基于^(18)F-FDG PET/MR成像和扩散加权成像分数微积分模型对肺腺癌增殖状态评估价值的研究

The value of monoexponentia,fractional order calculus models and ^(18)F-FDG PET imaging in evaluating the proliferation status of lung adenocarcinoma

目的探讨基于扩散加权成像(diffusion weighted imaging,DWI)的单指数、分数微积分(fractional order calculus,FROC)模型和氟代脱氧葡萄糖正电子发射断层扫描(^(18)F-fluorodeoxyglucose-positron emission tomography,^(18)F-FDG PET)在评估肺腺癌增殖状态中的价值。材料与方法选取经我院病理证实的64例肺腺癌患者,以Ki-67表达25%为界,>25%为高Ki-67组,≤25%为低Ki-67组。所有患者在治疗前均行肺部^(18)F-FDG PET/MR检查,其中DWI采取10个b值(0~1000 s/mm^(2))扫描。比较两组间表观扩散系数(apparent diffusion coefficient,ADC)、空间变量(a microstructural quantity,μ)、扩散系数(diffusion coefficient,D)、分数空间导数(fractional order parameter,β)、最大标准摄取值(maximum standardized uptake value,SUV_(max))有无显著差异。通过多因素logistic回归分析Ki-67增殖状态的独立预测因素,采用受试者工作特征曲线(receiver operating characteristic curve,ROC)评估鉴别效能,并分析各参数与Ki-67之间的相关性。结果低Ki-67组的ADC、D、β显著大于高Ki-67组(P<0.05),高Ki-67组的μ、SUV_(max)显著大于低Ki-67组(P<0.05)。参数D和SUV_(max)曲线下面积分别为0.873和0.727,且多因素logistic回归显示D值(OR:0.421,95%CI:0.245~0.723,P=0.002)和SUV_(max)值(OR:1.022,95%CI:1.002~1.042,P=0.031)是Ki-67高表达的独立危险因素。ADC值和D值与Ki-67呈负相关(r=-0.361,r=-0.420),μ和SUV_(max)值与Ki-67呈正相关(r=0.369,r=0.527)。结论单指数、FROC模型和^(18)F-FDG PET均是评估肺腺癌增殖状态的有效手段,其中FROC模型的D值具有最高的诊断效能。FROC模型为探索肿瘤组织微环境信息提供了新的视角,在无创评估肺腺癌增殖状态方面具有很大潜力,其临床应用前景广阔。

Objective:To explore the value of monoexponential,fractional order calculus(FROC)models based on diffusion weighted imaging(DWI)and ^(18)F-fluorodeoxyglucose-positron emission tomography(^(18)F-FDG PET)in assessing the proliferation status of lung adenocarcinoma.Materials and Methods:Atotal of 64 patients with lung adenocarcinoma confirmed by pathology in our hospital were included.The expression of Ki-67 in lung cancer tissues was detected by immunohistochemistry and divided into the high Ki-67 group(>25%)and the low Ki-67 group(≤25%).Before treatment,all patients underwent a dedicated thoracic ^(18)F-FDG PET/MR examination.The DWI was scanned with 10 b-values(0-1000 s/mm^(2)).The apparent diffusion coefficient(ADC),a microstructural quantity(μ),diffusion coefficient(D),fractional order parameter(β)and maximum standardized uptake value(SUV_(max))were compared between the two groups.The independent predictors of Ki-67 proliferative status were analyzed by multivariate logistic regression,receiver operating characteristic(ROC)curve was used to evaluate the discriminant performance,and the correlation between each parameter and Ki-67 was analyzed.Results:TheADC,D,andβvalues in the low Ki-67 group were significantly higher than in the high Ki-67 group(P<0.05),and theμand SUV_(max) values in the high Ki-67 group were significantly higher than in the low Ki-67 group(P<0.05).The area under the curve(AUC)of parameters D and SUV_(max) were 0.873 and 0.727,respectively,and multivariate logistic regression showed that parameters D(OR:0.421,95%CI:0.245-0.723,P=0.002)and SUV_(max)(OR:1.022,95%CI:1.002-1.042,P=0.031)were independent risk factors for high Ki-67 expression.ADC and D values were negatively correlated with Ki-67(r=-0.361,r=-0.420),andμand SUV_(max) values were positively correlated with Ki-67(r=0.369,r=0.527).Conclusions:Monoexponential,FROC models and ^(18)F-FDG PET are effective methods to evaluate the proliferation status of lung adenocarcinoma,and the D value of FROC model shows the highest diagnostic performance.FROC model provides a new perspective for exploring the information of tumor tissue microenvironment,and has great potential in non-invasive evaluation of lung adenocarcinoma proliferation,and its clinical application has broad prospects.