根治性前列腺切除术后Gleason分级升高的相关因素及预测列线图模型的建立

Associated factors for Gleason grade upgrading after radical prostatectomy and establishment of a predictive Nomogram model

  • 摘要:
      目的  分析前列腺穿刺活检Gleason分级≤4级患者根治性前列腺切除术(RP)后Gleason分级升高(GGU)的相关因素,并建立预测GGU的列线图模型。
      方法  回顾性分析2017年1月—2020年1月河南省人民医院收治的78例穿刺活检Gleason分级≤4级并行RP治疗的患者的临床和病理资料。根据RP后Gleason分级情况分为升高组(n=42)和非升高组(n=36),采用单因素和多因素Logistic回归分析筛选GGU的相关因素,建立预测列线图模型并进行内部验证。
      结果  本研究中RP后Gleason分级升高42例(53.8%),不变24例(30.8%),下降12例(15.4%)。多因素Logistic回归分析显示,RP前内分泌治疗(OR=3.888,P=0.015)、游离前列腺特异性抗原(FPSA)(OR=1.185,P=0.036)、淋巴细胞(OR=0.271,P=0.027)与GGU相关。以上述因素构建预测列线图模型并进行内部验证,该模型的曲线下面积(AUC)为0.779(95% CI为0.677~0.881),C指数为0.657,平均绝对误差为2.0%。
      结论  在前列腺穿刺活检Gleason分级≤4级的患者中,RP术前有内分泌治疗史、FPSA越高、淋巴细胞越低者,RP术后GGU的可能性越大。基于上述因素建立的预测列线图模型具有一定的临床应用价值。

     

    Abstract:
      Objective  To analyze the related factors for Gleason grade upgrading (GGU) after radical prostatectomy (RP) in patients with Gleason grade ≤ 4 by biopsy, and to establish a predictive Nomogram model.
      Methods  Clinical and pathological materials of 78 patients with Gleason grade ≤ 4 by biopsy and RP therapy in Henan Provincial People's Hospital from January 2017 to January 2020 were analyzed retrospectively. According to the Glason grade after RP, the patients was divided into elevated group (n=42) and non-elevated group (n=36). Univariate and multivariate Logistic regression analyses were used to analyze the associated factors for GGU, and the predictive Nomogram model was established on the basis of internal validation.
      Results  After receiving RP treatment, the Gleason grade were upgraded in 42 cases (53.8%), invariant in 24 cases (30.8%) and downgraded in 12 cases (15.4%). Multivariate Logistic regression analysis showed that the endocrine therapy before RP (OR=3.888, P=0.015), free prostate specific antigen (FPSA) (OR=1.185, P=0.036) and lymphocyte (OR=0.271, P=0.027) were correlated with GGU. Based on the above factors, the predictive Nomogram model was established and verified internally, the area under the curve (AUC) of the model was 0.779 (95%CI, 0.677 to 0.881), the C-index was 0.657, and the average absolute error was 2.0%.
      Conclusion  In patients with Gleason grade ≤ 4 by prostate biopsy, the patients with history of endocrine therapy before RP, increased FPSA and lower lymphocyte are more likely to suffer GGU after RP. The predictive Nomogram model based on the above factors has a certain clinical application value.

     

/

返回文章
返回