浸润性乳腺癌新辅助化疗疗效预测指数

A efficacy predictive index for invasive breast cancer with neoadjuvant chemother

目的 探讨乳腺癌患者营养和免疫相关实验室指标(简称实验室指标)与新辅助化疗(neoadjuvant chemotherapy,NAC)后病理完全缓解(pathologic complete response,pCR)之间的关系,构建实验室指标组合,确立乳腺癌NAC后pCR的临床预测指标。方法 回顾性收集2020年9月1日至2022年10月31日在西南医科大学附属医院乳腺外科接受NAC的310例浸润性乳腺癌患者NAC前实验室指标包括[白蛋白(albumin,ALB)、总胆固醇、甘油三酯、高密度脂蛋白胆固醇(high density lipoprotein cholesterol,HDL-C)、低密度脂蛋白胆固醇、载脂蛋白A-Ⅰ、载脂蛋白B、白细胞、中性粒细胞、淋巴细胞、单核细胞(monocyte,MON)和血小板]以及临床病理资料。采用logistic回归分析方法分析实验室指标与NAC后pCR的关系,通过简单数学运算构建实验室指标组合,使用受试者工作特征曲线下面积(area under curve,AUC)评估实验室指标不同组合预测pCR的效能,确立最佳实验室指标组合(简称最佳组合)。采用多因素logistic回归分析方法分析乳腺癌患者临床病理学特征与NAC后pCR的相关性,确立NAC后pCR的独立预测指标。结果 310例患者中有153例(49.4%)患者NAC后达到pCR。Logistic回归分析结果显示,实验室指标中ALB(Z=5.203,P<0.001)和HDL-C (Z=2.129,P=0.033)与NAC后pCR呈正相关关系,MON(Z=–4.883,P<0.001)与NAC后pCR呈负相关关系。实验室指标6种不同组合的AUC分析结果显示,ALB/MON组合的预测性能最高(中位AUC=0.708),确立为最佳组合,并确定该指标为新辅助疗效预测指数(neoadjuvant therapy predictive index,NTPI)。多因素logistic回归分析结果显示:雌激素受体(Z=–3.273,P=0.001)、人表皮生长因子2(Z=7.041,P<0.001)、Ki-67(Z=2.457,P=0.014)和NTPI(Z=4.661,P<0.001)为NAC后pCR的独立预测因子。结论 NTPI可以作为乳腺癌NAC后pCR的预测指标。

Objective To analyze the association between nutritional and immune-related laboratory indices and pathologic complete response(pCR)after neoadjuvant chemotherapy(NAC)in breast cancer patients and focused on constructing a combination of laboratory indices to serve as a clinical predictor of pCR after NAC in breast cancer.Methods Retrospectively collected the pre-NAC laboratory indices[albumin(ALB),total cholesterol,triglyceride,high density lipoprotein cholesterol(HDL-C),low density lipoprotein cholesterol,apolipoprotein A-Ⅰ,apolipoprotein B,white blood cell,neutrophil,lymphocyte,monocyte(MON),and platelet]and clinicopathologic data of 310 patients with invasive breast cancer who had received NAC in the Department of Breast Surgery,Affiliated Hospital of Southwest Medical University,from September 1,2020 to October 31,2022.Logistic regression analysis was conducted to determine the correlation between laboratory indices and post-NAC pCR.The combinations of laboratory indices were constructed by simple mathematical operation.The area under the receiver operating characteristic curve(AUC)was used to evaluate the efficacy of different combinations of laboratory indices in predicting pCR and to determine the optimal combination of liboratory indices.Multivariate logistic regression analysis was used to analysis the relevance between clinicopathologic features and post-NAC pCR in breast cancer patients and to determine the independent predictor of post-NAC pCR.Results Among the 310 patients,49.4%(153/310)of them achieved pCR after NAC.Logistic regression analysis revealed that ALB (Z=5.203, P<0.001) and HDL-C (Z=2.129, P=0.033) were positively correlated with post-NAC pCR, while MON(Z=–4.883, P<0.001) was negatively correlated with post-NAC pCR. The AUC analysis of 6 different combinations oflaboratory indices showed that the ALB/MON combination (the optimal combination of liboratory indices) had thehighest predictive performance (median AUC=0.708) and was determined to be the neoadjuvant therapy predictive index(NTPI). Multivariate logistic regression analysis showed that estrogen receptor (Z=–3.273, P=0.001), human epidermalgrowth factor 2 (Z=7.041, P<0.001), Ki-67 (Z=2.457, P=0.014), and NTPI (Z=4.661, P<0.001) were the independentpredictors for post-NAC pCR. Conclusion NTPI could serve as a predictive index for post-NAC pCR in patients withbreast cancer.