【作者】 许梅；

【导师】 马捷；

【作者基本信息】 暨南大学 ， 影像医学与核医学 放射学（专业学位）， 2020， 硕士

【摘要】 目的分析新辅助化疗(NAC)后磁共振(MRI)上乳腺原发肿瘤的不同退缩模式,探讨不同退缩模式的相关因素,不同分子分型乳腺癌的相关影响因素,以及MRI评价肿瘤化疗反应性的多因素分析,拟提出MRI对新辅助化疗后肿瘤治疗方式的研究价值,为乳腺癌患者个体化治疗及预后评估提供影像学依据。资料与方法回顾性分析本院2016年1月至2019年12月NAC前经核心针穿刺活检及NAC后手术病理证实为单侧乳腺癌患者,且NAC前及NAC后手术前2个星期均进行了乳腺MRI检查的58名女性患者。所有患者均行双侧乳腺平扫及增强MRI扫描,且NAC前均行乳腺X线摄影检查。2名医生分别阅片,分析肿瘤退缩模式,肿瘤退缩模式分为3型:向心性退缩;碎片样退缩;混合型退缩。根据NAC后手术病理检查结果,将患者分为病理完全缓解(pCR)组和非pCR组。记录NAC前经穿刺检测肿瘤内雌激素受体(ER),孕激素受体(PR)和人体表皮生长因子受体(HER2)和细胞增殖核抗原Ki-67的表达情况。采用Kappa检验评价2名医师对退缩模式分型的一致性,采用χ2检验、Fisher精确检验及Wilcoxon秩和检验比较pCR组及非pCR组临床、影像及病理资料的差异,采用χ2检验、Fisher精确检验及Wilcoxon秩和检验比较NAC后不同肿瘤退缩模式的临床、病理及影像学指标的差异性,采用Fisher精确检验比较4种分子分型的临床、影像资料是否存在差异,并且采用Binary logistic回归分析疗效的相关因素。并采用四格表的方法评价磁共振评估NAC后肿瘤达pCR的的诊断效能。结果NAC后经手术病理证实,pCR组22例,非pCR组36例。2名医师对化疗后肿瘤退缩模式定性评估的Kappa值分别为0.766,一致性为好。pCR和非pCR组间,年龄、绝经状态、T分期,治疗前淋巴结是否转移,分子分型,ER,PR、NAC前BPE情况,病灶最大径、MRI上肿瘤强化方式、是否单一病灶、治疗前病灶是否含有钙化,差异均无统计学异常(p均>0.05);退缩模式、HER2状态及Ki-67状态差异有统计学意义(p均<0.05)。NAC后肿瘤向心性退缩36例,碎片性退缩7例,混合型退缩15例,三组间的年龄、绝经状态、T分期,是否淋巴结转移、分子分型、病灶化疗前是否含钙化情况,肿瘤大小、BPE、Ki-67、ER、PR、HER2表达情况及是否单一病灶的差异均无统计学意义(p均>0.05),而MRI上强化方式在三组间的差异有统计学意义(p<0.05)。不同分子分型乳腺癌的年龄、绝经状态、T分期、淋巴结情况、肿瘤MRI强化方式、是否单一病灶及病灶化疗前是否含钙化情况差异均无统计学意义(p均>0.05)。多因素Logistic回归分析,结果表明NAC后肿瘤退缩模式及HER2表达状态是新辅助化疗后达到pCR的独立预测因素。MRI评价乳腺NAC后pCR的准确性82.8%,灵敏度59.1%,特异度97.2%。结论(1)乳腺癌新辅助化疗原发肿瘤在MRI上强化方式与后肿瘤退缩模式有相关性,可在一定程度预测NAC后肿瘤退缩模式,进而预测患者保乳手术的可能性。(2)肿瘤的退缩模式与乳腺癌分子分型无关。(3)乳腺癌HER2状态及NAC后肿瘤退缩模式可作为NAC疗效评估的独立因素。(4)NAC前HER2受体阳性患者,NAC后更容易达到pCR;pCR患者中更大概率为向心性退缩,碎片样退缩及混合型退缩难以达到pCR。(5)MRI评价乳腺癌NAC后达pCR的准确性82.8%,灵敏度59.1%,特异度97.2%,对残余病灶存在一定高估及低估。更多还原

