基于T1 mapping序列的定量参数鉴别肺癌病理类型的应用研究

Application of quantitative parameters based on T1 mapping sequence in identifying pathological types of lung cancer

目的评估基于纵向弛豫时间定量序列(T1 mapping)的定量参数预测肺癌病理类型的可行性。材料与方法共收治经病理确诊的肺癌患者117例,包括腺癌62例,鳞状细胞癌26例,小细胞肺癌(small cell lung cancer,SCLC)29例,于治疗前行常规MRI、T1 mapping检查。采用B1场校正的3D可变翻转角VIBE序列分别于增强前及增强后5 min采集T1 mapping图像,测量肿瘤大小、增强前T1值(T1pre)、增强后T1值(T1post),并计算增强前后T1值变化△T1、增强前后T1值变化率△T1%。采用SPSS和MedCalc软件进行分析,运用logistic回归联合ROC曲线下面积(area under the curve,AUC)评价各组各定量参数的差异及多参数联合的诊断价值。结果腺癌、鳞状细胞癌和SCLC的ΔT1、ΔT1%和T1post差异有统计学意义(P<0.05),而T1pre差异无统计学意义(P=0.506)。鉴别SCLC和非小细胞肺癌(non-small cell lung cancer,NSCLC)的T1post、△T1和△T1%值的AUC分别为0.856、0.805和0.864,3个参数的组合可以提高鉴别SCLC和NSCLC的诊断准确度(AUC=0.870,P<0.05)。△T1、△T1%鉴别腺癌和鳞状细胞癌的AUC分别为0.755和0.767,两者联合可提高诊断准确度(AUC=0.771,P>0.05)。T1post、△T1%鉴别鳞状细胞癌和SCLC的AUC分别为0.788和0.818,两者联合可提高诊断准确度(AUC=0.831,P>0.05)。T1post、△T1%鉴别腺癌和SCLC的AUC分别为0.895、0.873,二者联合可提高诊断准确度(AUC=0.898,P>0.05)。结论T1 mapping可以无创定量地获得腺癌、鳞状细胞癌和SCLC的T1值,可用于区分SCLC与NSCLC以及鳞状细胞癌和腺癌,具有一定的临床应用价值。

Objective:To evaluate the feasibility of quantitative parameters based on T1 mapping sequence in predicting the pathological types of lung cancer.Materials and Methods:A total of 117 lung cancer patients,including 62 cases of adenocarcinoma,26 cases of squamous cell carcinoma,29 cases of small cell lung cancer(SCLC),were enrolled in this study.Prior routine sequence scans,then the B1 field corrected variable flip angle VIBE sequence was used to acquire T1 mapping images.Afterwards,Gd-DTPA was used for dynamic enhanced scanning.T1 mapping images were collected 5 minutes before and after enhancement.Measure tumor size,T1 value before enhancement(T1pre),T1 value after enhancement(T1post),and calculateΔT1,ΔT1%.SPSS and MedCalc software were used for analysis the differential diagnostic value of each quantitative parameter in each group,logistic regression combined with area under the curve(AUC)was constructed to evaluate the diagnostic value of each quantitative parameter and multi-parameter combination.Results:There were statistically significant differences inΔT1,ΔT1%and T1post among adenocarcinoma,squamous cell carcinoma and SCLC(P<0.05),but no difference in T1pre(P=0.506).The AUCs of T1post,ΔT1,andΔT1%values to differentiate SCLC and non-small cell lung cancer(NSCLC)were 0.856,0.805 and 0.864,combination of the three parameters can improve the diagnostic accuracy of differentiating SCLC and NSCLC(AUC=0.870,P<0.05).The AUCs ofΔT1 andΔT1%values to differentiate adenocarcinoma and squamous cell carcinoma were 0.755 and 0.767,combination of the two could slightly improve the diagnostic accuracy(AUC=0.771,P>0.05).The AUCs ofΔT1%,T1post values to differentiate squamous cell carcinoma and SCLC were 0.788 and 0.818,combination of the two could improve the diagnostic accuracy(AUC=0.831,P>0.05).The AUCs ofΔT1%,T1post values to differentiate adenocarcinoma and SCLC is 0.895 and 0.873,combination of the two could improve the diagnostic accuracy(AUC=0.898,P>0.05).Conclusions:T1 mapping can non-invasively and quantitatively obtain the T1 value of adenocarcinoma,squamous cell carcinoma and small cell lung cancer,and can be used to distinguish SCLC from NSCLC,as well as squamous cell carcinoma and adenocarcinoma,provide more accurate histological correlations and prognostic value in lung cancer.