-
前列腺癌是老年男性最常见的恶性肿瘤之一[1]。随着前列腺特异性抗原(prostate specific antigen,PSA)、直肠指诊等诊断方法的不断进步与普及,更多的早期前列腺癌被检出[2]。通过手术、放疗等治疗手段,其治愈率也得到显著提高[3]。准确地预测前列腺癌病理特征可以帮助前列腺癌患者选择更加适合的治疗方案,从而提高治愈率。Partin表是通过4133例大样本统计分析得到的图表[4]。本研究将通过比较MRI与Partin表对术后器官局限性癌(organ-confined disease,OCD)、包膜侵犯(established capsular penetration,ECP)、精囊侵犯(seminal vesicle involvement,SVI)以及淋巴结转移(lymph nodal involvement,LNI)4个病理特征的预测结果,对二者的准确性进行分析。
-
本组51例患者中,术后病理证实OCD、ECP、SVI及LNI的患者分别为20例、18例、10例和3例。
-
分析MRI图像,对51例前列腺癌患者病理特征进行预测,OCD患者21例,ECP患者25例,SVI患者5例,LNI患者0例。将预测结果与术后病理结果进行比较,MRI预测4种术后病理特征的ROC曲线及AUC见图 1。MRI预测OCD、ECP、SVI及LNI的准确率、灵敏度、特异度、AUC及临床价值见表 1。
图 1 MRI预测前列腺癌患者病理特征的受试者工作特征曲线
Figure 1. The receiver operating characteristic curve of MRI pridict pathological feature of prostate cancer
病理特征 准确率(%) 灵敏度(%) 特异度 AUC 临床价值 OCD 90.0 90.0 90.3 0.902 较高 ECP 88.3 88.3 69.7 0.765 中等 SVI 20.0 20.0 92.7 0.563 较低 LNI 0 0 100 0.500 无 表中,OCD:器官局限性癌;ECP:包膜侵犯;SVI:精囊侵犯;LNI:淋巴结转移;AUC:曲线下面积。 表 1 MRI预测OCD、ECP、SVI及LNI准确率、灵敏度、特异度、AUC及临床价值
Table 1. The accuracy, sentivity, spectivity, AUC and clinic value of MRI
-
将患者PSA值、临床分期、Gleason评分带入Partin表,预测出的OCD、ECP、SVI及LNI 4种术后病理特征的ROC曲线及AUC见图 2。选取ROC曲线中的相应临界点,Partin表预测出的OCD、ECP、SVI及LNI的灵敏度、特异度、AUC及临床价值见表 2。
图 2 Partin表预测前列腺癌患者病理特征的受试者工作特征曲线
Figure 2. The receiver operating characteristic curve of Partin tables pridict pathological feature of prostate cancer
病理特征 临界点 灵敏度(%) 特异度(%) AUC 临床价值 OCD 0.14 90.00 80.65 0.911 较高 ECP 0.44 72.22 72.73 0.742 中等 SVI 0.24 90.00 73.17 0.827 中等 LNI 0.26 100 83.33 0.899 中等 表中,OCD:器官局限性癌;ECP:包膜侵犯;SVI:精囊侵犯;LNI:淋巴结转移;AUC:曲线下面积。 表 2 Partin表预测OCD、ECP、SVI及LNI灵敏度、特异度、AUC及临床价值
Table 2. The accuracy, sentivity, spectivity, AUC of Partin tables
-
MRI与Partin表ROC曲线下面积比较见表 3。在预测OCD和ECP方面,MRI与Partin表的AUC差异无统计学意义(Z=0.071、0.020,P均 > 0.05);在预测SVI和LNI方面,MRI与Partin表的AUC差异有统计学意义(Z=0.286、0.499,P均 < 0.01)。
器官局限性癌 包膜侵犯 精囊侵犯 淋巴结转移 MRI 0.902(95%CI:0.785~0.967) 0.765(95%CI:0.626~0.872) 0.563(95%CI:0.417~0.702) 0.500(95%CI:0.357~0.643) Partin表 0.911(95%CI:0.798~0.973) 0.742(95%CI:0.600~0.854) 0.827(95%CI:0.695~0.918) 0.899(95%CI:0.782~0.966) Z值 0.071 0.020 0.286 0.499 P值 > 0.05 > 0.05 < 0.01 < 0.01 表 3 MRI与Partin表的受试者工作特征曲线下面积
Table 3. The area under receiver operating characteristic curve of MRI and Partin tables
MRI和1997年版Partin表对前列腺癌病理特征预测准确性的对比研究
Accuracy of MRI and 1997 edition of Partin tables in predicting the pathological features of prostate cancer
-
摘要:
目的 比较MRI与1997年版Partin表对前列腺癌病理特征预测结果的准确性。 方法 随机选取2012年1月至2014年2月51例前列腺癌患者并行根治性前列腺切除术,统计资料具备术前血清前列腺特异性抗原结果、临床分期、Gleason评分、盆腔MRI资料以及术后病理报告。分别将术前MRI及Partin表对前列腺癌器官局限性癌、包膜侵犯、精囊侵犯以及淋巴结转移4种病理特征的预测结果与术后病理结果进行比较,通过受试者工作特征曲线(ROC)分析法检验MRI与Partin表预测前列腺癌病理特征的准确性并进行比较。 结果 应用Partin表预测器官局限性癌、包膜侵犯、精囊侵犯以及淋巴结转移4种病理特征的曲线下面积分别为0.911、0.742、0.827、0.899;应用MRI预测的曲线下面积分别为0.902、0.765、0.563、0.500。MRI与Partin表预测器官局限性癌和包膜侵犯的ROC曲线下面积差异无统计学意义(Z=0.071、0.020,P均>0.05);预测精囊侵犯和淋巴结转移的ROC曲线下面积差异有统计学意义(Z=0.286、0.499,P均 < 0.01)。 结论 Partin表预测前列腺癌4种病理特征具有临床诊断价值;MRI与Partin表预测前列腺癌器官局限性癌、包膜侵犯2种病理特征的准确性差异无统计学意义,Partin表预测精囊侵犯、淋巴结转移2种病理特征的准确性高于MRI。 