Liu X M, Duan H Y, Zhang D Q, Chen C, Ji Y T, Zhang Y M, Feng Z W, Liu Y, Li J J, Zhang Y, Li C Y, Zhang Y C, Yang L, Lyu Z Y, Song F F, Song F J, Huang Y B
Department of Cancer Epidemiology and Biostatistics, National Clinical Research Center for Cancer, Tianjin's Clinical Research Center for Cancer, Key Laboratory of Molecular Cancer Epidemiology, Tianjin Medical University Cancer Institute and Hospital, Tianjin 300060, China.
Department of Hospital Information System, National Clinical Research Center for Cancer, Tianjin's Clinical Research Center for Cancer, Key Laboratory of Molecular Cancer Epidemiology, Tianjin Medical University Cancer Institute and Hospital, Tianjin 300060, China.
Zhonghua Zhong Liu Za Zhi. 2024 Apr 23;46(4):354-364. doi: 10.3760/cma.j.cn112152-20230805-00062.
To determine the total and age-specific cut-off values of total prostate specific antigen (tPSA) and the ratio of free PSA divided total PSA (fPSA/tPSA) for screening prostate cancer in China. Based on the Chinese Colorectal, Breast, Lung, Liver, and Stomach cancer Screening Trial (C-BLAST) and the Tianjin Common Cancer Case Cohort (TJ4C), males who were not diagnosed with any cancers at baseline since 2017 and received both tPSA and fPSA testes were selected. Based on Cox regression, the overall and age-specific (＜60, 60-＜70, and ≥70 years) accuracy and optimal cut-off values of tPSA and fPSA/tPSA ratio for screening prostate cancer were evaluated with time-dependent receiver operating characteristic curve (tdROC) and area under curve (AUC). Bootstrap resampling was used to internally validate the stability of the optimal cut-off value, and the PLCO study was used to externally validate the accuracy under different cut-off values. A total of 5 180 participants were included in the study, and after a median follow-up of 1.48 years, a total of 332 prostate cancer patients were included. In the total population, the tdAUC of tPSA and fPSA/tPSA screening for prostate cancer were 0.852 and 0.748, respectively, with the optimal cut-off values of 5.08 ng/ml and 0.173, respectively. After age stratification, the age specific cut-off values of tPSA in the <60, 60-<70, and ≥70 age groups were 3.13, 4.82, and 11.54 ng/ml, respectively, while the age-specific cut-off values of fPSA/tPSA were 0.153, 0.135, and 0.130, respectively. Under the age-specific cut-off values, the sensitivities of tPSA screening for prostate cancer in males <60, 60-70, and ≥70 years old were 92.3%, 82.0%, and 77.6%, respectively, while the specificities were 84.7%, 81.3%, and 75.4%, respectively. The age-specific sensitivities of fPSA/tPSA for screening prostate cancer were 74.4%, 53.3%, and 55.9%, respectively, while the specificities were 83.8%, 83.7%, and 83.7%, respectively. Both bootstrap's internal validation and PLCO external validation provided similar results. The combination of tPSA and fPSA/tPSA could further improve the accuracy of screening. To improve the screening effects, it is recommended that age-specific cut-off values of tPSA and fPSA/tPSA should be used to screen for prostate cancer in the general risk population.
确定中国前列腺癌筛查中总前列腺特异性抗原（tPSA）及游离PSA与总PSA比值（fPSA/tPSA）的总体及年龄特异性临界值。基于中国结直肠癌、乳腺癌、肺癌、肝癌和胃癌筛查试验（C-BLAST）及天津常见癌症病例队列（TJ4C），选取2017年以来基线时未诊断出任何癌症且接受了tPSA和fPSA检测的男性。基于Cox回归，采用时间依赖性受试者工作特征曲线（tdROC）和曲线下面积（AUC）评估tPSA和fPSA/tPSA比值筛查前列腺癌的总体及年龄特异性（＜60岁、60-＜70岁和≥70岁）准确性及最佳临界值。采用Bootstrap重抽样对最佳临界值的稳定性进行内部验证，并采用PLCO研究对不同临界值下的准确性进行外部验证。该研究共纳入5180名参与者，中位随访1.48年后，共纳入332例前列腺癌患者。在总体人群中，tPSA和fPSA/tPSA筛查前列腺癌的tdAUC分别为0.852和0.748，最佳临界值分别为5.08 ng/ml和0.173。年龄分层后，＜60岁、60-＜70岁和≥70岁年龄组tPSA的年龄特异性临界值分别为3.13、4.82和11.54 ng/ml，fPSA/tPSA的年龄特异性临界值分别为0.153、0.135和0.130。在年龄特异性临界值下，＜60岁、60-70岁和≥70岁男性中tPSA筛查前列腺癌的敏感性分别为92.3%、82.0%和77.6%，特异性分别为84.7%、81.3%和75.4%。fPSA/tPSA筛查前列腺癌的年龄特异性敏感性分别为74.4%、53.3%和55.9%，特异性分别为83.8%、83.7%和83.7%。Bootstrap内部验证和PLCO外部验证均得到相似结果。tPSA和fPSA/tPSA联合使用可进一步提高筛查准确性。为提高筛查效果，建议在一般风险人群中采用tPSA和fPSA/tPSA的年龄特异性临界值进行前列腺癌筛查。
