ADC直方图分析在诊断外周带临床显著性前列腺癌及判断其恶性程度中的价值研究

The value of apparent diffusion coefficient histogram in detecting clinically significant peripheral prostate cancer and assessing tumor aggressiveness

目的探讨表观扩散系数(apparent diffusion coefficient,ADC)直方图参数对外周带临床显著性前列腺癌(clinically significant peripheral prostate cancer,csPCa)的诊断价值及其与前列腺癌(prostate cancer,PCa)病理分级分组的相关性。方法回顾性分析延安大学附属医院泌尿外科疑诊为PCa并行超声引导下前列腺穿刺活检的患者107例,根据穿刺活组织病理学检查结果将其分为csPCa组33例和非csPCa组74例。比较两组临床指标及ADC直方图参数。采用单因素及多因素Logistic逐步回归分析筛选csPCa的独立预测因子。绘制受试者工作特征曲线(receiver operating characteristic curve,ROC)分析并比较回归模型预测概率值与各独立预测因子的诊断效能。Spearman相关检验分析ADC直方图参数与前列腺癌病理学分级分组之间的相关性。结果两组总前列腺特异性抗原、游离前列腺特异性抗原、前列腺特异性抗原密度、PI-RADS评分、ADC直方图参数(ADCmin、ADCmax、ADCmean、ADCmedian、ADC5、ADC10、ADC25、ADC50、ADC75、ADC90、ADC95、偏度、峰度及熵)比较差异均有统计学意义(均P<0.05)。Logistic回归结果显示,ADC25、ADCmedian及ADCmax是csPCa的独立预测因子(OR值分别为1.013、0.972、1.004,P<0.05)。ROC曲线分析结果显示预测概率值的曲线下面积(area under the curve,AUC)为0.943,截断值为0.269,敏感度为0.909,特异度为0.892,其AUC均高于各独立预测因子。ADC直方图参数ADC10、ADC25、ADC50、ADC75、ADC90、ADCmin、ADCmean、ADCmedian与前列腺癌病理学分级分组存在相关性(P<0.05)。ADCmean与病理学分级分组的相关性最强(r=-0.414)。结论ADC直方图参数及其所建立的预测模型对csPCa有较高的诊断效能,而且ADC直方图参数对进一步判断PCa的病理学分级也有一定的价值,可以帮助泌尿外科医师准确判断病情并精准干预。

Objective To explore the diagnostic value of apparent diffusion coefficient(ADC)histogram parameters in clinically significant peripheral prostate cancer(csPCa)and its correlation with pathological grading of prostate cancer(PCa).Methods A retrospective analysis was made on 107 patients suspected of PCa and ultrasoundguided prostate biopsy in urology department,Affiliated Hospital of Yan´an University.According to the pathological results of ultrasound-guided prostate biopsy,they were divided into csPCa group(n=33)and non-csPCa group(n=74).The clinical indexes and ADC histogram parameters of the two groups were compared.Univariate and multivariate logistic stepwise regression analysis were used to screen the independent predictors of csPCa.The receiver operating characteristic(ROC)curve was drawn and the predictive probability value of the regression model was compared with the diagnostic efficiency of each independent predictor.Spearman correlation test was used to analyze the correlation between ADC histogram parameters and pathological grading of prostate cancer.Results There were significant differences in total prostate specific antigen(tPSA),free prostate specific antigen(PSA),prostate specific antigen density,PI-RADS score,ADC histogram parameters(ADCmin,ADCmax,ADCmean,ADCmedian,ADC5,ADC10,ADC25,ADC50,ADC75,ADC90,ADC95,skewness,kurtosis and entropy)between the two groups(all P<0.05).The results of logistic regression showed that ADC25,ADCmedian and ADCmax were independent predictors of csPCa(OR=1.013,0.972,1.004,P<0.05).The results of ROC curve analysis showed that the area under the curve(AUC),cutoff value,sensitivity and specificity of the predicted probability value were 0.943,0.269,0.909 and 0.892,respectively.The AUC of the predicted probability value was higher than that of each independent predictive factor.ADC histogram parameters ADC10,ADC25,ADC50,ADC75,ADC90,ADCmin,ADCmean,ADCmedian were correlated with pathological grade of prostate cancer.The correlation between ADCmean and pathological grading was the strongest(r=-0.414).Conclusion Based on the ADC histogram parameters of the established prediction model,the diagnostic efficiency of csPCa is high,and the ADC histogram parameters also have a certain value in further judging the pathological grade of PCa,which can help urologists to accurately judge the disease and accurate intervention.