During the course of a lifetime, approximately one out of every 10 men will be identified as having an elevated PSA level (considered greater than or equal to 4 ng/mL). However, only one in four of these men will actually turn out to have prostate cancer. Even though men with prostate cancer are quite likely to have an elevated PSA, an elevated PSA level often does not indicate prostate cancer. In fact, three out of four men who undergo a prostate biopsy based on abnormal results from a PSA test turn out not to have cancer. The PSA test, in other words, is sensitive but not specific.
In an effort to improve its specificity, several variations on the PSA test have been introduced. These include
age specific PSA
(PSA levels which take age into account),
(PSAD; in which the size of prostate gland is considered when interpreting the PSA level),
(how quickly the PSA level rises over time), and
percent free PSA
(fPSA; a test which identifies the amount of free versus the amount of bound PSA circulating in the blood). Unfortunately, none of these tests, with the exception of fPSA, has significantly reduced the number of biopsies performed.
To address this issue, a group of researchers set out to develop a predictive model for the detection of prostate cancer that would reduce the number of men having unnecessary prostate biopsies. The results of their study, which will be published in the October 1, 2003 issue of the journal
, finds that taking into account the results of clinical and laboratory tests can significantly improved the accuracy of the PSA test, reducing the number of unnecessary biopsies.
About the study
The researchers collected data on 1237 men who had a mildly elevated PSA levels between 4 and 10 ng/mL, and who had undergone an initial prostate biopsy. The researchers then collected the following data on each participant: age, race, family history, referral indication(s), prior vasectomy, DRE findings, PSA level, PSAD, and transrectal ultrasound (TRUS) findings. (TRUS uses ultrasound to make an image of the prostate on a video screen.)
The researcher found that there were four independent predictors of a positive biopsy. These included elevated PSAD, abnormal DRE, positive TRUS finding, and age of 75 years or older. Using these four independent predictors, the researchers constructed a numeric model in which a patient accumulates a certain number of points for each of the four categories. These four scores are then totaled to determine the patient’s overall risk of prostate cancer. For example, a man with a total score of 6.8 would have a 10% risk of having prostate cancer.
The researchers then determined that if only those men whose score indicated a 10% or greater risk for prostate cancer went on to have prostate biopsies, 92% would have prostate cancer.
How does this affect you?
By incorporating additional clinical and laboratory information, these researchers were able to develop a numeric scoring system that could significantly improve a physician’s ability to predict prostate cancer over the use of standard PSA testing alone. In fact, if other researchers can validate the results of this study, it may be possible to reduce the number of unnecessary biopsies by 24% using this model.
The widespread application of this new method would not be without its drawbacks, however. TRUS is a relatively expensive procedure that is not a part of routine prostate cancer screening. Whether or not the cost of TRUS for routine screening will offset the amount saved by reducing the number of unnecessary prostate biopsies has yet to be determined.
A method for reducing the number of prostate biopsies is made all the more urgent by the fact that PSA screening has not yet been shown to reduce the risk of death from prostate cancer in clinical trials. Performing unwarranted invasive procedures is costly and exposes patients to unnecessary discomfort and the risk of bleeding and infection. Yet simply raising the threshold for a positive result in order to reduce the number of biopsies would mean many missed cancers. Anything that can significantly improve the specificity of the test, without compromising its sensitivity, is sorely needed. This new method may be a step in that direction.
Garzotto M, Hudson RG, Peters L, et al. Predictive modeling for the presence of prostate carcinoma using clinical, laboratory, and ultrasound parameters in patients with prostate specific antigen levels ≤ 10 ng/mL.
2003;98:000-000. Early release article.
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