基于机器学习构建乳腺癌骨转移预测模型

欧阳飞; 王阳; 陈瑜; 裴国清; 王陵; 张扬; 石磊

doi:10.19401/j.cnki.1007-3639.2024.10.001

您当前的位置：

首页 >

文章列表页 >

基于机器学习构建乳腺癌骨转移预测模型

论著 | 更新时间：2025-12-31

- 基于机器学习构建乳腺癌骨转移预测模型
- Construction of the prediction model of breast cancer bone metastasis based on machine learning
- 中国癌症杂志 2024年34卷第10期页码：903-914
- 作者机构：
  
  1. 空军军医大学第一附属医院骨科，陕西西安，710032
  2. 空军军医大学第一附属医院神经内科，陕西西安，710032
  3. 空军军医大学军事预防医学系卫生统计学教研室，陕西西安，710032
- 作者简介：
  
  [ "第一作者：欧阳飞（ORCID: 0009-0007-6226-1440），在读硕士研究生，住院医师。" ]
  石磊（ORCID: 0000-0002-3396-4645），博士，副主任医师。
- 基金信息：
- DOI：10.19401/j.cnki.1007-3639.2024.10.001
  中图分类号： R737.9
- 收稿：2024-06-13，
  
  修回：2024-09-05，
  
  纸质出版：2024-10-30
- 稿件说明：
移动端阅览
欧阳飞, 王阳, 陈瑜, 等. 基于机器学习构建乳腺癌骨转移预测模型[J]. 中国癌症杂志, 2024,34(10):903-914.

Fei OUYANG, Yang WANG, Yu CHEN, et al. Construction of the prediction model of breast cancer bone metastasis based on machine learning[J]. China Oncology, 2024, 34(10): 903-914.
欧阳飞, 王阳, 陈瑜, 等. 基于机器学习构建乳腺癌骨转移预测模型[J]. 中国癌症杂志, 2024,34(10):903-914. DOI： 10.19401/j.cnki.1007-3639.2024.10.001.

Fei OUYANG, Yang WANG, Yu CHEN, et al. Construction of the prediction model of breast cancer bone metastasis based on machine learning[J]. China Oncology, 2024, 34(10): 903-914. DOI： 10.19401/j.cnki.1007-3639.2024.10.001.

摘要

背景与目的：

乳腺癌是全球重大公共卫生问题，骨是乳腺癌远处转移最常见的部位，约占所有转移病例的70%。乳腺癌骨转移可引起一系列并发症，包括剧烈疼痛、病理性骨折、高钙血症、脊髓压迫等，给患者身体活动带来极大不便，影响生活质量。转移性复发是乳腺癌患者死亡的主要原因。因此迫切需要构建乳腺癌骨转移预测模型，以识别具有高骨转移风险的患者。本研究旨在开发基于机器学习的预测模型来预测乳腺癌发生骨转移的概率。

方法：

从监测、流行病学和最终结果（The Surveillance

Epidemiology

and End Results，SEER）数据库中提取2010年—2015年诊断的乳腺癌患者数据，并通过最小绝对收敛和选择算子（least absolute shrinkage and selection operator，LASSO）回归、单因素和多因素logistic回归分析对变量进行筛选，纳入具有统计学意义的风险因素构建预测模型。本研究使用决策树、弹性网络、K最近邻、轻量级梯度提升机、logistic回归、神经网络、随机森林、支持向量机和极限梯度提升等9种机器学习算法，通过随机搜索和5倍交叉验证调整模型超参数，构建乳腺癌骨转移预测模型。利用受试者工作特征曲线（receiver operating characteristic，ROC）的曲线下面积（area under curve，AUC）、校准曲线和决策曲线对模型进行评价，得到最优模型，并基于最优模型分析变量的重要性。最后，应用最优模型建立预测乳腺癌骨转移风险的网络计算器。本队列研究严格遵循《加强流行病学中观察性研究报告质量》（Strengthening the Reporting of Observational Studies in Epidemiology，STROBE）指南中的各项条目。

