The amount of prostate-specific antigen (PSA) in a man's bloodstream at the time of his prostate cancer diagnosis or its rate of change over the course of the disease does not adequately predict lethal prostate cancer, according to a study in the April 4 Journal of the National Cancer Institute.
Although men with untreated localized prostate cancer have high long-term survival rates, many patients undergo treatment anyway. In order to avoid unnecessary treatment, researchers want to identify methods to determine which patients will develop lethal prostate cancer. The rate of increase of PSA:a protein produced by the prostate'before prostate cancer treatment has been associated with the patient's prognosis, which suggests that early measurements of PSA may predict the behavior of the tumor.
To assess the accuracy of using PSA to predict prostate cancer outcome, Katja Fall, M.D., Ph.D., of the Karolinska Institute in Stockholm, and colleagues analyzed the rate of change of PSA levels in 267 men from Sweden, Finland, and Iceland who were diagnosed with early localized prostate cancer between 1989 and 1999. The researchers recorded the PSA levels for the first two years after diagnosis to capture the patients' early PSA patterns. The men in the study received no curative treatment for the first two years but were closely watched for signs of progression, which is called watchful waiting.
At the end of the follow-up in December 2003, 34 patients had died from prostate cancer, and 18 had developed metastatic prostate cancer but were still alive. Although initial PSA values and the rate of change were associated with later development of lethal prostate cancer, they were not accurate enough to predict lethal cancer.