摘要
目的探讨扩散峰度成像(DKI)定量参数在预测直肠癌KRAS基因突变中的应用价值。方法回顾性分析山西省肿瘤医院2016年11月—2018年6月经病理证实的152例直肠腺癌患者的资料,其中男92例,女60例;年龄33~86岁,平均61岁。患者均于术前行常规MRI和功能DKI序列检查,由2名放射科医师双盲勾画感兴趣区,应用MatLab软件计算两组患者DKI定量参数平均表观扩散系数(MD)、平均峰度(MK)以及表观扩散系数(ADC),并采用组内相关系数进行一致性分析。术后均进行KRAS基因检测,依据检测结果将患者分为野生组和突变组两组,采用独立样本t检验对比两组患者的MD、MK、ADC。绘制DKI定量参数诊断KRAS基因突变的受试者工作特征曲线(ROC),根据约登指数确定各定量参数的最佳诊断阈值,以及相应的灵敏度、特异度,并采用DeLong检验比较各定量参数的ROC的曲线下面积(AUC)。结果152例直肠癌患者中,KRAS基因突变组74例,野生组78例,基因突变率为48.6%(74/152)。突变组患者的ADC、MD、MK值分别为(1.18±0.18)×10^-3 mm^2/s、(1.28±0.18)×10^-3 mm^2/s、0.97±0.11,野生组分别为(1.33±0.18)×10^-3 mm^2/s、(1.42±0.17)×10^-3 mm^2/s、0.82±0.09;突变组ADC、MD值均小于野生组患者,而MK值则大于野生组患者,差异均有统计学意义(t=5.424、4.882、-8.809,P值均<0.01)。ROC曲线显示,ADC、MD、和MK值预测KRAS基因的AUC分别为0.758、0.740、0.845,灵敏度分别为75.7%、82.4%、77.0%,特异度分别为68.0%、57.8%、84.6%。DeLong检验结果显示,MK值的AUC明显高于ADC和MD值(P值均<0.01),而ADC值和MD值间AUC差异无统计学意义(P>0.05)。结论DKI定量参数MD、MK和ADC值,在预测直肠癌的KRAS基因突变方面均有一定的价值,其中MK值有更高的AUC和特异度,有更高的诊断价值。
Objective To investigate the applicable value of diffusion kurtosis imaging(DKI)in assessing the status of KRAS mutation in patients with rectal cancer.Methods The data of 152 patients with rectal cancer confirmed by pathology,in the Shanxi Province Tumor Hospital from November 2016 to June 2018,were retrospectively analyzed,including 92 males and 60 females,average age was 61 years(33-86 years).Patients underwent morphologic MRI and functional DKI imaging sequences before operation.The tumor region of interest(ROI)was delineated by two radiologist,and they were blind to KRAS mutational analysis.MatLab software was applied to calculated the quantitative parameters of DKI,including mean diffusivity(MD)and mean kurtosis(MK),as well as apparent diffusion coefficient(ADC)in two groups,respectively,and the consistency analyses were performed according to intraclass correlation coefficients(ICC).The patients were divided into KRAS wild type and mutant groups,according to the results of KRAS mutational analysis on surgical specimens.Student's t-test was used to compare the quantitative parameters of DKI between two groups.Receiver-operating characteristic curve(ROC)was generated to evaluate the diagnostic value of each parameter in predicting the mutation status of KRAS in rectal cancer.The optimal cut-off values of each quantitative parameter was chosen according to Youden's index.The corresponding sensitivity and specificity were also calculated.Differences in performance of them were analyzed by comparing the areas under the ROC curves(AUC)of the quantitative each parameters using the method developed by DeLong et al.Results For a total of 152 rectal lesions,the KRAS mutation rate was 48.6%,divide into mutant group(n=74),and wildtype group(n=78).The ADC and MD values as well as MK value were(1.18±0.18)×10^-3 mm^2/s,(1.28±0.18)×10^-3 mm^2/s,and 0.97±0.11 in the mutated group,respectively;while they were(1.33±0.18)×10^-3 mm^2/s,(1.42±0.17)×10^-3 mm^2/s,and 0.82±0.09 in the wildtype group,respectively.The ADC and MD values were significantly lower in the mutated group than those in the wildtype group,while MK value was higher in the mutated group compared with those in the wildtype group(t=5.424,4.882,-8.809,all P values<0.01).The AUC,sensitivity and specificity of ADC,MK and MD values for predicting KRAS mutations were 0.758,75.7%,68.0%,and 0.740,82.4%,57.8%,and 0.84,77.0%,84.6%,respectively,which indicated the MK value have higher specificity in predicting KRAS mutation status in rectal cancer compared to MD and ADC values,while MD showed the highest sensitivity.In addition,the AUC of MK value was higher than with the MD and ADC values(all P values<0.01).However,there was no significant difference between ADC value and MD value in terms of DeLong et al(P>0.05).Conclusions Both ADC and MD as well as MK metrics exhibit potential for predicting KRAS mutation status in rectal cancer.However,the MK value has higher AUC and specificity,which shows it has higher diagnostic value for analysis of KRAS mutation status in rectal cancer.
作者
李丹丹
崔艳芬
杨晓棠
Li Dandan;Cui Yanfen;Yang Xiaotang(Department of Medical Imaging,Shanxi Medical University,Taiyuan 030000,China;Department of Radiology,Shanxi Province Tumor Hospital,Taiyuan 030000,China)
出处
《中华解剖与临床杂志》
2020年第2期102-108,共7页
Chinese Journal of Anatomy and Clinics
基金
山西省应用基础研究计划(201801D121307)。
关键词
直肠肿瘤
KRAS基因突变
扩散峰度成像
表观扩散系数
Rectal neoplasms
KRAS mutation
Diffusion kurtosis imaging
Apparent diffusion coefficient