术前炎性指标对PSA灰区且PI-RADS评分3分患者患临床有意义前列腺癌的预测价值

Efficacy of preoperative inflammatory markers in predicting clinically significant prostate cancer in patients with gray zone PSA levels and PI-RADS 3 lesions and a novel model construction

目的评估术前炎性指标对前列腺特异性抗原(PSA)灰区且PI-RADS评分3分的患者预测临床有意义前列腺癌的效能并构建预测模型。方法回顾性分析2018年10月至2019年12月期间苏州大学附属第一医院收治的前列腺穿刺患者信息。研究共纳入患者364例,根据穿刺结果分为阳性组(临床有意义前列腺癌)50例及阴性组(临床无意义前列腺癌及前列腺增生)314例,比较两组间临床病理参数的差异,采用logistic单因素及多因素分析,寻找临床有意义前列腺癌的独立危险因素,绘制诺莫图构建预测模型,并借助受试者工作特征(ROC)曲线检验其效能。结果阳性、阴性两组间年龄、直肠指检(DRE)、总前列腺特异性抗原(tPSA)、游离前列腺特异性抗原比值(f/tPSA)、前列腺特异性抗原密度(PSAD)、中性粒淋巴细胞比(NLR)、血小板淋巴细胞比(PLR)、淋巴单核细胞比(LMR)、系统免疫炎症指数(SII)方面差异有统计学意义(P<0.05)。将阳性、阴性两组间有差异的临床病理特征因素进行logistic单因素及多因素分析,发现年龄(OR=0.909,95%CI 0.845~0.977)、DRE(OR=0.087,95%CI 0.025~0.301)、f/tPSA(OR=10530810.891,95%CI 2105.692~52665822743.053)、PSAD(OR=0.000,95%CI 0.000~0.038)、PLR(OR=0.973,95%CI 0.954~0.991)及LMR(OR=7.473,95%CI 3.376~16.541)为诊断临床有意义前列腺癌的独立危险因素(P<0.05)。绘制logistic多因素模型的诺莫图及calibration校准曲线,模型拟合度较好(χ^(2)=8.072,P=0.427)。ROC曲线结果显示,全模型AUC最高(0.971),与其他模型相比,差异具有统计学意义(P<0.05)。决策曲线分析显示,全模型的临床净获益均高于其他模型。结论年龄、DRE阳性、f/tPSA、PSAD、PLR及LMR与临床有意义前列腺癌有关,且由此构建的诊断模型有较强的诊断效能。

Objective To evaluate the efficacy of preoperative inflammatory indexes in predicting clinically significant prostate cancer in patients with prostate-specific antigen(PSA)ranged from 4-10 ng/mL and PI-RADS=3 lesions and to construct a prediction model.Methods The information of patients with prostate biopsy admitted to the First Affiliated Hospital of Soochow University from October 2018 to December 2019 was retrospectively reviewed,and the patients were divided into positive group(clinically significant prostate cancer)and negative group(clinically insignificant prostate cancer and benign prostate hyperplasia)according to the biopsy results.The differences in clinicopathological parameters were compared between the two groups,and logistic univariate and multivariate analyses were performed on factors with statistical significance to identify independent risk factors in order to draw a nomogram to construct a prediction model.Finally,the efficacy of the prediction model was tested using receiver operating characteristic(ROC)curve.Results A total of 364 patients were enrolled in the study,including 50 in the positive group and 314 in the negative group.Univariate analysis indicated that digital rectal examination(DRE),age,total prostate specific antigen(tPSA),free-to-total prostate-specific antigen ratio(f/tPSA),prostatespecific antigen density(PSAD),neutrophil lymphocyte ratio(NLR),platelet lymphocyte ratio(PLR),lymphomonocyte ratio(LMR),and systemic immunoinflammatory index(SII)were correlated with clinically significant prostate cancer(P<0.05).Multivariate analysis revealed that age(OR=0.909,95%CI 0.845-0.977),DRE(OR=0.087,95%CI 0.025-0.301),f/tPSA(OR=10530810.891,95%CI 2105.692-52665822743.053),PSAD(OR=0.000,95%CI 0.000-0.038),PLR(OR=0.973,95%CI 0.954-0.991),and LMR(OR=7.473,95%CI 3.376-16.541)were independent risk factors for clinically significant prostate cancer(P<0.05).A nomogram was drawn based on the independent risk factors and calibration curves showed good fitness(χ^(2)=8.072,P=0.427).The full model had the highest AUC(0.971)in comparison with other models(P<0.05).Decision curve analysis showed that the clinical net benefit of the full model was higher than that of other models.Conclusions Age,DRE,f/tPSA,PSAD,PLR and LMR are correlated with clinically significant prostate cancer,and the diagnostic model established has strong diagnostic efficacy.