基于单细胞与机器学习的三阴性乳腺癌上皮细胞亚群解析与预后模型构建

吴金鹏; 郭雪; 刘恩谷; 林峰; 李鸿涛

doi:10.3971/j.issn.1000-8578.2026.25.0694

您当前的位置：

首页 >

文章列表页 >

基于单细胞与机器学习的三阴性乳腺癌上皮细胞亚群解析与预后模型构建

基础研究 | 更新时间：2026-11-05

- 基于单细胞与机器学习的三阴性乳腺癌上皮细胞亚群解析与预后模型构建
- Single-Cell and Machine Learning-Based Identification of Epithelial Subsets and Prognostic Modeling in Triple-Negative Breast Cancer
- 肿瘤防治研究 2026年53卷第4期页码：251-266
- 作者机构：
  
  1. 新疆医科大学附属肿瘤医院乳腺甲状腺外科,乌鲁木齐,830000
  2. 新疆医科大学临床医学院,乌鲁木齐,830017
- 作者简介：
  
  吴金鹏，男，硕士在读，主要从事乳腺癌的相关研究，ORCID: 0009-0004-6275-7452
  郭雪，女，硕士在读，主要从事乳腺癌的相关研究，ORCID: 0009-0008-6999-2999
  李鸿涛，男，博士，主任医师，教授，主要从事乳腺癌相关临床与教学、科研工作，E-mail: lht4656@163.com，ORCID: 0000-0003-2003-0015
- 基金信息：
  
  “天山英才”培养计划科技创新领军人才项目（No. 2023TSYCLJ0039）
- DOI：10.3971/j.issn.1000-8578.2026.25.0694
  中图分类号： R737.9
- 纸质出版：2026
- 稿件说明：
移动端阅览
吴金鹏, 郭雪, 刘恩谷, 林峰, 李鸿涛. 基于单细胞与机器学习的三阴性乳腺癌上皮细胞亚群解析与预后模型构建[J]. 肿瘤防治研究, 2026, 53(4): 251-266.

WU Jinpeng, GUO Xue, LIU Engu, et al. Single-Cell and Machine Learning-Based Identification of Epithelial Subsets and Prognostic Modeling in Triple-Negative Breast Cancer[J]. 2026, 53(4): 251-266.
吴金鹏, 郭雪, 刘恩谷, 林峰, 李鸿涛. 基于单细胞与机器学习的三阴性乳腺癌上皮细胞亚群解析与预后模型构建[J]. 肿瘤防治研究, 2026, 53(4): 251-266. DOI： 10.3971/j.issn.1000-8578.2026.25.0694.

WU Jinpeng, GUO Xue, LIU Engu, et al. Single-Cell and Machine Learning-Based Identification of Epithelial Subsets and Prognostic Modeling in Triple-Negative Breast Cancer[J]. 2026, 53(4): 251-266. DOI： 10.3971/j.issn.1000-8578.2026.25.0694.

摘要

目的

探讨三阴性乳腺癌（TNBC）上皮细胞的异质性与关键分子特征，筛选预后相关标志物并构建稳定的生存预测模型。

方法

基于TNBC单细胞转录组数据，对上皮细胞进行提取、标准化及聚类分型，并解析其分子特征与功能差异；采用hdWGCNA构建上皮细胞共表达模块，结合多种机器学习算法筛选关键预后基因并建立风险评分模型；通过ROC曲线与Kaplan-Meier（K-M）生存分析验证模型性能。进一步评估高、低风险组的免疫微环境特征及潜在药物反应差异，并采用PCR验证关键基因在肿瘤与正常组织中的表达差异。

结果

本研究解析了TNBC上皮细胞的亚群组成及其分子特征，识别出与TNBC相关的上皮细胞亚群。基于hdWGCNA与机器学习联合筛得10个关键基因，构建的预后模型能够有效区分不同生存风险人群，并在ROC与K-M分析中表现出良好的预测能力。免疫分析显示，高、低风险组在7类免疫细胞浸润水平及免疫功能相关特征上存在差异；药物敏感性评估提示8种药物在不同风险组间存在潜在响应差异。PCR进一步验证了10个关键基因在肿瘤与正常组织中的表达差异。

结论

本研究在单细胞分辨率下揭示了TNBC上皮细胞异质性，并建立了基于10基因的预后模型，可用于TNBC患者风险分层及免疫特征与潜在治疗策略评估。

Abstract

Objective

To investigate the heterogeneity and key molecular features of epithelial cells in triple-negative breast cancer (TNBC)

identify prognostic biomarkers

and develop a robust survival prediction model.

