磁共振扩散加权成像(DWI)评估前列腺癌Gleason评分(GS):单指数模型与峰度模型比较

 目的 探讨扩散加权成像（diffusion-weighted imaging，DWI）中单指数模型和峰度模型在评估前列腺癌的Gleason评分（Gleason score，GS）方面的应用及差别。方法 连续收集复旦大学附属中山医院2017年5月至2018年12月期间疑诊前列腺疾病患者，进行多参数磁共振成像（multiparametric magnetic resonance imaging，mpMRI），包括常规序列、DWI、DCE-MRI。使用后处理软件分析DWI图像分别获得单指数模型参数表观扩散系数（apparent diffusion coefficient，ADC）及峰度模型参数Kapp及校正后的ADC-Dapp。使用Spearman相关分析检验癌区DWI参数与GS的相关性；单因素方差分析和独立样本t检验用于三组及两组间参数差异显著性分析，使用Kruskal-Wallis单因素方差分析（即κ样本）及Mann-Whitney U检验分别行三组及两组间非参数检验分析；通过ROC曲线来评估各参数鉴别诊断不同GS分组的效能。结果 参数ADC、Dapp均与肿瘤的GS存在负相关，Kapp与肿瘤的GS呈正相关。3个参数在GS≤3+4和GS>3+4、GS≤3+4和GS>3+4组间均存在显著差异（P均<0.05），Kapp仅在GS=4+3和GS>7组间差异有统计学意义（P<0.01）；ADC、Dapp、Kapp鉴别诊断GS≤3+4和GS>3+4组的AUC分别为0.751、0.756和0.785，AUC间差异无统计学意义（P=0.58~0.76）。结论 DWI单指数模型参数和峰度模型参数均有助于术前评估前列腺癌的Gleason分级；Kapp可能是肿瘤侵袭性的优秀预测因子，推荐将扩散峰度成像应用于临床。

A comparative study of monoexponential model and diffusion kurtosis model based diffusion-weighted imaging (DWI) for evaluating Gleason score (GS) of prostate cancer

Objective To compare the utility of various metrics derived from monoexponential model and diffusion kurtosis model based diffusion-weighted imaging (DWI) in evaluating Gleason score (GS) of prostate cancer. Methods From May 2017 to Dec.2018 in Zhongshan hospital,Fudan University patients with clinically suspected prostate lesions underwent multiparametric magnetic resonance imaging (mpMRI),including routine sequence,DWI,dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI).DWI data were postprocessed using monoexponential and DK models to quantify the apparent diffusion coefficient (ADC),apparent kurtosis coefficient (Kapp),and corrected ADC related to Gaussian behavior (Dapp).Spearman correlation analysis was used to analyze the relationship between those parameters and the GS.Among the three groups and the two groups,one-way Analysis of Variance (ANOVA) and independent sample t test were used to analyze the difference of parameters,and Kruskal-Wallis one-way ANOVA (k sample) and Mann-Whitney u test were used to perform nonparametric test analysis.The ROC curve was used to analyze the differential diagnosis ability of parameters. Results Both ADC and Dapp were negatively correlated with the GS of prostate cancer,while Kapp was positively correlated with the GS of PCa.The three parameters all had statistical differences between GS ≤ 3+4 and GS>3+4,GS ≤ 3+4 and GS>3+4 (P<0.05 for all);only Kapp showed significant difference between GS=4+3 and GS>7 groups (P<0.01).The AUC of ADC,Dapp and Kapp in differential diagnosis of GS ≤ 3+4 and GS>3+4 were 0.751,0.756 and 0.785,respectively,but differences among AUCs were not statisticant (P=0.58-0.76). Conclusions DWI parameters derived from ME model and DK model were helpful for preoperative evaluation of Gleason grading of prostate cancer;Kapp may be an excellent predictor of tumor aggressiveness,and diffusion kurtosis imaging is recommended for clinical use.