PSA Testing and Screening

Limitations in the use of serum prostate specific antigen levels to monitor patients after treatment for prostate cancer

Journal of Urology 173(5):1456-1462, May 2005
Frank A Vicini, Carlos Vargas, Anthony Abener, Larry Kestin, Eric Horwitz, Alvaro Martinez

Study purpose: This study reviews the medical literature to clarify the benefits and/or hazards of monitoring serum prostate specific antigen (PSA) levels after surgery or radiation therapy for non-metastasized prostate cancer (cancer that hasn’t spread outside the prostate). Prostate-specific antigen (PSA) is a protein made by the cells of the prostate gland and by prostate cancer cells. The PSA test for prostate cancer measures the level of PSA in the blood. The higher a man’s PSA level, the more likely it is that cancer is present.

Study description: Internet medical databases were searched from 1990-2004 for studies that used PSA tests to evaluate treatments for prostate cancer. The studies were reviewed to see 1) if regular PSA testing gives a true picture of the success or failure of a treatment; 2) if patterns in PSA results after treatment can tell apart therapies that knock out only some areas of disease as opposed to completely unsuccessful ones; 3) whether learning sooner that radiation therapy has failed and going to other treatments earlier helps patients.

Findings: Men with PSA levels that are much lower after radiation therapy are often thought to being on their way to a cure. However, this study indicates that 5% to 25% of patients who initially had excellent results become as ill as they originally were after a five-year period. No PSA patterns after treatment have been linked to a cancer coming back (recurrence).

Conclusions: A rising PSA level after radiation therapy seems to predict the failure of the treatment, often 6-18 months before that fact is otherwise evident.

Note: Although high PSA levels after treatment may show the presence of cancer, the benefit of getting prompt new treatment in response remains unproven. There is little information on the effects of such common treatments as hormone therapy, radiation therapy, and surgery. The overall benefits of monitoring serum PSA after treatment remain controversial.

Meta-Analysis of Prostate-Specific Antigen and Digital Rectal Examination as Screening Tests for Prostate Carcinoma

the Journal of the American Board of Family Practice 16:95-101, 2003
Kishor Mistry and Greg Cable

Study purpose: Physicians check men for prostate cancer using digital rectal examination and serum prostate-specific antigen (PSA) testing. The usefulness of such “screening” methods is not well established, however. This meta-analysis looked at 13 studies to see how well screening methods detect prostate cancer (a meta-analysis is a statistical method that combines study results in order to re-interpret overall findings).

Study description: Studies were selected from the medical literature: most contained men with no symptoms over the age of 50. The meta-analysis focused on a screening method’s sensitivity (how often it picks out actual cancer), its specificity (its ability to recognize that cancer is present or absent), and its accuracy in making predictions, known as the “positive predictive value”. Overall results from the 13 studies were calculated from their individual findings for sensitivity, specificity, and positive predictive value. Biopsy results were used as a comparative standard for results from PSA tests and digital rectal exams.

Findings: The screening methods’ overall detection rate of prostate cancer was 1.8% based on a positive biopsy. Of the prostate cancer detected, 83.4% was contained within the prostate gland. Pooled results from all the studies showed that the sensitivity and specificity of PSA testing was 72.1% and 93.2% respectively. The positive predictive value for PSA was 25.1%. The digital rectal exam had a pooled sensitivity rate of 53.2%, a specificity rate of 83.6%, and a positive predictive value of 17.8%.

Conclusions: The meta-analysis had two major outcomes. One was the potential for detecting early-stage prostate cancer with both of the screening methods studied — 83.4% of the total prostate cancers detected were still localized within the prostate gland. The second was evidence that PSA testing outperforms digital rectal examination when used as a screening tool. The PSA test rated higher in overall sensitivity, specificity, and ability to make accurate predictions.

Note: When a man has abnormal findings using PSA and digital rectal examination, the chance of cancer is one in four or five. However, when findings from both are normal, the chance of missing a cancer remains about 10%.