临床-病理-影像联合模型构建列线图预测乳腺癌新辅助治疗达病理完全缓解

Nomogram Based on Clinical-Pathological-Imaging Model for Predicting the Pathological Complete Response of Breast Cancer Patients Undergoing Neoadjuvant Chemotherapy

目的分析MRI与临床病理联合构建列线图预测乳腺癌新辅助治疗(NAC)达病理完全缓解(pCR)的价值。资料与方法回顾性收集宁夏医科大学总医院2018年1月—2022年5月接受NAC并手术的首诊女性乳腺癌患者,其中2018年1月—2021年5月就诊的159例患者为建模组,2021年6月—2022年5月69例为验证组。收集患者临床病理及NAC前MRI参数。通过单因素和多因素Logistic回归分析确定pCR的独立预后因素并构建列线图。通过受试者工作特征曲线及校准图评估列线图的诊断效能。结果肿瘤最大径线(OR=0.96,95%CI 0.93~0.99,P=0.01),HER-2状态(OR=0.07,95%CI 0.02~0.22,P<0.001),Ki-67指数(OR=1.04,95%CI 1.01~1.07,P=0.002)和信号强化率(OR=1.02,95%CI 1.01~1.03,P=0.002)是pCR的独立预后因素,其构建的列线图预测pCR具有较高的诊断效能(曲线下面积为0.82,95%CI 0.75~0.89)和校准能力(C指数=0.89,95%CI 0.83~0.94)。结论肿瘤最大径线、HER-2状态、Ki-67指数和信号强化率联合构建的列线图在预测NAC达pCR具有良好的诊断效能和校准能力。

Purpose To analyze the value of a nomogram based on MRI and clinicopathological variables to predict pathologic complete response(pCR)for breast cancer undergoing neoadjuvant chemotherapy.Materials and Methods In this single-center retrospective study,consecutive women with breast cancer who underwent neoadjuvant chemotherapy followed by surgery between January 2018 and May 2022 were considered for inclusion.The women were divided into a development cohort between January 2018 and May 2021 and a validation cohort between June 2021 and May 2022.A total of 159 women were in the development cohort and 69 in the validation cohort.Clinicopathological data were collected,and MRI data were obtained before neoadjuvant chemotherapy were analyzed.Univariate and multivariate Logistic regression analyses were performed to identify independent prognostic factors of pCR in the development cohort,and the nomogram was finally created.Nomogram performance was assessed with the area under the receiver operating characteristic curve and calibration slope.Results Tumor size[odd ratio(OR)=0.96,95%CI 0.93-0.99,P=0.01],human epidermal growth factor receptor(HER-2)status(OR=0.07,95%CI 0.02-0.22,P<0.001),Ki-67 index(OR=1.04,95%CI 1.01-1.07,P=0.002)and signal enhancement ratio(OR=1.02,95%CI 1.01-1.03,P=0.002)were independently associated with pCR.The nomogram incorporating these variables showed good discrimination(area under the curve was 0.82,95%CI 0.75-0.89)and calibration abilities(calibration slope=0.89,95%CI 0.83-0.94)in the development cohort.Conclusion A nomogram incorporating tumor size,HER-2 state,Ki-67 index and signal enhancement ratio show good discrimination and calibration abilities in predicting pCR.