动态增强MR定量分析早期预测乳腺癌新辅助化疗疗效的研究

Quantitative dynamic contrast enhanced MR in the prediction of response in breast cancer patients undergoing neoadjuvant chemotherapy

目的探讨动态增强MR定量分析早期预测乳腺癌新辅助化疗（NAC）疗效的可行性。方法收集分析47例经核芯针穿刺病理证实的，乳腺浸润性导管癌患者化疗前和NAC2周期后的动态增强MR量化参数：容量转移常数（Ktrans）、速率常数（Kep）和细胞外血管外间隙容积比（Vc）。按病理结果分为组织学非显著反应（NMHR）组和组织学显著反应（MHR）组。采用非参数检验（Mann—WhitneyU检验）比较MHR组和NMHR组NAC2周期后各量化参数的变化值，采用两独立样本t检验比较化疗前两组的基线参数值，并通过绘制受试者工作曲线（ROC）找到最佳预测参数及其诊断阈值。结果MHR组15例，NMHR组32例。化疗前MHR组的Ktrans、Kep和Vc值分别为（1．51±0．33）／min、（2．97±1．06）／min和（0．55±0．16），NMHR组各参数值分别为（1．53±0．40）／min、（2．82±0．99）／min和（0．57±0．20），两组间差异无统计学意义（t值分别为-0．123、0．450和-0．380，P值均〉0．05）。NAC2周期后MHR组Ktrans、Kep和Vc的变化值分别为-88．2％（-96．0％-1．5％）、-62．5％（-94．3％～-8．7％）和-57．0％（-82．1％～55．5％），NMHR组各参数的变化值分别为-8．1％（-88．5％-32．4％）、-18．2％（-62．1％～145．9％）和-4．7％（-83．1％～95．7％），两组间差异均有统计学意义（Z值分别为4．359、4．359和3．332，P值均〈0．01）。通过绘制ROC曲线可得，Ktrans和Kep的变化值及NAC2周期后Ktrans值的曲线下面积最大，均为0．898，三者预测NMHR的敏感度分别为87．5％、90．6％和78．1％，特异度分别为86．7％、80．0％和93．3％。结论动态增强MR定量分析可早期（化疗2周期）预测乳腺癌患者NAC的最终疗效。

Objective To investigate whether quantitative dynamic contrast enhanced MR can predict final pathologic response in primary breast cancer patients undergoing neoadjuvant chemotherapy （NAC）. Methods Forty seven patients who were pathologically proved infiltrating ductal carcinoma with core needle puncture biopsy were examined before NAC and after 2 cycles of treatment and the quantitative parameters （Ktrans, Kep and Ve ） were analyzed prospectively. Histological response is categorized as non-major histological response （NMHR） and major histological response （MHR）. Quantitative parameter changes measured after 2 cycles of NAC were compared between MHR and NMHR using non parametric tests （Mann- Whitney U test ） and pretreatment parameters were compared using independent samples t tests. Receiver operating characteristic curve （ ROC ） was used to determine the best predictor and cutoff value. Results Fifteen patients were grouped into MHR and 32 patients were NMHR. Pretreatment parameters（ Ktrans, Kep and Vc） were（1.51 ±0.33） /min, （2.97 ±1.06） /min and （0.55 ±0.16） in MHR and （1.53 ± 0.40） /min,（2. 82 ±0. 99 ） /min and （0. 57±0. 20） in NMHR. There was no significant difference between the two groups （t values were -0. 123,0. 450 and -0. 380, respectively, P 〉 0. 05 ）. Changes inkinetic parameters（K , Kep and V, ） were - 88.2% （ - 96. 0% to 1.5% ） , - 62. 5% （ - 94. 3% to -8.7%） and -57.0%（ -82. 1% to 55.5%） in MHR and -8.1%（ -88.5% to 32. 4%）, -18.2% （ - 62. 1% to 145.9% ） and - 4. 7% （ - 83.1% to 95.7% ） in NMHR. There were significant difference between the two groups （Z values were 4. 359,4. 359 and 3. 332, respectively, P 〈 0. 01 ）. The areas under ROC curve of △Ktrans, △Kep and Ktrans after 2 cycles of NAC were all 0. 898. Sensitivity of the three parameters for predicting NMHR were 87. 5% ,90. 6% and 78.1% ,and specificity were 86. 7% ,80. 0% and 93.3%, respectively. Conclusion Quantitative dynamic contrast enhanced MRI can predict final pathologic response in primary breast cancers after 2 cycles of NAC.