There has been a gradual 30% decline in prostate cancer mortality in the U.S. and Canada which started 4-5 years after the introduction of PSA testing. Statistical modelling studies indicate that PSA testing has played a role in this decline. PSA testing has also led to a marked increase in the diagnosis of early cancers that would not be expected to cause any symptoms had they not been detected (‘over diagnosis’).

Recent European randomized trials [5],[6] demonstrated a 44% and 20% relative reduction in prostate cancer deaths among those offered screening when compared to those that were not. In the more mature Göteborg study (14 years median follow-up)[5] the benefit was greater than in the less mature ERSPC study (11 years median).[6] The Göteborg trial estimated that 293 men would need to be offered screening and 12 men diagnosed for prevention of one prostate cancer death over a 14 year period. The ERSPC trial estimates were 936 and 33, respectively, but are expected to reduce as the trial matures further. The NCI-US PLCO study found no difference in prostate cancer deaths at 13 years of follow-up when comparing unscreened men[7]. However, in this underpowered study, 44% of men were pre-screened and screening in the control group was very substantial (52% in the 6th year) which would have masked any impact of screening on mortality. Given the protracted natural history of many prostate cancers, longer follow-up is necessary to address the balance of benefits and harms of screening for prostate cancer. All cause mortality has not been found to be different in any prostate cancer screening trial, which may be expected given the sample size of the trials and duration of follow-up, but also emphasizes the low risk of dying from prostate cancer (in the US PLCO trial, at 13 years there were 303 prostate cancer deaths out of 77,685 men), the long natural history of the disease and the many competing causes of death that aging men face.

Nonetheless it is still unclear whether prostate cancer screening results in more benefit than harm, and thus a thoughtful and balanced approach to PSA testing is critical. The results of the European trials identify a clinically significant benefit from PSA screening, as well as an opportunity to guide prostate cancer screening (and treatment) algorithms in BC. However the use of PSA testing for the early detection of prostate cancer remains controversial owing to the risk for over-diagnosis and over-treatment. The latter may be addressed by uncoupling treatment as an inevitable response to diagnosis. Once this is done, the issue of prostate cancer screening becomes less controversial.

The Canadian Partnership Against Cancer, Screening Action Group recently assessed the current evidence regarding PSA screening and concluded that while expansion of PSA testing practices was not justified and might produce net harm, but there was insufficient evidence to discourage current ‘ad hoc’ PSA testing practices.[8]

At the present time, PSA testing is common (~50% of men over 55yrs have been tested in BC) but is not ideally performed in BC or across Canada. Considerable confusion and misinformation exists amongst public and primary care physicians, leading to inappropriate PSA testing (too young or too old, too frequent, duplication), inappropriate biopsies, leading to over-detection and over-treatment, as well as abnormal PSA values being ignored and not acted upon. Hence, there is a need to refine PSA testing (in whom, and how) with guidelines to systematically manage indications for testing, biopsy, and subsequently treat only those who are likely to benefit from it. Based upon the algorithm that was used in the European trial, we recommend that men who wish to be tested start at age 55, and be tested every 4 years until they reach 70. A result greater than 3 ng/ml should be regarded as abnormal and after being checked, the man should be referred to a Urologist for investigation. There is no role for testing and then ignoring or misinterpreting abnormal results; and likewise no man should be screened unless both he and his primary care physician are in agreement that an abnormal result will lead to further investigation.

It is noted that a baseline PSA level from men in their 40s above the age-adjusted median value is a strong predictor of future risk of prostate cancer; those with a PSA value below the median (0.6 to 0.7ng/mL) are at very low risk while those above the median are at higher risk. [9] At the present time, it is uncertain how and whether to incorporate this information into a PSA testing algorithm.

DRE of the prostate has traditionally played a role in the early detection of prostate cancer. The sensitivity of DRE to detect cancer is less than that of PSA, and the ERSPC trial dropped DRE as part of its screening protocol. Nonetheless some men will develop significant cancers without a rise in PSA, and therefore it may also be considered in those who wish to pursue early detection of prostate cancer.

Taken together, because of the overall potential for a loss of quality-adjusted life years due to over-diagnosis and over-treatment of prostate cancer, the GUTG/VPC does not recommend an unselected, population–wide, PSA screening strategy. The utility analysis however, does support the GUTG/VPC recommended strategy for use of PSA testing for in asymptomatic men who are well informed about the risks of over-diagnosis and over-treatment but still wish to pursue early diagnosis of prostate cancer. In addition, a strategy of early detection of prostate cancer linked to a treatment algorithm that includes prioritization of active surveillance for men with low risk prostate cancer serves to mitigate the potential harms of over-diagnosis, over-treatment and loss of quality of life associated with early detection of prostate cancer by PSA testing.