结果：

本研究纳入10 106例乳腺癌患者，训练集7 073例患者，验证集3 033例患者，在这两个队列中，分别有4 494例（63.5%）和1 927例（63.5%）患者发生骨转移。种族、病理学分级、雌激素受体（estrogen receptor，ER）状态、孕激素受体（progesterone receptor，PR）状态、人表皮生长因子受体2（human epidermal growth factor receptor

2，HER2）状态、N分期、肺转移、放疗、化疗、手术是骨转移的独立预测因素。使用训练集和验证集对模型进行验证，综合ROC曲线的AUC、校准曲线和决策曲线等评价指标发现极限梯度提升算法优于其他机器学习算法。最后，本研究利用极限梯度提升算法构建预测乳腺癌骨转移的网络计算器，链接为https://bcbm.shinyapps.io/DynNomapp/。

结论：

本研究开发基于机器学习的预测模型，用于预测乳腺癌患者发生骨转移的概率，希望有助于临床医师作出更合理的治疗决策。

Abstract

Background and purpose:

Breast cancer is a major global public health problem. Bone is the most common site of distant metastasis of breast cancer

accounting for about 70% of all metastatic cases. Bone metastasis of breast cancer can cause a series of complications

including severe pain

pathological fracture

hypercalcemia

spinal cord compression

etc.

which bring great inconvenience to patients' physical activities and affect their quality of life. Metastatic recurrence is the leading cause of death in breast cancer patients. Therefore

there is an urgent need to build a diagnostic model of bone metastasis in breast cancer to identify patients with a high risk of bone metastasis. The aim of this study was to develop a predictive model based on machine learning to predict the probability of breast cancer developing bone metastasis.

Methods:

Data of breast cancer patients diagnosed between 2010 and 2015 were extracted from The Surveillance

Epidemiology

and End Results (SEER) database. The variables were screened by least absolute shrinkage and selection operator (LASSO) regression

univariate and multivariate logistic regression analysis

and statistically significant risk factors were included to build a prediction model. In this study

nine machine learning algorithms

including decision tree

elastic network

K-nearest neighbor

lightweight gradient elevator

logistic regression

neural network

random forest

support vector machine and limit gradient lifting

were used to adjust the model hyperparameters through random search and 5x cross-validation to build a breast cancer bone metastasis prediction model. The area under the receiver operating cha

racteristic (ROC) curve

calibration curve and decision curve were used to evaluate the model

the optimal model was obtained

and the importance of variables was analyzed based on the optimal model. Finally

a network calculator for predicting the risk of bone metastasis of breast cancer was established using the optimal model.

Results:

The study included 10 106 patients with breast cancer

7 073 patients in the training set

and 3 033 patients in the validation set. We found that 4 494 (63.5%) patients in the training set and 1 927 (63.5%) patients in the validation set developed bone metastases

respectively. Race

pathologic grade

estrogen receptor (ER) status

progesterone receptor (PR) status

human epidermal growth factor receptor 2 (HER2) status

N stage

lung metastasis

radiotherapy

chemotherapy and surgery were independent predictors of bone metastasis. The training set and verification set were used to verify the model

and the limit gradient lifting algorithm was superior to other machine learning algorithms by integrating the evaluation indexes such as the area under the ROC curve

calibration curve and decision curve. Finally

we used limit gradient algorithm to build network calculator for prediction of breast cancer bone metastases (https://bcbm.shinyapps.io/DynNomapp/).

Conclusion:

This study developed a predictive model based on machine learning to predict the probability of bone metastases in breast cancer patients

hoping to help clinicians make more rational treatment decisions.

关键词

Keywords

references

SUNG H , FERLAY J , SIEGEL R L , et al. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries [J ] . CA Cancer J Clin , 2021 , 71 ( 3 ): 209 - 249 .

MEDEIROS B , ALLAN A L . Molecular mechanisms of breast cancer metastasis to the lung: clinical and experimental perspectives [J ] . Int J Mol Sci , 2019 , 20 ( 9 ): 2272.

