基于XD-VIBE的DCE-MRI在预测局部进展期胃癌组织病理学特征中的应用价值

The Application Value of DCE-MRI Based on XD-VIBE in Predicting Histopathological Features of Locally Advanced Gastric Cancer

目的探讨基于额外维度容积内插屏气检查(XD-VIBE)的动态对比增强磁共振成像(DCE-MRI)在预测胃癌组织病理学特征中的应用价值。方法前瞻性纳入2021年12月至2023年6月采用XD-VIBE序列进行DCE-MRI检查且经术后病理证实为局部进展期胃癌的患者。两名放射科医师在不知病理信息情况下分别对病灶进行测量,测量的定量参数如下:容积转运常数(K^(trans))、血管外细胞外间隙容积分数(V_(e))、逆流速率常数(K_(ep))、血浆分数体积(V_(p))以及体积。基于术后不同的病理组织学亚型,获得肿瘤侵袭性的总组织病理学特征评分,并将其分为高危组及低危组。采用组内相关系数(ICC)评估DCE-MRI参数的观察者间一致性。采用t检验及Mann-Whitney U检验,比较胃癌组织病理学高危及低危组间的DCE-MRI定量参数特征差异,P<0.05为差异具有统计学意义,将差异有统计学意义的参数纳入Logistic回归分析,建立多参数MRI预测模型。结果191例局部进展期胃癌患者于MRI检查后行根治性胃切除术且取得术后病理资料,其中低危组122例、高危组69例。体积、K^(trans)mean、K^(trans)median、V_(e)mean、V_(e)median、V_(p)mean、V_(p)median值在鉴别总组织病理学特征高危组与低危组间均有统计学差异。通过Logistic回归分析建立DCE-MRI定量参数预测组织病理学特征模型,体积、K^(trans)mean、V_(e)median是独立预测因子,模型的敏感度、特异度、阳性预测值及阴性预测值分别是87.0%、37.7%、87.7%、55.1%,并绘制受试者工作特征(ROC)曲线,曲线下面积(AUC)为0.811(95%CI:0.748~0.874)。结论基于XD-VIBE的DCE-MRI定量参数在术前预测局部进展期胃癌组织病理学特征中具有较好的效能。

Objective The application value of DCE-MRI based on XD-VIBE in predicting the histopathological feature of gastric cancer.Methods Patients with locally advanced gastric cancer confirmed by postoperative pathology who underwent DCE-MRI with XD-VIBE sequence in the First Affiliated Hospital of Nanjing Medical University from December 2021 to June 2023 were prospectively enrolled.Two radiologists independently assessed the lesions without any pathological information.The following quantitative parameters were measured:K^(trans),V_(e),K_(ep),V_(p),and volume.Based on different histopathological subtypes observed after surgery,a total histopathological score of tumor invasiveness was obtained and used to categorize patients into high-risk and low-risk groups.The inter-observer agreement of DCE-MRI parameters was assessed using the Intraclass correlation coefficient(ICC).The differences in quantitative parameters of DCE-MRI between the high risk and low risk groups of gastric cancer histopathology were compared using t-test and Mann-Whitney U test.Statistical significance was considered at P<0.05.The parameters showing statistically significant differences were included in a Logistic regression analysis to establish a multi-parameter MRI prediction model.Results A total of 191 patients diagnosed with locally advanced gastric cancer underwent radical gastrectomy after undergoing MR Examination.Postoperative pathological data were collected,including 122 cases in the low-risk group and 69 cases in the high-risk group.The values of volume,K^(trans)mean,K^(trans)median,V_(e)mean,V_(e)median,V_(p)mean,and V_(p)median were significantly different between the high-risk and low-risk groups of overall histological features.Volume,K^(trans)mean,and V_(e)median were independent predictors.The Logistic regression model demonstrated a sensitivity of 87.0%,specificity of 37.7%,positive predictive value of 87.7%,and negative predictive value of 55.1%.Additionally,the area under the ROC curve was calculated to be 0.811(95%CI:0.748-0.874).Conclusion The quantitative parameters obtained from XD-VIBE-based DCE-MRI exhibit robust effectiveness in preoperative prediction of the histopathological features of locally advanced gastric cancer.