影像科学与光化学2026，Vol.44Issue(3)：42-48,55,8.DOI:10.7517/issn.1674-0475.2026.03.05
DCE-MRI定量参数与直肠癌分化程度的关系及预测术后局部复发的价值
DCE-MRI Biomarkers:Correlation with Histologic Differentiation Grade and Predictive Performance for Local Recurrence in Rectal Cancer
摘要
Abstract
Objective:To analyze the relationship between dynamic contrast-enhanced magnetic resonance imaging(DCE-MRI)parameters and rectal cancer differentiation grade,and their predictive value for postoperative local recurrence.Methods:A prospective study was conducted,enrolling 205 rectal cancer patients admitted to our hospital between April 2021 and April 2022.All patients underwent DCE-MRI examinations with measurement of the volume transfer constant(Ktrans),rate constant(Kep),and extracellular extravascular volume fraction(Ve).The relationships between these parameters and tumor differentiation grade were analyzed using bivariate correlation(Kendall's Tau-b),Cox regression,Nomogram modeling,and ROC curves to evaluate their predictive value for recurrence.Results:Kendall's Tau-b correlation analysis revealed significant negative correlations between Ktrans,Kep,Ve,and the differentiation degree of rectal cancer patients(r=-0.739,-0.641,-0.635,all P<0.001).The local recurrence group exhibited significantly higher proportions of poorly differentiated tumors[75.75%(25/33)vs.10.47%(18/172)]and T4 stage[39.39%(13/33)vs.19.77%(34/172)],as well as elevated levels of Ktrans[(0.74±0.15)min-1 vs.(0.53±0.14)min-1],Kep[(1.51±0.26)min-1 vs.(1.29±0.25)min-1],and Ve(0.49±0.13 vs.0.41±0.12)compared to the non-recurrence group(all P<0.05).Cox regression analysis identified high Ktrans(HR=82.869,P<0.001),high Kep(HR=7.242,P=0.005),high Ve(HR=16.852,P=0.043),and poor differentiation(HR=3.919,P<0.001)as risk factors for postoperative local recurrence in rectal cancer patients.A Nomogram prediction model for local recurrence was constructed based on Ktrans,Kep,Ve,and validated via Bootstrap resampling,demonstrating a C-index of 0.909.ROC curve analysis showed that the model achieved an AUC of 0.909(95%CI:0.589～0.959,P<0.001)in predicting postoperative local recurrence,with a cut-off value of 33.80 points,specificity of 0.837,sensitivity of 0.818,and Youden's index of 0.655.Conclusion:Ktrans,Kep,Ve are negatively correlated with the differentiation degree of rectal cancer patients,Ktrans,Kep,Ve and the degree of differentiation are influencing factors for local recurrence in rectal cancer patients after surgery.A Nomogram model based on Ktrans,Kep,Ve can effectively predict the risk of local recurrence in rectal cancer patients after surgery,providing imaging evidence for clinical stratified intervention.关键词
直肠癌/磁共振动态对比增强/复发/分化程度Key words
rectal cancer/dynamic contrast-enhanced magnetic resonance imaging/recurrence/differentiation grade分类
医药卫生引用本文复制引用
郭涛,潘敦,彭万英,蒋林波,敬博..DCE-MRI定量参数与直肠癌分化程度的关系及预测术后局部复发的价值[J].影像科学与光化学,2026,44(3):42-48,55,8.基金项目
2023年四川省医学青年创新科研课题项目(Q23009). （Q23009）
