Zhao Rong, Xing Jun, Gao Jinnan
Department of Breast Surgery, Shanxi Bethune Hospital, Shanxi Academy of Medical Sciences, Tongji Shanxi Hospital, Third Hospital of Shanxi Medical University, Taiyuan, China.
Front Oncol. 2022 May 12;12:875665. doi: 10.3389/fonc.2022.875665. eCollection 2022.
The chances of second surgery due to positive margins in patients receiving breast-conversing surgery (BCS) were about 20-40%. This study aims to develop and validate a nomogram to predict the status of breast-conserving margins.
The database identified patients with core needle biopsy-proven ductal carcinoma (DCIS) or invasive breast carcinoma who underwent BCS in Shanxi Bethune Hospital between January 1, 2015 and December 31, 2021 (n = 573). The patients were divided into two models: (1) The first model consists of 398 patients who underwent BCS between 2015 and 2019; (2) The validation model consists of 175 patients who underwent BCS between 2020 and 2021. The development of the nomogram was based on the findings of multivariate logistic regression analysis. Discrimination was assessed by computing the C-index. The Hosmer-Lemeshow goodness-of-fit test was used to validate the calibration performance.
The final multivariate regression model was developed as a nomogram, including blood flow signals (OR = 2.88, p = 0.001), grade (OR = 2.46, p = 0.002), microcalcifications (OR = 2.39, p = 0.003), tumor size in ultrasound (OR = 2.12, p = 0.011) and cerbB-2 status (OR = 1.99, p = 0.042). C-indices were calculated of 0.71 (95% CI: 0.64-0.78) and 0.68 (95% CI: 0.59-0.78) for the modeling and the validation group, respectively. The calibration of the model was considered adequate in the validation group (p > 0.05).
We developed a nomogram that enables the estimation of the preoperative risk of positive BCS margins. Our nomogram provides a valuable tool for identifying high-risk patients who might have to undergo a wider excision.
接受保乳手术（BCS）的患者因切缘阳性而需二次手术的几率约为20%-40%。本研究旨在开发并验证一种列线图，以预测保乳切缘状态。
数据库纳入了2015年1月1日至2021年12月31日期间在山西白求恩医院接受BCS且经粗针穿刺活检证实为导管原位癌（DCIS）或浸润性乳腺癌的患者（n = 573）。患者被分为两个模型：（1）第一个模型由2015年至2019年期间接受BCS的398例患者组成；（2）验证模型由2020年至2021年期间接受BCS的175例患者组成。列线图的开发基于多因素逻辑回归分析结果。通过计算C指数评估辨别力。采用Hosmer-Lemeshow拟合优度检验验证校准性能。
最终的多因素回归模型被开发为列线图，包括血流信号（OR = 2.88，p = 0.001）、分级（OR = 2.46，p = 0.002）、微钙化（OR = 2.39，p = 0.003）、超声肿瘤大小（OR = 2.12，p = 0.011）和cerbB-2状态（OR = 1.99，p = 0.042）。建模组和验证组的C指数分别计算为0.71（95%CI：0.64-0.78）和0.68（95%CI：0.59-0.78）。验证组中模型的校准被认为是充分的（p>0.05）。
我们开发了一种列线图，能够估计保乳手术切缘阳性的术前风险。我们的列线图为识别可能需要进行更广泛切除的高危患者提供了一个有价值的工具。
