ADC图纹理参数在预测局部进展期直肠癌新辅助放化疗抗拒的应用研究

Application Value of ADC Textures in Predicting Resistance to Neoadjuvant Chemoradiotherapy in Locally Advanced Rectal Cancer

目的探讨ADC图纹理参数在预测局部进展期直肠癌新辅助放化疗(NCRT)抗拒中的应用价值。方法回顾性分析接受NCRT和根治手术且NCRT前MRI数据齐全的局部进展期直肠癌患者资料共53例。对治疗前ADC图进行后处理,提取肿瘤区域的133个纹理参数。根据术后病理分期和NCRT前MRI分期的对比结果将患者分为对NCRT抗拒组和有效组。对比纹理参数在两组间的差异,将有统计学意义的参数纳入逻辑回归模型,确定NCRT抗拒的独立预测指标,采用ROC曲线分析,评估其对NCRT抗拒的预测效能。结果 17个ADC图纹理参数在两组间差异有统计学意义,通过多因素逻辑回归分析,能量方差和Sd Ga47两个纹理参数可作为NCRT抗拒的独立预测指标,利用该逻辑回归模型对NCRT抗拒进行预测的ROC曲线下面积为0. 889。结论基于ADC图的纹理参数有助于在治疗前筛选出直肠癌NCRT抗拒患者。

Objective To evaluate the application value of ADC textures in predicting resistance to neoadjuvant chemoradiotherapy(NCRT)in locally advanced rectal cancer(LARC).Methods A total of 53 LARC patients who underwent diffusion-weighted imaging before NCRT were enrolled.Textures of ADC imaging of the primary tumor were extracted for each patient,then were compared between resistant and responsive groups.Logistic regression was used to determine the independent predictors for NCRT-resistance.Receiver operating characteristic curve(ROC)was performed to evaluate the predictive performance by a logistic model consisting of the independent predictors.Results Of the 133 parameters,seventeen ADC textures significantly differed between resistant and responsive groups(P<0.05).Furthermore,energy variance and Sd Ga47 were identified as independent predictors for resistance to NCRT;this logistic model achieved an area under the curve(AUC)of 0.889.Conclusion Texture analysis based on pre-therapy ADC imaging could potentially be helpful to predict resistance to NCRT in LARC.