ARTICLE: "PREVALENCE OF PROSTATE CANCER AMONG MEN WITH A PROSTATE-SPECIFIC ANTIGEN LEVEL < 4.0 NG PER MILLILETER"
New England Journal of Medicine, May 27, 2004 Volume 350:22,2239-2246

The article published in the New England Journal of Medicine on May 27, 2004, discusses a subset of patients from the Prostate Cancer Prevention Trial, which was recently concluded. In this study men were randomized to receive either Finasteride or placebo for a total of 7 years. At the end of that study all men underwent a needle biopsy of the prostate.

This article specifically concerns the group of men in the placebo arm of the study who had PSAs less than 4.0 ng/ml. When these men were biopsied approximately 15% of them were found to have prostate cancer and of that group about 15% had cancers that were considered high grade (that is, a Gleason grade that is 7 or higher). A Gleason grade of 7 or higher is often associated with more advanced or more aggressive disease and on this basis the authors have raised questions about whether there needs to be a change in PSA screening technique. They correctly point out that it is possible to have prostate cancer when the PSA is in the range between 0 and 4 ng/ml and also correctly point out that were the level of normal to be reduced to 2.5 ng/ml more cancers would be detected. The problem with the strategy of reducing the PSA threshold from 4 to 2.5 is that many additional men will undergo biopsy who do not have prostate cancer and of the men who have prostate cancer detected many will undergo treatment which could have otherwise been avoided.

The findings presented in this paper are not new findings. Urologists have known for decades that PSA testing is not perfect as a screening tool. Urologists have therefore used the strategy of sequential PSA testing and age-specific reference ranges to try to identify patients who are at increased risk for prostate cancer. This strategy has been very effective. Current biopsy rates in most modern series for prostate cancer show that a man considered at risk has about a 35% probability of having prostate cancer on needle biopsy. As such, PSA screening in the hands of urologists has been a remarkably successful strategy. Mortality rates for prostate cancer have been steadily declining since the mid 1990s and the stage at diagnosis has been likewise improving since the late 1990s.

Our ability to detect prostate cancer early in the course of the disease and to differentiate prostate cancers, which need to be treated from those that do not need to be treated will continue to improve. The use of free and total PSA and the use of complexed PSA will further refine our ability to make the early diagnosis of prostate cancer and spare men who do not need a biopsy from going through the procedure.

Some national organizations are now calling for lowering the threshold value for a normal PSA to 2.5 ng/ml. It may be premature to make that blanket recommendation. Certainly men who are higher risk because of race or family history should undergo PSA testing at an earlier age and should be biopsied at lower PSA values. This is current practice in urology. PSA remains a valuable tool in the early diagnosis of prostate cancer and should be combined with digital rectal examination, a thorough physical examination, and a medical history in identifying patients who are considered at risk for prostate cancer and who should be biopsied.

Improvements in biopsy technique have resulted in fewer and fewer false negative biopsies. It is clear that if we lower the threshold PSA level to a low enough value, all men who can be diagnosed with prostate cancer will be diagnosed this begs the question of whether those men deserve treatment. Improvements in the evaluation of biopsy specimens may also help identify men who may be at risk for the subsequent development of prostate cancer even when current biopsies are considered benign.

Patients should be reassured about the validity of PSA as a valuable tool. It is most accurate when interpreted longitudinally and not as an isolated test. It is most practical when combined with physical examination and medical history and interpreted by the clinician.

Definitions:
Sequential PSA testing: Performing PSA tests over time to evaluate the rate of increase over time. This is the way that PSA velocity is determined.

Age-specific reference ranges: PSA values increase with age. To adjust for the normal PSA associated with aging different "normal" values are assigned according to age groups (Oesterling et al).
Age Group Normal PSA Range
45 - 49 years < 2.5 ng/ml
50 - 59 years < 3.5 ng/ml
60 - 69 years < 4.5 ng/ml
70 - 79 years < 6.5 ng/ml

Free and total PSA: PSA exists in two forms: one is bound to a carrier molecule and the other is "free" in the serum or blood. The ratio of the free form to the bound form is the %free PSA. A free PSA of < 15% indicates an increased risk of prostate cancer and a free PSA or > 25% suggests a lower risk of cancer.

Complexed PSA: The complexed PSA test selectively measures the complexed form of the PSA. Some studies have demonstrated that perhaps the complexed PSA test is more specific for detecting elevations in PSA associated with prostate cancer as opposed to elevations in PSA associated with BPH. The total PSA detects both the complexed and noncomplexed forms of PSA.