Tabei Tadashi, Taguri Masataka, Sakai Naoki, Koh Hideshige, Yosida Minoru, Fujikawa Atsushi, Nirei Takuma, Tsutsumi Sohgo, Ito Hiroki, Furuhata Souichi, Kawahara Takashi, Miyoshi Yasuhide, Noguchi Sumio, Uemura Hiroji, Kobayashi Kazuki
Department of Urology, Yokosuka Kyosai Hospital, Yokosuka, Japan.
Department of Data Science, School of Data Science, Yokohama City University, Yokohama, Japan.
Prostate. 2020 Aug;80(11):824-830. doi: 10.1002/pros.23997. Epub 2020 May 20.
Studies of prostate-specific antigen (PSA)-based population screening have been conducted in western countries, but there is little data in Asian populations. The objective of this study was to determine the efficacy of PSA screening in Asian men using real-world data over a period of 15 years after introducing population screening in Yokosuka City, Japan.
We investigated patients with pathologically diagnosed prostate cancer at four hospitals and two clinics across the Yokosuka area (Miura peninsula) between April 2001 and March 2015. Patients were divided into two groups; the S group consisted of those diagnosed by PSA-based population screening in Yokosuka City and the NS group consisted of those diagnosed by methods other than screening. Cancer-specific survival (CSS) and overall survival (OS) rates were calculated using the Kaplan-Meier method with the log-rank test to compare survival between the two groups. Clinical and pathological factors for cancer-specific mortality were assessed with Cox regression analyses to calculate the hazard ratio (HR) and 95% confidence interval (CI).
A total of 3094 patients had been diagnosed with prostate cancer over the 15-year period. The median follow-up period was 77 months. The S group and the NS group consisted of 977 and 2117 patients, respectively. Patients in the S group were younger (age: 71 years vs 73 years, P < .001) and had a lower Charlson comorbidity index (CCI) with favorable oncological factors, such as lower initial PSA, Gleason score (GS), and risk category. Kaplan-Meier curves for OS and CSS revealed significant differences between the two groups (OS: P < .001, CSS: P < .001). Analysis with Cox proportional hazards model indicated the NS group (HR: 1.584, 95% CI, 1.065-2.356, P = .023), a CCI > 4 (HR: 1.552, 95% CI, 1.136-2.120, P = .006), a GS ≥ 8 (HR: 4.869, 95% CI, 2.631-9.001, P < .001), and nonlocalized cancer (locally advanced; HR: 2.632, 95% CI, 1.676-4.133, P < .001, advanced; HR: 9.468, 95% CI, 6.279-14.278, P < .001) as independent risk factors for cancer-specific mortality.
PSA-based population screening of prostate cancer might be useful in the Japanese population.
西方国家已开展基于前列腺特异性抗原（PSA）的人群筛查研究，但亚洲人群的数据较少。本研究的目的是利用日本横须贺市引入人群筛查后15年的真实世界数据，确定PSA筛查在亚洲男性中的效果。
我们调查了2001年4月至2015年3月期间横须贺地区（三浦半岛）四家医院和两家诊所中经病理诊断为前列腺癌的患者。患者分为两组；S组由在横须贺市通过基于PSA的人群筛查诊断出的患者组成，NS组由通过筛查以外的方法诊断出的患者组成。采用Kaplan-Meier法和对数秩检验计算癌症特异性生存率（CSS）和总生存率（OS），以比较两组之间的生存率。通过Cox回归分析评估癌症特异性死亡率的临床和病理因素，以计算风险比（HR）和95%置信区间（CI）。
在这15年期间，共有3094例患者被诊断为前列腺癌。中位随访期为77个月。S组和NS组分别有977例和2117例患者。S组患者更年轻（年龄：71岁对73岁，P < .001），Charlson合并症指数（CCI）更低，且具有有利的肿瘤学因素，如初始PSA、Gleason评分（GS）和风险类别更低。OS和CSS的Kaplan-Meier曲线显示两组之间存在显著差异（OS：P < .001，CSS：P < .001）。Cox比例风险模型分析表明，NS组（HR：1.584，95%CI，1.065 - 2.356，P = .023）、CCI > 4（HR：1.552，95%CI，1.136 - 2.120，P = .006）、GS ≥ 8（HR：4.869，95%CI，2.631 - 9.001，P < .