LIANG Y R , ZHANG H W , SONG X J , et al. Metastatic heterogeneity of breast cancer: molecular mechanism and potential therapeutic targets [J ] . Semin Cancer Biol , 2020 , 60 : 14 - 27 . DOI: S1044-579X(19)30063-X http://doi.org/S1044-579X(19)30063-X

JIANG Z C , LI J Y , CHEN S J , et al. Zoledronate and SPIO dual-targeting nanoparticles loaded with ICG for photothermal therapy of breast cancer tibial metastasis [J ] . Sci Rep , 2020 , 10 ( 1 ): 13675.

ZAJKOWSKA M , LUBOWICKA E , FIEDOROWICZ W , et al. Human plasma levels of VEGF-A, VEGF-C, VEGF-D, their soluble receptor-VEGFR-2 and applicability of these parameters as tumor markers in the diagnostics of breast cancer [J ] . Pathol Oncol Res , 2019 , 25 ( 4 ): 1477 - 1486 .

SIDEY-GIBBONS J A M , SIDEY-GIBBONS C J . Machine learning in medicine: a practical introduction [J ] . BMC Med Res Methodol , 2019 , 19 ( 1 ): 64.

ESTEVA A , KUPREL B , NOVOA R A , et al. Dermatologist-level classification of skin cancer with deep neural networks [J ] . Nature , 2017 , 542 ( 7639 ): 115 - 118 .

ANDERSON J P , PARIKH J R , SHENFELD D K , et al. Reverse engineering and evaluation of prediction models for progression to type 2 diabetes: an application of machine learning using electronic health records [J ] . J Diabetes Sci Technol , 2015 , 10 ( 1 ): 6 - 18 . DOI: 10.1177/1932296815620200 http://doi.org/10.1177/1932296815620200

RAHIMIAN F , SALIMI-KHORSHIDI G , PAYBERAH A H , et al. Predicting the risk of emergency admission with machine learning: development and validation using linked electronic health records [J ] . PLoS Med , 2018 , 15 ( 11 ): e1002695.

GUO Q P , WANG Y Q , AN J , et al. A prognostic model for patients with gastric signet ring cell carcinoma [J ] . Technol Cancer Res Treat , 2021 , 20 : 15330338211027912 .

YANG Y P , MA Y X , SHENG J , et al. A multicenter, retrospective epidemiologic survey of the clinical features and management of bone metastatic disease in China [J ] . Chin J Cancer , 2016 , 35 : 40 . DOI: 10.1186/s40880-016-0102-6 http://doi.org/10.1186/s40880-016-0102-6

PLUNKETT T A , SMITH P , RUBENS R D . Risk of complications from bone metastases in breast cancer. implications for management [J ] . Eur J Cancer , 2000 , 36 ( 4 ): 476 - 482 . DOI: 10.1016/s0959-8049(99)00331-7 http://doi.org/10.1016/s0959-8049(99)00331-7

PAREEK A , SINGH O P , YOGI V , et al. Bone metastases incidence and its correlation with hormonal and human epidermal growth factor receptor 2 neu receptors in breast cancer [J ] . J Cancer Res Ther , 2019 , 15 ( 5 ): 971 - 975 . DOI: 10.4103/jcrt.JCRT_235_18 http://doi.org/10.4103/jcrt.JCRT_235_18

Recommended breast cancer surveillance guidelines . American Society of Clinical Oncology [J ] . J Clin Oncol , 1997 , 15 ( 5 ): 2149 - 2156 .

ROSSELLI DEL TURCO M , PALLI D , CARIDDI A , et al. Intensive diagnostic follow-up after treatment of primary breast cancer. A randomized trial. National research council project on breast cancer follow-up [J ] . JAMA , 1994 , 271 ( 20 ): 1593 - 1597 .

Impact of follow-up testing on survival and health-related quality of life in breast cancer patients . A multicenter randomized controlled trial. The GIVIO investigators [J ] . JAMA , 1994 , 271 ( 20 ): 1587 - 1592 .

MO X L , CHEN X J , IEONG C , et al. Early prediction of clinical response to etanercept treatment in juvenile idiopathic arthritis using machine learning [J ] . Front Pharmacol , 2020 , 11 : 1164 . DOI: 10.3389/fphar.2020.01164 http://doi.org/10.3389/fphar.2020.01164

ZHU J , ZHENG J X , LI L F , et al. Application of machine learning algorithms to predict central lymph node metastasis in T1-T2, non-invasive, and clinically node negative papillary thyroid carcinoma [J ] . Front Med , 2021 , 8 : 635771 .

