Fang Huiying, Yue Jian, Li Hongzhong, Luan Tiankuo, Wang Pin, Ren Guosheng
Chongqing Key Laboratory of Molecular Oncology and Epigenetics, The First Affiliated Hospital of Chongqing Medical University, Chongqing, China.
Department of Breast Cancer Center, Chongqing University Cancer Hospital, Chongqing, China.
Front Oncol. 2024 Aug 23;14:1444531. doi: 10.3389/fonc.2024.1444531. eCollection 2024.
The study aimed to develop a nomogram model for individual prognosis prediction in patients with hormone receptors positive (HR+) mucinous breast carcinoma (MBC) and assess the value of neoadjuvant chemotherapy (NAC) in this context.
A total of 6,850 HR+ MBC patients from the SEER database were identified and randomly (in a 7:3 ratio) divided into training cohorts and internal validation cohorts. 77 patients were enrolled from the Chongqing University Cancer Hospital as the external validation cohort. Independent risk factors affecting overall survival (OS) were selected using univariate and multivariate Cox regression analysis, and nomogram models were constructed and validated. A propensity score matching (PSM) approach was used in the exploration of the value of NAC versus adjuvant chemocherapy (AC) for long-term prognosis in HR+ MBC patients.
Multivariate Cox regression analysis showed 8 independent prognostic factors: age, race, marital status, tumor size, distant metastasis, surgery, radiotherapy, and chemotherapy. The constructed nomogram model based on these 8 factors exhibited good consistency and accuracy. In the training group, internal validation group and external validation group, the high-risk groups demonstrated worse OS (<0.0001). Subgroup analysis revealed that NAC had no impact on OS ( = 0.18), or cancer specific survival (CSS) ( = 0.26) compared with AC after PSM.
The established nomogram model provides an accurate prognostic prediction for HR+ MBC patients. NAC does not confer long-term survival benefits compared to AC. These findings provide a novel approach for prognostic prediction and clinical practice.
本研究旨在开发一种列线图模型，用于预测激素受体阳性（HR+）黏液性乳腺癌（MBC）患者的个体预后，并评估新辅助化疗（NAC）在此背景下的价值。
从监测、流行病学和最终结果（SEER）数据库中识别出6850例HR+ MBC患者，并随机（按7:3的比例）分为训练队列和内部验证队列。从重庆大学附属肿瘤医院招募77例患者作为外部验证队列。采用单因素和多因素Cox回归分析选择影响总生存期（OS）的独立危险因素，并构建和验证列线图模型。在探索NAC与辅助化疗（AC）对HR+ MBC患者长期预后的价值时，采用倾向评分匹配（PSM）方法。
多因素Cox回归分析显示8个独立预后因素：年龄、种族、婚姻状况、肿瘤大小、远处转移、手术、放疗和化疗。基于这8个因素构建的列线图模型显示出良好的一致性和准确性。在训练组、内部验证组和外部验证组中，高危组的OS较差（<0.0001）。亚组分析显示，PSM后，与AC相比，NAC对OS（ = 0.18）或癌症特异性生存期（CSS）（ = 0.26）无影响。
所建立的列线图模型为HR+ MBC患者提供了准确的预后预测。与AC相比，NAC未带来长期生存益处。这些发现为预后预测和临床实践提供了一种新方法。
