Gao Ying, Liu Lei, Wang Shoujun, Tao Weijie, Wang Jinmiao, Duan Ran, Xie Hai, Takahashi Hideaki, Hao Jie, Gao Ming
Department of Breast and Thyroid Surgery, Tianjin Key Laboratory of General Surgery in Construction, Tianjin Union Medical Center, The First Affiliated Hospital of Nankai University, Tianjin, China.
Department of Thyroid and Neck Cancer, Tianjin Medical University Cancer Institute and Hospital, National Cancer Clinical Research Center, Tianjin Cancer Clinical Research Center, Tianjin Key Laboratory of Cancer Prevention and Treatment, Tianjin, China.
Gland Surg. 2025 Jul 31;14(7):1366-1378. doi: 10.21037/gs-2025-168. Epub 2025 Jul 28.
Breast cancer (BC) is the leading cancer in women. It often metastasizes to bone, worsening the prognosis. Diagnostic methods often fail to predict bone metastasis (BM). This study developed a machine learning (ML) model using the Surveillance, Epidemiology, and End Results (SEER) database for BM prediction, to refine treatments and improve outcomes.
Using SEER data, we studied 24,584 BC patients diagnosed 2010-2015 with radiologically confirmed BM. Tumor size, grade, tumor (T)/node (N) stages, and estrogen receptor (ER)/progesterone receptor (PR)/human epidermal growth factor receptor 2 (HER2) status were assessed. Stratified randomization divided the data into 70% training (n=18,438) and 30% validation (n=6,146). Six ML algorithms were developed, emphasizing random forest (RF). Receiver operating characteristic (ROC) curve analysis [area under the curve (AUC), sensitivity, specificity, negative predictive value (NPV)] assessed performance. The SHapley Additive exPlanations (SHAP) framework identified key BM predictors.
Our analysis of 24,584 patients identified 1,298 (5.26%) patients with BM. Logistic regression (LR) provided the highest specificity [0.897, 95% confidence interval (CI): 0.889-0.905], contrasting with gradient boosting machine (GBM)'s maximal sensitivity (0.658, 95% CI: 0.609-0.707). With sensitivity at 0.658, better algorithms or multimodal methods are needed for case identification. The multilayer perceptron neural network (MLPNN) model demonstrated superior performance, with the highest AUC of 0.808 (95% CI: 0.798-0.818), surpassing the LR and adaptive boosting (AdaBoost) models, both with AUCs of 0.803 (95% CI: 0.793-0.813). The RF model was particularly adept at ruling out BM, with an NPV above 97%. The SHAP analysis identified tumor size, grade, T/N stages, ER/PR/HER2 status, and brain/liver/lung metastases as key predictors for risk stratification. Decision curve analysis showed RF's superior utility over the American Joint Committee on Cancer (AJCC) Staging System.
Our ML model demonstrates potential for predicting BM in patients with BC. It may serve as a clinical aid to identify at-risk individuals early. However, moderate sensitivity requires refinement for better case detection. This study supports integrating ML into clinical practice, advancing personalized oncology medicine.
乳腺癌（BC）是女性中最常见的癌症。它常转移至骨骼，使预后恶化。诊断方法往往无法预测骨转移（BM）。本研究利用监测、流行病学和最终结果（SEER）数据库开发了一种机器学习（ML）模型用于BM预测，以优化治疗并改善预后。
利用SEER数据，我们研究了2010 - 2015年诊断为经放射学证实的BM的24584例BC患者。评估了肿瘤大小、分级、肿瘤（T）/淋巴结（N）分期以及雌激素受体（ER）/孕激素受体（PR）/人表皮生长因子受体2（HER2）状态。分层随机化将数据分为70%训练集（n = 18438）和30%验证集（n = 6146）。开发了六种ML算法，重点是随机森林（RF）。通过受试者工作特征（ROC）曲线分析[曲线下面积（AUC）、敏感性、特异性、阴性预测值（NPV）]评估性能。SHapley加性解释（SHAP）框架确定了关键的BM预测因子。
我们对24584例患者的分析确定了1298例（5.26%）有BM的患者。逻辑回归（LR）具有最高的特异性[0.897，95%置信区间（CI）：0.889 - 0.905]，与梯度提升机（GBM）的最大敏感性（0.658，95%CI：0.609 - 0.707）形成对比。在敏感性为0.658的情况下，需要更好的算法或多模态方法来进行病例识别。多层感知器神经网络（MLPNN）模型表现出卓越的性能，最高AUC为0.808（95%CI：0.798 - 0.818），超过了LR和自适应增强（AdaBoost）模型，这两种模型的AUC均为0.803（95%CI：|0.793 - 0.813）。RF模型特别擅长排除BM，NPV高于97%。SHAP分析确定肿瘤大小、分级、T/N分期、ER/PR/HER2状态以及脑/肝/肺转移是风险分层的关键预测因子。决策曲线分析显示RF比美国癌症联合委员会（AJCC）分期系统具有更高的效用。
我们的ML模型显示了预测BC患者BM的潜力。它可作为一种临床辅助手段，早期识别高危个体。然而，中等的敏感性需要改进以实现更好的病例检测。本研究支持将ML整合到临床实践中，推动个性化肿瘤医学发展。