QIU B X , SHEN Z X , WU S , et al. A machine learning-based model for predicting distant metastasis in patients with rectal cancer [J ] . Front Oncol , 2023 , 13 : 1235121 .

FENG X W , HONG T , LIU W C , et al. Development and validation of a machine learning model to predict the risk of lymph node metastasis in renal carcinoma [J ] . Front Endocrinol , 2022 , 13 : 1054358 .

XI N M , WANG L , YANG C J . Improving the diagnosis of thyroid cancer by machine learning and clinical data [J ] . Sci Rep , 2022 , 12 ( 1 ): 11143.

SAKHUJA S , DEVEAUX A , WILSON L E , et al. Patterns of de-novo metastasis and breast cancer-specific mortality by race and molecular subtype in the SEER population-based dataset [J ] . Breast Cancer Res Treat , 2021 , 186 ( 2 ): 509 - 518 .

GAO T Y , SHAO F . Risk factors and prognostic factors for inflammatory breast cancer with bone metastasis: a population-based study [J ] . J Orthop Surg (Hong Kong) , 2021 , 29 ( 2 ): 23094990211000144.

AKINYEMIJU T , SAKHUJA S , WATERBOR J , et al. Racial/ethnic disparities in de novo metastases sites and survival outcomes for patients with primary breast, colorectal, and prostate cancer [J ] . Cancer Med , 2018 , 7 ( 4 ): 1183 - 1193 .

CHEN J , ZHU S , XIE X Z , et al. Analysis of clinicopathological factors associated with bone metastasis in breast cancer [J ] . J Huazhong Univ Sci Technol (Med Sci) , 2013 , 33 ( 1 ): 122 - 125 .

LIEDE A , JERZAK K J , HERNANDEZ R K , et al. The incidence of bone metastasis after early-stage breast cancer in Canada [J ] . Breast Cancer Res Treat , 2016 , 156 ( 3 ): 587 - 595 .

GAO C W , WANG J G , HE P S , et al. Metastatic pattern of breast cancer by histologic grade: a SEER population-based study [J ] . Discov Med , 2022 , 34 ( 173 ): 189 - 197 .

JAMES J J , EVANS A J , PINDER S E , et al. Bone metastases from breast carcinoma: histopathological-radiological correlations and prognostic features [J ] . Br J Cancer , 2003 , 89 ( 4 ): 660 - 665 .

ARPINO G , WEISS H , LEE A V , et al. Estrogen receptor-positive, progesterone receptor-negative breast cancer: association with growth factor receptor expression and tamoxifen resistance [J ] . J Natl Cancer Inst , 2005 , 97 ( 17 ): 1254 - 1261 . DOI: 10.1093/jnci/dji249 http://doi.org/10.1093/jnci/dji249

ARCIERO C A , GUO Y , JIANG R J , et al. ER + /HER2 + breast cancer has different metastatic patterns and better survival than ER - /HER2 + breast cancer [J ] . Clin Breast Cancer , 2019 , 19 ( 4 ): 236 - 245 .

HAYASHI N , IWAMOTO T , QI Y , et al. Bone metastasis-related signaling pathways in breast cancers stratified by estrogen receptor status [J ] . J Cancer , 2017 , 8 ( 6 ): 1045 - 1052 . DOI: 10.7150/jca.13690 http://doi.org/10.7150/jca.13690

LOI S , HAIBE-KAINS B , DESMEDT C , et al. Definition of clinically distinct molecular subtypes in estrogen receptor-positive breast carcinomas through genomic grade [J ] . J Clin Oncol , 2007 , 25 ( 10 ): 1239 - 1246 . DOI: 10.1200/JCO.2006.07.1522 http://doi.org/10.1200/JCO.2006.07.1522

KOIZUMI M , YOSHIMOTO M , KASUMI F , et al. An open cohort study of bone metastasis incidence following surgery in breast cancer patients [J ] . BMC Cancer , 2010 , 10 : 381 . DOI: 10.1186/1471-2407-10-381 http://doi.org/10.1186/1471-2407-10-381

TAYYEB B , PARVIN M . Pathogenesis of breast cancer metastasis to brain: a comprehensive approach to the signaling network [J ] . Mol Neurobiol , 2016 , 53 ( 1 ): 446 - 454 . DOI: 10.1007/s12035-014-9023-z http://doi.org/10.1007/s12035-014-9023-z

LU X , KANG Y B . Organotropism of breast cancer metastasis [J ] . J Mammary Gland Biol Neoplasia , 2007 , 12 ( 2/3 ): 153 - 162 .

