Xiao Wen-Jun, Ye Ding-Wei, Yao Xu-Dong, Zhang Shi-Lin, Dai Bo, Wang Chao-Fu, Wang Jian, Zhang Hai-Liang, Shen Yi-Jun, Zhu Yao, Zhu Yi-Ping, Shi Guo-Hai, Ma Chun-Guang, Qin Xiao-Jian, Lin Guo-Wen
Department of Urology, Fudan University Shanghai Cancer Centre, Shanghai, China.
Can J Urol. 2011 Apr;18(2):5619-24.
To perform a head to head comparison among three generations of Partin tables, namely from 1997, 2001 and the last updated version of 2007, in a Chinese cohort of prostate cancer.
Clinical and pathological data of 198 consecutive Chinese patients were retrospectively analyzed, who underwent radical prostatectomy for clinically localized prostate cancer between January 2005 and May 2010. Three versions of the Partin tables were compared for their accuracy and performance to predict final pathological stage using receiver operating characteristic (ROC) curve.
Of the whole cohort 58.6% were presented with organ-confined disease (OCD), 10.1% had lymph node involvement (LNI), and 31.3% had locally advanced disease (LAD), while 21.2% had extraprostatic extension (ECE) and 10.1% showed seminal vesicle involvement (SVI). The area under the ROC curve (AUC) of the Partin Tables 1997, 2001 and 2007 was 0.732, 0.722 and 0.695 for OCD; 0.647, 0.594 and 0.577 for LAD; 0.856, 0.872 and 0.829 for LNI, respectively.
All three generations of the Partin tables showed a good accuracy to predict OCD, and LNI. However, the predictive accuracy for LAD was more limited. Overall, the newer versions of the Partin tables could not exceed the version of 1997 in their predictive accuracy for the present Chinese cohort. Our results suggest caution when using newly introduced predictive tools that are not supported by population-specific accuracy tests.
在一组中国前列腺癌患者中,对三代Partin表(即1997年、2001年以及2007年的最新版本)进行直接比较。
回顾性分析了198例连续的中国患者的临床和病理数据,这些患者在2005年1月至2010年5月期间因临床局限性前列腺癌接受了根治性前列腺切除术。使用受试者工作特征(ROC)曲线比较了三个版本的Partin表在预测最终病理分期方面的准确性和性能。
在整个队列中,58.6%的患者表现为器官局限性疾病(OCD),10.1%有淋巴结受累(LNI),31.3%有局部晚期疾病(LAD),而21.2%有前列腺外扩展(ECE),10.1%有精囊受累(SVI)。1997年、2001年和2007年Partin表的ROC曲线下面积(AUC),对于OCD分别为0.732、0.722和0.695;对于LAD分别为0.647、0.594和0.577;对于LNI分别为0.856、0.872和0.829。
三代Partin表在预测OCD和LNI方面均显示出良好的准确性。然而,对LAD的预测准确性较为有限。总体而言,对于当前的中国队列,Partin表的新版本在预测准确性上并未超过1997年的版本。我们的数据表明,在使用未经特定人群准确性测试支持的新引入预测工具时应谨慎。