水通道蛋白MRI在前列腺癌风险评估中的价值及其与水通道蛋白1的相关性研究  

The value of aquaporin MRI in risk assessment of prostate cancer and its correlation with aquaporin 1

作　　者：刘昀松 陈丽华[1] 张钦和 王楠[1] 武志刚 孙鹏 王家正 刘爱连[1,3] LIU Yun-song;CHEN Li-hua;ZHANG Qin-he;WANG Nan;WU Zhi-gang;SUN Peng;WANG Jia-zheng;LIU Ai-lian(Department of Radiology,the First Affiliated Hospital of Dalian Medical University,Dalian Liaoning 116011,China;Philips Healthcare,Beijing 100016,China;Dalian Medical Imaging Artificial Intelligence Engineering Technology Research Center,Dalian Liaoning 116011,China)

机构地区：[1]大连医科大学附属第一医院放射科,辽宁大连11601 [2]飞利浦医疗科,北京100016 [3]大连市医学影像人工智能工程技术研究中心,辽宁大连11601

出　　处：《中国临床医学影像杂志》2022年第8期533-539,共7页Journal of China Clinic Medical Imaging

摘　　要：目的:探讨基于多个超高b值的水通道蛋白(AQP)MRI在前列腺癌(PCa)风险预测中的价值及其与水通道蛋白1(AQP1)的相关性。方法:回顾性收集行3.0T MRI检查且经手术病理证实的PCa患者43例,分为高风险组26例、低风险组17例。均包括单b值(0、1 000 s/mm^(2))DWI和多个超高b值(2 000 s/mm^(2)、3 000 s/mm^(2)、4 000 s/mm^(2))AQP MRI序列。免疫组化检测AQP1染色的平均光密度(AOD)。两名观察者独立测量ADC值和AQP-ADC值。采用组内相关系数(ICC)评价两名观察者测量的一致性;采用两独立样本t检验或Mann-Whitney U检验分析高低风险组间AQP1-AOD值、AQP-ADC值、ADC值的差异;AQP-ADC值、ADC值与AQP1-AOD值间相关性检验采用线性回归分析;采用Cocor分析比较相关系数间的差异;采用多因素Logistic回归分析高风险PCa的独立影响因素;采用受试者工作特征(ROC)曲线评价各参数评估PCa风险性的价值;采用Delong检验对参数间的ROC曲线下面积(AUC)进行比较。结果:高风险组的AQP1-AOD值和AQP-ADC值均大于低风险组(AQP1-AOD值:(0.33±0.03) vs (0.29±0.02);AQP-ADC值(×10-3mm^(2)/s):(0.30±0.05) vs (0.25±0.02));高风险组的ADC值小于低风险组((1.06±0.17) vs (1.21±0.17))(P值均<0.01)。高、低风险组AQP-ADC值的方差分别为0.25×10^(-2)、0.04×10^(-2),ADC值的方差分别为3×10^(-2)、3×10^(-2),AQP-ADC值离散程度更小;AQP-ADC值、ADC值与AQP1-AOD值的相关系数分别为0.44和-0.34,决定系数R2分别为0.20和0.11,AQP-ADC值与AQP1-AOD值间的模型拟合度更好;AQP-ADC值、ADC值均是高风险PCa的独立危险因素;ADC值、AQP-ADC值及AQP1-AOD值评估PCa风险分组的AUC分别为0.75、0.86和0.84,敏感度分别为69%、73%和62%,特异度分别为77%、82%和100%;ADC值、AQP-ADC值间的敏感度和特异度均无统计学差异(P=1,χ~2=0)。结论:AQP MRI能够很好地评估PCa的风险性,AQP-ADC值较ADC值能更好地反映AQP1表达。Objective:To investigate the value of aquaporin(AQP) MRI based on multiple ultra-high b values in the risk prediction of prostate cancer(PCa) and its correlation with aquaporin 1(AQP1).Methods:Forty-three PCa patients,confirmed by surgical pathology,who underwent prostate 3.0T MRI examinations(including DWI(0,1 000 s/mm^(2)) and multiple ultra-high b values DWI(2 000 s/mm^(2),3 000 s/mm^(2),4 000 s/mm^(2)) sequences) preoperatively in our hospital,were enrolled in the retrospective study.All the patients were divided into two groups,including 26 cases in the high-risk group and 17 cases in the low-risk group.Immunohistochemistry(IHC) technique was used to detect the values of average optical density(AOD) of AQP1 staining parts of tissues.ADC values and AQP-ADC values were independently measured by two observers.ICC was used by two observers to verify the consistency of the data.Two independent sample t test or Mann-Whitney U test were used to analyze the differences of AQP1-AOD values,AQP-ADC values and ADC values between the high and low risk groups.Linear regression analysis was used to test the correlation between AQP-ADC value,ADC value and AQP1-AOD value.Cocor test was used to compare the difference between correlation coefficients.Multivariate Logistic regression was used to analyze the independent influencing factors of high risk PCa.ROCs were used to evaluate the risk of PCa.Delong test was used to compare AUCs between the parameters.Results:The AQP1-AOD values and AQP-ADC values in high-risk group were higher than those in low-risk group(AQP1-AOD values:(0.33 ±0.03) vs(0.29 ±0.02);AQP-ADC values(×10-3mm^(2)/s):(0.30 ±0.05) vs(0.25±0.02)),while ADC value(10-3mm^(2)/s) in high-risk group was lower than that in low-risk group((1.06±0.17) vs(1.21 ±0.17))(all P <0.01).The variance of AQP-ADC value in the high and low risk groups were 0.25 ×10^(-2)and 0.04 ×10^(-2),respectively,and the variance of ADC value was 3×10^(-2)and 3×10^(-2),respectively.The correlation coefficients between AQPADC values an

关 键 词：前列腺肿瘤 水通道蛋白质1 磁共振成像 

分 类 号：R737.25[医药卫生—肿瘤] R445.2[医药卫生—临床医学]