Abstract:Objective To compare the accuracies of MRI and 1997 edition of Partin tables in predicting the pathological features of prostate cancer. Methods A total of 51 patients with prostate carcinoma underwent radical prostatectomy from January 2012 to February 2014. Preoperative serum prostate specific antigen, clinical stage, biopsy Gleason score, percentage of positive biopsy scores, pelvic MRI, and pathological report of prostatectomy specimen were collected. Postoperative pathological results were compared with MRI results and Partin tables. Receiver operating characteristic curves were plotted to determine the accuracies of MRI and Partin tables in predicting the pathological features of prostate cancer. Results The areas under the curve(AUCs)of the Partin tables to predict organ-confined disease(OCD), established capsular penetration(ECP), seminal vesicle involvement(SVI), and lymph nodal involvement(LNI) were 0.911, 0.742, 0.827, and 0.899, respectively. The corresponding AUCs of MRI were 0.902, 0.765, 0.563, and 0.5. The AUCs of OCD and ECP did not significantly differ between MRI and Partin tables(Z=0.071 and 0.020, both P > 0.05). By contrast, the AUCs of SVI and LNI significantly differed between MRI and Partin tables(Z=0.286 and 0.499, both P < 0.01). Conclusion Partin tables exhibit a high clinical diagnostic value in the prediction of the pathological feature of prostate cancer. The accuracies of predicting OCD and ECP did not significantly differ between MRI and Partin tables. However, Partin tables were more accurate in predicting SVI and LNI than MRI. -
表 1 MRI预测OCD、ECP、SVI及LNI准确率、灵敏度、特异度、AUC及临床价值
Table 1. The accuracy, sentivity, spectivity, AUC and clinic value of MRI
病理特征 准确率(%) 灵敏度(%) 特异度 AUC 临床价值 OCD 90.0 90.0 90.3 0.902 较高 ECP 88.3 88.3 69.7 0.765 中等 SVI 20.0 20.0 92.7 0.563 较低 LNI 0 0 100 0.500 无 表中,OCD:器官局限性癌;ECP:包膜侵犯;SVI:精囊侵犯;LNI:淋巴结转移;AUC:曲线下面积。 表 2 Partin表预测OCD、ECP、SVI及LNI灵敏度、特异度、AUC及临床价值
Table 2. The accuracy, sentivity, spectivity, AUC of Partin tables
病理特征 临界点 灵敏度(%) 特异度(%) AUC 临床价值 OCD 0.14 90.00 80.65 0.911 较高 ECP 0.44 72.22 72.73 0.742 中等 SVI 0.24 90.00 73.17 0.827 中等 LNI 0.26 100 83.33 0.899 中等 表中,OCD:器官局限性癌;ECP:包膜侵犯;SVI:精囊侵犯;LNI:淋巴结转移;AUC:曲线下面积。 表 3 MRI与Partin表的受试者工作特征曲线下面积
Table 3. The area under receiver operating characteristic curve of MRI and Partin tables
器官局限性癌 包膜侵犯 精囊侵犯 淋巴结转移 MRI 0.902(95%CI:0.785~0.967) 0.765(95%CI:0.626~0.872) 0.563(95%CI:0.417~0.702) 0.500(95%CI:0.357~0.643) Partin表 0.911(95%CI:0.798~0.973) 0.742(95%CI:0.600~0.854) 0.827(95%CI:0.695~0.918) 0.899(95%CI:0.782~0.966) Z值 0.071 0.020 0.286 0.499 P值 > 0.05 > 0.05 < 0.01 < 0.01 -
[1] Ferlay J, Shin HR, Bray F, et al. Estimates of worldwide burden of cancer in 2008:GLOBOCAN 2008[J]. Cancer, 2010127):2893-2917. [2] Siegel R, Naishadham D, Jemal A. Cancer statistics, 2012[J]. CA Cancer J Clin, 2012, 621):10-29. doi: 10.3322/caac.20138 [3] Epstein JI, Partin AW, Sauvageot J, et al. Prediction of progression following radical prostatectomy. A multivariate analysis of 721 men with long-term follow-up[J]. Surg Pathol, 1996, 203):286-292. [4] Partin AW, Kattan MW, Subong