【Abstract】 Objective:To analyze the different shrinkage patterns of breast primary tumor in magnetic resonance imaging(MRI)after neoadjuvant chemotherapy(NAC),explore the influencing factors of the shrinkage patterns,the influencing factors of the molecular types of breast cancer,as well as the value of MRI in tumor treatment after neoadjuvant chemotherapy was proposed and to provide imaging basis for individualized treatment and prognosis assessment of breast cancer.Material and methods:In all,the datas of 58 patients with histologically confirmed unilateral breast cancer between January 2016 and december 2019 were collected.All of these patients were examined with magnetic resonance imaging(MRI)before and after NAC,and underwent mammography before NAC,and underwent surgery within two weeks after NAC.Two experienced radiologists independently reviewed the imaging and analyzed the tumor shrinkage patterns.The shrinkage pattern was classified into 3 categories: concentric shrinkage;nodular shrinkage;mixed shrinkage.All the patients were divided into pathological complete response(pCR)group and non-pCR group according to the histopathologic tumor response.The status of estrogen receptor(ER),progesterone receptor(PR),human epidermal growth factor receptor(HER2)and cell proliferating cell nuclear antigen Ki-67 were recorded before underwent NAC.The kappa test was used to test the inter-observer agreement of the tumor shrinkage pattern.The χ2 test,Fisher’s exact test and Wilcoxon test were used to compare the baseline characteristics of pCR and Non-pCR groups.The χ2 test and Fisher’s exact test were used to compare the differences of clinical,pathological and imaging indexes of different tumor shrinkage patterns after NAC.Fisher’s exact test was used to compare the clinical and imaging data of the molecular types.Binary logistic regression was used to analyze the correlation factors of tumor response.The diagnostic efficacy of pCR after NAC was evaluated by four-grid table.Results:Twenty-two patients achieved pCR,and 36 patients had residual disease(non-pCR). Inter-observer agreement regarding shrinkage pattern was good after NAC(Kappa value 0.766).The age,menopausal status,T stage,lymph node status,subtypes,ER,PR,NAC pre-BPE,tumor size,tumor enhancement pattern on MRI,single lesion or not,and calcifications showed no significance between pCR and non-pCR groups(allp>0.05),while only HER2、Ki-67 status and the shrinkage pattern a significant difference(all p<0.05).A total of 36 lesions show concentric shrinkage,7 nodular shrinkage,and 15 mixed shrinkage.And the age,menopausal status,T stage,lymph node status,molecular subtypes,calcification,tumor size,BPE,Ki-67,ER,PR,and HER2 status and whether a single lesion showed no significance between the shrinkage patterns(all p>0.05),the difference of MRI enhancement of the tumor was statistically significant(p<0.05).There were no statistically significant differences in age,menopausal status,T stage,lymph node status,tumor MRI enhancement,single lesion or not and calcifications among different subtypes of breast cancer(p<0.05). Logistic model showed that tumor shrinkage pattern after NAC and HER2 status were independent factors for predicting pCR.The accuracy,sensitivity and specificity of NAC pCR were evaluated by MRI with 82.8%,59.1% and 97.2%,respectively.Conclusions:(1)The tumor shrinkage pattern after neoadjuvant chemotherapy of breast cancer was correlated with,which can predict the tumor shrinkage pattern after NAC to a certain extent,and then predict the possibility of breast conserving surgery in patients.(2)The pattern of tumor shrinkage was independent of the molecular biology of breast cancer.(3)The shrinkage pattern and HER2 status can be independent factors for predicting pCR.(4)The HER2 receptor positive was more likely to achieve pCR,while the nodular and mixed shrinkage were difficult to achieve pCR,and most of the pCR patients were concentric shrinkage.(5)The accuracy,sensitivity and specificity of pCR were evaluated by MRI with82.8%,59.1% and 97.2% respectively,with certain overestimation and underestimation of residual lesions.更多还原