New Study Looks at Accuracy of Prostate Cancer Screening Tests
Three-quarters of U.S. men over the age of 50 have had a ]]>prostate-specific antigen (PSA) test]]> as part of routine screening for ]]>prostate cancer]]> . An abnormal PSA test often leads to a ]]>biopsy]]> to determine if cancer is present. In recent years, health professionals have questioned whether the PSA test is an effective way of detecting prostate cancer. Does it miss too many cancers? Does it lead to too many unnecessary biopsies? Is there a better way to screen for prostate cancer?
An ideal cancer-screening test detects the most cancers possible, while minimizing the number of people who will mistakenly screen positive. Currently, a PSA test is considered abnormal if the results are higher than 4 ng/ml. Some health professionals suggest that this number is too high, and that lowering it will increase the value of PSA tests.
A new study in the July 24, 2003 issue of the New England Journal of Medicine took a closer look at the accuracy of PSA testing and concluded that lowering the PSA threshold from 4.1 to 2.6 ng/ml in men under 60 would double the cancer-detection rate from 18% to 36%, without substantially increasing the number of false-positive screenings.
About the Study
From May 1995 to November 2001, 6,691 men had a PSA test and a digital rectal examination, both routinely used to screen for prostate cancer. To be included in the study, the men had to be at least 50 years old, unless they had a family history of prostate cancer or were black, in which case the minimum age was 40. Men were excluded from the study if they had undergone prostate biopsy, had been diagnosed with prostate cancer, were using finasteride (a medication used for prostate enlargement), or had an active ]]>urinary tract infection]]> or ]]>prostatitis]]> .
A biopsy was recommended for the men whose PSA values were above 2.5 ng/ml, or if the digital rectal exam indicated a lump suspicious for prostate cancer. Men whose biopsies detected prostate cancer and men who were diagnosed with prostate cancer within 18 months after the PSA test were included in the final count of prostate cancer cases.
From the results of the PSA tests and the diagnoses of prostate cancer, the researchers used a mathematical equation to estimate the number of cancers that would be missed and the number of unnecessary biopsies that would be performed using different PSA thresholds.
Eleven percent (705) of the men in this study underwent a biopsy, and one-quarter (182) of these men were subsequently diagnosed with prostate cancer. In this study, being black and under the age of 60 was significantly associated with prostate cancer, but having a family history of prostate cancer was not.
The researchers calculated that if a PSA test result over 4 ng/ml was considered abnormal in men under 60, 82% of prostate cancers would be missed and 2% of men without prostate cancer would undergo an unnecessary biopsy. If the threshold were 2.6 ng/ml, on the other hand, 64% prostate cancers would be missed and 6% of men without prostate cancers would undergo an unnecessary biopsy. The following table shows the expected results of using different PSA thresholds when testing men under 60:
|PSA threshold for biopsy recommendation (ng/ml)||Estimated percentage of cancers that would be missed||Estimated percentage of unnecessary biopsies|
These results are certainly interesting, but not without their limitations. First, to determine the true accuracy of PSA testing, all men—regardless of whether or not their PSA test was abnormal—would need to undergo biopsy to detect cancer. Because this isn’t practical, these researchers used an equation to estimate the number of cancers that would be missed. If their mathematical assumptions were incorrect, the results would not accurately predict the experience of actual patients. Also, since biopsies can’t detect all cancers, the number of cancers missed by the PSA test may have been underestimated.
How Does This Affect You?
The perfect screening test would identify all patients with the condition and no patients without the condition. Such a screening test does not exist. There are always false positives and false negatives. In this case, a false positive is no trivial matter, since it leads to pointless anxiety and an unnecessary invasive procedure. A false negative, on the other hand, can be equally distressing later when the cancer is discovered at an advanced stage. While the researchers in this study didn’t pinpoint the optimal PSA threshold, they did find that lowering the threshold from 4.1 to 2.6 ng/ml would double the prostate cancer-detection rate, without having a huge effect on the number of unnecessary biopsies—at least in younger men.
So will your doctor soon be adopting a lower threshold? Probably not. The ultimate goal of the PSA test is not to decide who should be biopsied and who should not. It is to save lives. Even though a lower PSA threshold would certainly detect more cancers, it is not yet clear whether detecting these missed cancers will lead to better outcomes among men with prostate cancer. The answer to this question awaits the results of ongoing studies that will be completed over the next few years. Until then, the true benefits of PSA testing at any threshold level will remain unknown.
American Cancer Society
National Cancer Institute
Punglia RS, D’Amico AV, Catalona WJ, Roehl KA, Kuntz KM. Effect of verification bias on screening for prostate cancer by measurement of prostate-specific antigen. New England Journal of Medicine. 2003;349:335-342.
Last reviewed July 31, 2003 by ]]>Richard Glickman-Simon, MD]]>
Please be aware that this information is provided to supplement the care provided by your physician. It is neither intended nor implied to be a substitute for professional medical advice. CALL YOUR HEALTHCARE PROVIDER IMMEDIATELY IF YOU THINK YOU MAY HAVE A MEDICAL EMERGENCY. Always seek the advice of your physician or other qualified health provider prior to starting any new treatment or with any questions you may have regarding a medical condition.
Copyright © 2007 EBSCO Publishing All rights reserved.