YATES L R , KNAPPSKOG S , WEDGE D , et al. Genomic evolution of breast cancer metastasis and relapse [J ] . Cancer Cell , 2017 , 32 ( 2 ): 169 - 184 .e7. DOI: S1535-6108(17)30297-0 http://doi.org/S1535-6108(17)30297-0

中国抗癌协会乳腺癌专业委员会, 中华医学会肿瘤学分会乳腺肿瘤学组 . 中国抗癌协会乳腺癌诊治指南与规范(2024年版) [J ] . 中国癌症杂志 , 2023 , 33 ( 12 ): 1092 - 1187 . DOI: 10.19401/j.cnki.1007-3639.2023.12.004 http://doi.org/10.19401/j.cnki.1007-3639.2023.12.004

The Society of Breast Cancer China Anti-Cancer Association, Breast Oncology Group of the Oncology Branch of the Chinese Medical Association . Guidelines for breast cancer diagnosis and treatment by China Anti-cancer Association (2024 edition) [J ] . China Oncol , 2023 , 33 ( 12 ): 1092 - 1187 .

TU Q H , HU C , ZHANG H , et al. Establishment and validation of novel clinical prognosis nomograms for luminal A breast cancer patients with bone metastasis [J ] . Biomed Res Int , 2020 , 2020 : 1972064 .

GAO B , OU X L , LI M F , et al. Risk stratification system and visualized dynamic nomogram constructed for predicting diagnosis and prognosis in rare male breast cancer patients with bone metastases [J ] . Front Endocrinol , 2022 , 13 : 1013338 .

PURUSHOTHAM A , SHAMIL E , CARIATI M , et al. Age at diagnosis and distant metastasis in breast cancer: a surprising inverse relationship [J ] . Eur J Cancer , 2014 , 50 ( 10 ): 1697 - 1705 .

CHEN X , LI D W . Sequencing facility and DNA source associated patterns of virus-mappable reads in whole-genome sequencing data [J ] . Genomics , 2021 , 113 ( 1 Pt 2 ): 1189 - 1198 . DOI: 10.1016/j.ygeno.2020.12.004 http://doi.org/10.1016/j.ygeno.2020.12.004

LIU G C , CHEN X , LUAN Y H , et al. VirusPredictor: XGBoost-based software to predict virus-related sequences in human data [J ] . Bioinformatics , 2024 , 40 ( 4 ): btae192.

PALECZEK A , GROCHALA D , RYDOSZ A . Artificial breath classification using XGBoost algorithm for diabetes detection [J ] . Sensors , 2021 , 21 ( 12 ): 4187.

DONG B T , ZHANG H , DUAN Y Y , et al. Development of a machine learning-based model to predict prognosis of alpha-fetoprotein-positive hepatocellular carcinoma [J ] . J Transl Med , 2024 , 22 ( 1 ): 455.

LI B , EISENBERG N , BEATON D , et al. Using machine learning (XGBoost) to predict outcomes after infrainguinal bypass for peripheral artery disease [J ] . Ann Surg , 2024 , 279 ( 4 ): 705 - 713 .

浏览量

1637

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

早期乳腺癌术中快速冰冻切片病理学检查延迟诊断率分析：一项真实世界的回顾性研究

早期乳腺癌女性患者的骨健康管理中国专家共识（2022年版）

乳腺癌骨转移研究的新进展及展望——从机制到临床

改良EDTA脱钙法和酸脱钙法在乳腺癌骨转移免疫组织化学中的比较研究

中国抗癌协会乳腺癌诊治指南与规范（2026年版）