OBJECTIVES: To investigate the age-specific reference range of prostate-specific antigen (PSA) in Japanese men older than 60 years of age by analyzing the receiver operating characteristic (ROC) curve. Several reports have noted that many clinically serious cancers are missed by raising the cutoff value of the age-specific PSA reference range for men older than 60 years. METHODS: We studied 6744 individuals who had undergone mass screening for prostate cancer in Gunma Prefecture from 1994 to 1998. PSA determination was the first step of the mass screening in all subjects. Digital rectal examination and transrectal ultrasonography (TRUS) were performed in all except in a fraction of patients. Subjects with an abnormal PSA level or abnormal digital rectal examination or TRUS findings underwent TRUS-guided systematic sextant biopsies. Patients older than 60 years of age were grouped according to their age at 5-year intervals, and the cutoff value of the age-specific PSA reference range was calculated for each age group by analyzing the ROC curve. RESULTS: The diagnostic efficiency of the age-specific PSA reference range was optimal with cutoff values of 3.0, 3.5, 4.0, 4.0, and 7.0 ng/mL in subjects 60 to 64, 65 to 69, 70 to 74, 75 to 79, and older than 80 years of age, respectively. By using the age-specific PSA reference range as determined by the ROC curve, the sensitivity, specificity, and efficiency increased to 92.4%, 91.2%, and 84.3%, respectively. When the standard PSA reference range was used for the diagnosis, the sensitivity, specificity, and efficiency was 89.1%, 92.4%, and 82.3%, respectively. All of the cases of prostate cancer detected by using the age-specific PSA reference range with the cutoff point based on the ROC curves were clinically significant. CONCLUSIONS: The age-specific PSA reference range cutoff value in this setting demonstrated better diagnostic efficiency than the standard cutoff value of PSA and the age-specific PSA reference range determined by the 95% confidence interval. It appears likely to be a useful diagnostic index for the first step of mass screening in Japanese men.
OBJECTIVES: To investigate the age-specific reference range of prostate-specific antigen (PSA) in Japanese men older than 60 years of age by analyzing the receiver operating characteristic (ROC) curve. Several reports have noted that many clinically serious cancers are missed by raising the cutoff value of the age-specific PSA reference range for men older than 60 years. METHODS: We studied 6744 individuals who had undergone mass screening for prostate cancer in Gunma Prefecture from 1994 to 1998. PSA determination was the first step of the mass screening in all subjects. Digital rectal examination and transrectal ultrasonography (TRUS) were performed in all except in a fraction of patients. Subjects with an abnormal PSA level or abnormal digital rectal examination or TRUS findings underwent TRUS-guided systematic sextant biopsies. Patients older than 60 years of age were grouped according to their age at 5-year intervals, and the cutoff value of the age-specific PSA reference range was calculated for each age group by analyzing the ROC curve. RESULTS: The diagnostic efficiency of the age-specific PSA reference range was optimal with cutoff values of 3.0, 3.5, 4.0, 4.0, and 7.0 ng/mL in subjects 60 to 64, 65 to 69, 70 to 74, 75 to 79, and older than 80 years of age, respectively. By using the age-specific PSA reference range as determined by the ROC curve, the sensitivity, specificity, and efficiency increased to 92.4%, 91.2%, and 84.3%, respectively. When the standard PSA reference range was used for the diagnosis, the sensitivity, specificity, and efficiency was 89.1%, 92.4%, and 82.3%, respectively. All of the cases of prostate cancer detected by using the age-specific PSA reference range with the cutoff point based on the ROC curves were clinically significant. CONCLUSIONS: The age-specific PSA reference range cutoff value in this setting demonstrated better diagnostic efficiency than the standard cutoff value of PSA and the age-specific PSA reference range determined by the 95% confidence interval. It appears likely to be a useful diagnostic index for the first step of mass screening in Japanese men.