当代北美前列腺癌患者中2012年更新版Partin表基于人群的外部验证

Population-Based External Validation of the Updated 2012 Partin Tables in Contemporary North American Prostate Cancer Patients.

作者信息

Leyh-Bannurah Sami-Ramzi, Gazdovich Stéphanie, Budäus Lars, Zaffuto Emanuele, Dell'Oglio Paolo, Briganti Alberto, Abdollah Firas, Montorsi Francesco, Schiffmann Jonas, Menon Mani, Shariat Shahrokh F, Fisch Margit, Chun Felix, Graefen Markus, Karakiewicz Pierre I

机构信息

Cancer Prognostics and Health Outcomes Unit, University of Montreal Health Center, Montreal, Canada.

Martini-Clinic, Prostate Cancer Center Hamburg-Eppendorf, Hamburg, Germany.

出版信息

Prostate. 2017 Jan;77(1):105-113. doi: 10.1002/pros.23253. Epub 2016 Sep 29.

DOI:10.1002/pros.23253

PMID:27683103

Abstract

OBJECTIVE

To externally validate the updated 2012 Partin Tables in contemporary North American patients treated with radical prostatectomy (RP) for localized prostate cancer (PCa) at community institutions.

MATERIALS AND METHODS

We examined records of 25,254 patients treated with RP and pelvic lymph node dissection (PLND) between 2010 and 2013, within the surveillance, epidemiology, and end results database. The ROC derived AUC assessed discriminant properties of the updated 2012 Partin Tables of organ confined disease (OC), extracapsular extension (ECE), seminal vesical invasion (SVI), and lymph node invasion (LNI). Calibration plots focused on calibration between predicted and observed rates.

RESULTS

Proportions of OC, ECE, SVI, and LNI at RP were 69.8%, 18.4%, 7.4%, and 4.4%, respectively. Accuracy for prediction of OC, ECE, SVI, and LNI was 70.4%, 59.9%, 72.9%, and 77.1%, respectively. In subgroup analyses in patients with nodal yield >10, accuracy for LNI prediction was 76.0%. Subgroup analyses in elderly patients and in African American patients revealed decreased accuracy for prediction of all four endpoints. Last but not least, SVI and LNI calibration plots showed excellent agreement, versus good agreement for OC (maximum underestimation of 10%) and poor agreement for ECE (maximum overestimation of 12%).

CONCLUSION

Taken together, the updated 2012 Partin Tables can be unequivocally endorsed for prediction of OC, SVI, and LNI in community-based patients with localized PCa. Conversely, ECE predictions failed to reach the minimum accuracy requirements of 70%. Prostate 77:105-113, 2017. © 2016 Wiley Periodicals, Inc.

摘要

目的

在当代北美社区机构中接受根治性前列腺切除术（RP）治疗局限性前列腺癌（PCa）的患者中，对外验证2012年更新版的Partin表。

材料与方法

我们在监测、流行病学和最终结果数据库中，检查了2010年至2013年间接受RP和盆腔淋巴结清扫术（PLND）治疗的25254例患者的记录。ROC曲线下面积（AUC）评估了2012年更新版Partin表对器官局限性疾病（OC）、包膜外侵犯（ECE）、精囊侵犯（SVI）和淋巴结侵犯（LNI）的判别特性。校准图重点关注预测率与观察率之间的校准。

结果

RP时OC、ECE、SVI和LNI的比例分别为69.8%、18.4%、7.4%和4.4%。OC、ECE、SVI和LNI预测的准确率分别为70.4%、59.9%、72.9%和77.1%。在淋巴结检出数>10的患者亚组分析中，LNI预测的准确率为76.0%。老年患者和非裔美国患者的亚组分析显示，所有四个终点预测的准确率均降低。最后，SVI和LNI校准图显示出极佳的一致性，而OC为良好一致性（最大低估10%），ECE为较差一致性（最大高估12%）。