DIAGNOSTIC VALUE OF PERCENT FREE PROSTATE- SPECIFIC ANTIGEN: RETROSPECTIVE ANALYSIS OF A POPULATION-BASED SCREENING STUDY WITH EMPHASIS ON MEN WITH PSA LEVELS LESS THAN 3.0 ng/mL MAGNUS TO ¨ RNBLOM, ULF NORMING, JAN ADOLFSSON, CHARLOTTE BECKER, PER-ANDERS ABRAHAMSSON, HANS LILJA, AND OVE GUSTAFSSON ABSTRACT Objectives. To retrospectively investigate the use of percent free prostate-specific antigen (PSA) compared with total PSA in serum as predictor of prostate cancer in men selected randomly from the general population who underwent biopsy on the basis of abnormal findings on digital rectal examination (DRE) or transrectal ultrasound (TRUS) and/or serum PSA levels greater than 10 ng/mL. Methods. A single intervention, population-based screening study was undertaken in 1988 and 1989. Of the 2400 men aged 55 to 70 years invited to participate, 1782 men responded and were examined with DRE, TRUS, and PSA testing (Tandem-Hybritech). In 1995, frozen serum samples from 1748 men were analyzed for percent free PSA (Prostatus, Wallac OY). Five-year follow-up data on new cancers in the screened population were obtained from the Swedish Cancer Registry (SCR). Results. Of the 1748 men, 367 underwent TRUS-guided biopsies because of abnormal findings on either DRE or TRUS or serum PSA levels of greater than 10 ng/mL. This resulted in the diagnosis of 64 cases of prostate cancer (3.7%). PSA levels of 3.0 ng/mL or greater were found in 55 (86%) of 64 cancer cases and in 399 (24%) of the 1684 benign cases. Among the 1294 men with PSA less than 3.0 ng/mL, 9 prostate cancers were diagnosed (14% of all prostate cancers). All 9 patients with cancer and with PSA less than 3.0 ng/mL had a percent free PSA of 18% or less. In the group of 1109 patients with PSA less than 3.0 ng/mL and a percent free PSA greater than 18%, 159 biopsies were performed because of abnormal DRE or TRUS. However, no prostate cancer was diagnosed in this category of patients. Five years after the screening intervention, 7 more cases of prostate cancer were clinically diagnosed in the screened population according to the SCR. Conclusions. The combination of PSA levels less than 3.0 ng/mL and percent free PSA greater than 18% defines a large part of the population at a very low risk of cancer of the prostate both at the time of screening and during the following 5 years. Men in this group may be spared DRE, and longer screening intervals may be considered. However, the risk of having prostate cancer is not negligible in men with PSA less than 3.0 ng/mL and percent free PSA of 18% or less. The results of this study indicate that biopsy should be recommended to men fulfilling these criteria, although these results should be confirmed in larger prospec- tive studies because of the limited number of patients with prostate cancer in the present series. UROLOGY 53: 945–950, 1999. © 1999, Elsevier Science Inc. All rights reserved. This study was supported by grants from the Gunvor and Josef Ane ´r Foundation; the Sigurd and Elsa Golje Memory Trust; the Swedish Association for Cancer and Traffic Victims; Biomed 2 Program, Area 4.1.7 (contract number BMH4-CT96-0453); and the Swedish Cancer Society (project number 3555). Dr. Hans Lilja is a scientific advisor to EG & G Wallac and is a patent inventor for this technology. From the Department of Urology, Huddinge Hospital, Karolin- ska Institute, Stockholm; Department of Urology, Lund Univer- sity Hospital, Lund; and Department of Clinical Chemistry, Malmo ¨ University Hospital, Malmo ¨ , Sweden Reprint requests: Magnus To ¨rnblom, M.D., Karolinska Insti- tute, Department of Urology, Huddinge Hospital, Huddinge S-141 86, Sweden Submitted: April 21, 1998, accepted (with revisions): Novem- ber 16, 1998 ADULT UROLOGY © 1999, ELSEVIER SCIENCE INC. 0090-4295/99/$20.00 ALL RIGHTS RESERVED PII S0090-4295(98)00640-2 945