EN, et al. Combination of prostate-specific antigen, clinical stage, and Gleason score to predict pathological stage of localized prostate cancer. A multi-institutional update[J]. JAMA, 1997, 27718):1445-1451. doi: 10.1001/jama.1997.03540420041027 [5] Schroder FH, Hermanek P, Denis L, et al. The TNM classification of prostate cancer[J]. Prostate Suppl, 19924):129-138. [6] Soylu FN, Eggener S, Oto A. Local staging of prostate cancer with MRI[J]. Diagn interv Radiol, 2012, 184):365-373. [7] Tanaka H, Hayashi S, Ohtakara K, et al. Usefulness of CT-MRI fusion in radiotherapy planning for localized prostate cancer[J]. Radiat Res, 2011, 526):782-788. doi: 10.1269/jrr.11053 [8] 晋晖. 3.0T磁共振对前列腺外周带癌的诊断价值[J].中国医疗前沿, 2013, 87):97-98.
[9] 李春媚, 陈敏, 李飒英, 等.前列腺癌MR动态增强扫描定量分析及其应用[J].中华放射学杂志, 2011, 455):508-5l0. doi: 10.3760/cma.j.issn.1005-1201.2011.05.023
[10] Doo KW, Sung DJ, Park BJ, et al. Detectability of low and intermediate or high risk prostate cancer with combined T2-weighted and diffusion-weighted MRI[J]. Eur Radiol, 2012, 228):1812-1819. doi: 10.1007/s00330-012-2430-5 [11] 赵阳, 徐勇, 韩悦, 等. MR波谱成像对前列腺癌分化程度预估价值的初步研究[J].中华放射学杂志, 2011, 4510):951-954. doi: 10.3760/cma.j.issn.1005-1201.2011.10.014
[12] Nix Jw, Turkhey B, Hoang A, et al.Very distal apical prostate tumors:identification on multiparametric MRI at 3 Tesla[J]. BJU Int, 2012, 11011 Pt B):694-700. [13] Kaji Y, Kurhanewicz J, Hricak H, et al. Localizing prostate cancer in the presence of post biopsy changes on MR images:role of proton MR spectroscopic imaging[J]. Radiology, 1998, 2063):785-790. [14] Partin AW, Mangold LA, Lamm DM, et al. Contemporary update of prostate cancer staging nomogramsPartin Tables)for the new millennium[J]. Urology, 2001, 586):843-848. doi: 10.1016/S0090-4295(01)01441-8 [15] Makarov DV, Trock BJ, Humphreys EB, et al. Updated nomogram to predict pathologic stage of prostate cancer given prostate specific antigen level, clinical stage, and biopsy Gleason scorePartin tables)based on cases from 2000 to 2005[J]. Urology, 2007, 696):1095-1101. doi: 10.1016/j.urology.2007.03.042 [16] Kattan MW. Re:Updated nomogram to predict pathologic stage of prostate cancer given prostate-specific antigen level, clinical stage, and biopsy gleason scorePartin Tables) based on cases from 2000 to 2005[J]. Eur Urol, 2007, 525):1528. [17] 高旭, 孙颖浩, 许传亮, 等.应用Partin tables预测国人前列腺癌术后病理特征准确性的研究[J].中华泌尿外科杂志, 2007, 289):614-618. doi: 10.3760/j.issn:1000-6702.2007.09.010
[18] Ren J, Huan Y, Wang H, et al. Seminal vesicle invasion in prostate cancer:prediction with combined T2-weighted and diffusion-weighted MR imaging[J]. Eur Radiol, 2009, 1910):2481-2486. doi: 10.1007/s00330-009-1428-0 [19] Wang YJ. Superparamagnetic iron oxide based MRI contrast agents:Current status of clinical application[J]. Quant Imaging Med Surg, 20111):35-40. [20] Hentschel B, Oehler W, Strauss D, et al. Definition of the CTV prostate in CT and MRI by using CT-MRI image fusion in IMRT planning for prostate cancer[J]. Strahlenther Onkol, 2011, 1873):183-190. doi: 10.1007/s00066-010-2179-1