目的：三阴性乳腺癌（Triple-negative breast cancer，TNBC）是一种高度侵袭性的乳腺癌亚型，预后较差。该类型乳腺癌缺乏雌激素受体、孕激素受体和HER2受体的表达，因此对常规的内分泌治疗和靶向治疗不敏感。新辅助化疗作为三阴性乳腺癌的标准治疗方法，能够有效缩小肿瘤体积，从而提高患者接受保乳手术的可能性。紫杉类药物是TNBC新辅助化疗的基础药物，其中白蛋白紫杉醇和紫杉醇脂质体是两种新型紫杉醇制剂，具有提高疗效、降低毒性的潜力。目前尚缺乏多中心、大样本的回顾性研究，直接对比白蛋白紫杉醇联合卡铂与紫杉醇脂质体联合卡铂在TNBC新辅助治疗中的疗效和安全性。本研究旨在通过回顾性分析三家医院TNBC患者的临床资料，比较两种治疗方案的有效性和安全性，为临床制定个体化治疗方案提供参考依据。

材料与方法：本研究收集了2017年1月至2023年12月期间在中国医学科学院肿瘤医院、中国人民解放军总医院和北京大学第一医院三家医院接受新辅助治疗的TNBC患者的临床资料。治疗方案包括：白蛋白紫杉醇联合卡铂、紫杉醇脂质体联合卡铂以及两种方案联合免疫治疗（如PD-1/PD-L1抑制剂）、靶向治疗（如PARP抑制剂）。主要疗效指标为病理完全缓解率（pCR）。次要疗效指标包括客观缓解率（ORR），生存情况包括无复发生存期（RFS）和总生存期（OS）。收集患者治疗过程中出现的不良事件（AE），并根据美国国家癌症研究所常见不良事件评价标准（NCI-CTCAE）5.0版进行分级。数据采用SPSS 26.0软件进行统计分析。

结果：共收集129例患者，白蛋白紫杉醇联合卡铂组82例，紫杉醇脂质体联合卡铂组47例。两组患者的基线特征均衡可比。白蛋白紫杉醇组病理完全缓解率（pCR）为56.1%，紫杉醇脂质体组为44.7%，差异无统计学意义（P=0.212）。tpCR相关单因素分析结果显示，组织学分级是其影响因素，组织学分级为III级较II级的患者更易达到pCR（风险比=0.278，95%CI：0.103-0.749，P=0.011）。白蛋白紫杉醇组客观缓解率（ORR）为91.5%，紫杉醇脂质体组为85.1%，差异亦无统计学意义（P=0.264）。白蛋白紫杉醇组更易发生手脚麻木、关节肌肉痛等外周神经毒性（P<0.05），紫杉醇脂质体组更易出现脱发及乏力（P<0.05）。中位随访时间37.8个月，两组患者的3年和5年无复发生存率（RFS）及总生存率（OS）均无统计学差异。Ki67指数≥50%（风险比=3.062，95%CI：1.181-7.943，P=0.021）对RFS有影响，使患者RFS更短。

结论：白蛋白紫杉醇病理完全缓解率（pCR）较紫杉醇脂质体呈现更高的趋势（56.1% vs 44.7%），但无统计学差异。组织学分级为III级较II级的患者更易达到pCR。白蛋白紫杉醇组更易出现外周神经毒性，紫杉醇脂质体组脱发及乏力的发生率更高。两组在RFS和OS上无统计学差异。Ki67≥50%使患者无复发生存期（RFS）更短。白蛋白紫杉醇可作为优选紫杉醇药物。

Objective: Triple-negative breast cancer (TNBC) is a highly aggressive subtype of breast cancer with a poor prognosis. This type of cancer lacks the expression of estrogen receptors, progesterone receptors, and HER2 receptors, making it insensitive to conventional endocrine therapy and targeted therapies. Neoadjuvant chemotherapy is the standard treatment for TNBC, effectively reducing tumor size and increasing the likelihood of breast-conserving surgery. Taxanes are the cornerstone of neoadjuvant chemotherapy for TNBC, with albumin-bound paclitaxel and liposomal paclitaxel being two novel formulations of paclitaxel that have the potential to enhance efficacy and reduce toxicity. Currently, there is a lack of large-scale, multi-center retrospective studies directly comparing the efficacy and safety of albumin-bound paclitaxel plus carboplatin versus liposomal paclitaxel plus carboplatin in neoadjuvant treatment for TNBC. This study aims to retrospectively analyze clinical data from three hospitals to compare the effectiveness and safety of the two treatment regimens, providing valuable insights for individualized treatment strategies in clinical practice.

Materials and Methods: This study retrospectively collected clinical data from TNBC patients who received neoadjuvant treatment between January 2017 and December 2023 at three institutions: the Cancer Hospital Chinese Academy of Medical Sciences, the Chinese PLA General Hospital (PLAGH), and Peking University First Hospital. The treatment regimens included albumin-bound paclitaxel combined with carboplatin, liposomal paclitaxel combined with carboplatin, and combinations of these regimens with immunotherapy (e.g., PD-1/PD-L1 inhibitors) or targeted therapy (e.g., PARP inhibitors). The primary efficacy endpoint was the pathological complete response (pCR) rate, while secondary endpoints included the objective response rate (ORR), as well as survival outcomes, such as relapse-free survival (RFS) and overall survival (OS). Adverse events (AEs) occurring during treatment were recorded and classified according to the National Cancer Institute's Common Terminology Criteria for Adverse Events (NCI-CTCAE) version 5.0. Statistical analyses were conducted using SPSS version 26.0.

Results: A total of 129 patients were enrolled, with 82 patients in the albumin-bound paclitaxel plus carboplatin group and 47 patients in the liposomal paclitaxel plus carboplatin group. Baseline characteristics were well balanced between the two groups. The pathological complete response (pCR) rate was 56.1% in the albumin-bound paclitaxel group and 44.7% in the liposomal paclitaxel group, with no statistically significant difference (P=0.212). Univariate analysis showed histological grade significantly influenced tpCR rates, with grade III tumors having higher pCR likelihood than grade II (HR=0.278, 95%CI 0.103-0.749, P=0.011). The objective response rate (ORR) was 91.5% in the albumin-bound paclitaxel group and 85.1% in the liposomal paclitaxel group, with no significant difference (P=0.264). Peripheral neurotoxicity, such as hand-foot numbness and joint-muscle pain, was more commonly observed in the albumin-bound paclitaxel group (P<0.05), whereas hair loss and fatigue were more prevalent in the liposomal paclitaxel group (P<0.05). The median follow-up time was 37.8 months, and there were no statistically significant differences in the 3-year and 5-year relapse-free survival (RFS) and overall survival (OS) between the two groups. The Ki67 index ≥ 50% (HR=3.062, 95%CI: 1.181-7.943,P=0.021) was associated with RFS, resulting in a shorter RFS in patients.

Conclusion: The pathological complete response rate (pCR) of albumin-bound paclitaxel showed a higher trend compared to liposomal paclitaxel (56.1% vs 44.7%), but the difference was not statistically significant. Patients with histological grade III were more likely to achieve pCR compared to those with grade II. The albumin-bound paclitaxel group had a higher incidence of peripheral neurotoxicity, while the liposomal paclitaxel group had higher rates of hair loss and fatigue. There was no statistical difference between the two groups in recurrence-free survival (RFS) and overall survival (OS). Ki67≥50% was associated with shorter RFS. Albumin-bound paclitaxel can be considered as a preferred formulation of paclitaxel.