Prostate specific antigen levels in men older than 50 years without clinical evidence of prostatic carcinoma.

In their derivation of normal prostate specific antigen (PSA) levels (0 to 4.0 ng./ml.) Hybritech used almost exclusively men less than 60 years old. The purpose of this study was to define PSA levels by age decade in men older than 50 years without clinical evidence of prostatic carcinoma or so-called cancer-free. We define the cancer-free population as men with a PSA less than or equal to 4.0 ng./ml. and nonsuspicious digital rectal examination, and those with an abnormality in either parameter with a nonmalignant prostate biopsy. A total of 755 men was recruited for a prostate cancer detection study using serum PSA and digital rectal examination, of whom 728 fulfilled our criteria of cancer-free. Newly derived mean and standard deviation were 1.32 +/- 1.10 ng./ml. in the 50 to 59-year group, 1.91 +/- 1.72 ng./ml. in the 60 to 69-year group and 2.36 +/- 1.98 ng./ml. in the 70 to 79-year group. The upper limits for PSA (mean+2 standard deviations) by age were 3.5 ng./ml. in the 50 to 59-year group, 5.4 ng./ml. in the 60 to 69-year group and 6.3 ng./ml. in the 70 to 79-year group. The apparent accuracy of these new limits is strong in the 60 to 69-year group but it declines in the next decade. The data support further attempts at using PSA, age and digital rectal examination to establish selection criteria for prostate biopsy with adequate sensitivity and specificity.