not convincing, it appears, as of May 2009, that routine screening for prostate cancer using the PSA count and/or rectal examination is not justified.

©Prescrire November 2009
 
"PSA-based screening for localised prostate cancer" Prescrire Int 2009; 18 (103): 225.

Download the full review