Methods

Using TNBC single-cell transcriptomic data

epithelial cells were extracted

normalized

and subclustered to cha

racterize their molecular signatures and functional differences. High-dimensional weighted gene co-expression network analysis (hdWGCNA) was applied to establish co-expression modules in epithelial cells. Multiple machine learning algorithms were integrated to select key prognostic genes and develop a risk-score model

whose performance was evaluated using receiver operating characteristic (ROC) curves and Kaplan-Meier (K-M) survival analysis. In addition

the immune microenvironment features and potential drug-response differences between the high- and low-risk groups were systematically assessed. Finally

PCR was performed to validate the expression differences of the key genes between tumor and normal tissues.

Results

We characterized the composition and molecular features of TNBC epithelial subpopulations and identified a TNBC-associated epithelial subset. By integrating hdWGCNA with machine learning approaches

10 key genes were selected to construct a prognostic model

which effectively stratified patients into distinct survival-risk groups and demonstrated favorable predictive performance in ROC and K-M analyses. Immune profiling revealed the differences in the infiltration levels of seven immune cell types and immune function-related features between the high- and low-risk groups. Drug-sensitivity analysis suggested potential differential responses to eight agents across the risk groups. PCR validation further confirmed the differential expression of the ten signature genes between tumor and normal tissues.

Conclusion

This study reveals epithelial heterogeneity in TNBC at single-cell resolution and establishes a 10-gene prognostic model

which may facilitate the stratification of TNBC risk and the evaluation of immune characteristics and potential therapeutic strategies.

关键词

Keywords

references

Garrido-Castro AC, Lin NU, Polyak K. Insights into Molecular Classifications of Triple-Negative Breast Cancer: Improving Patient Selection for Treatment[J]. Cancer Discov, 2019, 9(2): 176-198.

Bai X, Ni J, Beretov J, et al . Triple-negative breast cancer therapeutic resistance: where is the Achilles' heel?[J ] . Cancer Lett, 2021, 497: 100-111.

Loibl S, André F, Bachelot T, et al . Early breast cancer: ESMO Clinical Practice Guideline for diagnosis, treatment and follow-up[J ] . Ann Oncol, 2024, 35(2): 159-182.

Bao B, Prasad AS. Targeting CSC in a Most Aggressive Subtype of Breast Cancer TNBC[J]. Adv Exp Med Biol, 2019, 1152: 311-334.

So JY, Ohm J, Lipkowitz S, et al . Triple negative breast cancer (TNBC): non-genetic tumor heterogeneity and immune microenvironment: emerging treatment options[J ] . Pharmacol Ther, 2022, 237: 108253.

Hua Z, White J, Zhou J. Cancer stem cells in TNBC[J]. Semin Cancer Biol, 2022, 82: 26-34.

Spring LM, Fell G, Arfe A, et al . Pathologic complete response after neoadjuvant chemotherapy and impact on breast cancer recurrence and survival: a comprehensive meta-analysis[J ] . Clin Cancer Res, 2020, 26(12): 2838-2848.

Zhu H, Wei M, Xu J, et al . PARP inhibitors in pancreatic cancer: molecular mechanisms and clinical applications[J ] . Mol Cancer, 2020, 19(1): 49.

Morganti S, Marra A, De Angelis C, et al . PARP Inhibitors for Breast Cancer Treatment: A Review[J ] . JAMA Oncol, 2024, 10(5): 658-670.

Ye Z, Yuan J, Hong D, et al . Multimodal diagnostic models and subtype analysis for neoadjuvant therapy in breast cancer[J ] . Front Immunol, 2025, 16: 1559200.

Wang H, Zhu YN, Zhang S, et al . Transcriptome-wide analysis reveals potential roles of CFD and ANGPTL4 in fibroblasts regulating B cell lineage for extracellular matrix-driven clustering and novel avenues for immunotherapy in breast cancer[J ] . Mol Med, 2025, 31(1): 179.

Zhong L, Zhu J, Chen J, et al . MGAT4EP promotes tumor progression and ser ves as a prognostic marker for breast cancer[J ] . Cancer Biol Ther, 2025, 26(1): 2475604.

Zou Y, Xie J, Zheng S, et al . Leveraging diverse cell-death patterns to predict the prognosis and drug sensitivity of triple-negative breast cancer patients after surgery[J ] . Int J Surg, 2022, 107: 106936.

Chen C, Qi J, Lin W, et al . Single-cell ligand-receptor profiling reveals an immunotherapy-responsive subtype and prognostic signature in triple-negative breast cancer[J ] . Front Immunol, 2025, 16: 1590951.

Stuart T, Butler A, Hoffman P, et al . Comprehensive integration of single-cell data[J ] . Cell, 2019, 177(7): 1888-1902. e21.

Hafemeister C, Satija R. Normalization and variance stabilization of single-cell RNA-seq data using regularized negative binomial regression[J]. Genome Biol, 2019, 20(1): 296.

Tan Z, Chen X, Zuo J, et al . Comprehensive analysis of scRNA-Seq and bulk RNA-Seq reveals dynamic changes in the tumor immune microenvironment of bladder cancer and establishes a prognostic model[J ] . J Transl Med, 2023, 21(1): 223.

Jin S, Guerrero-Juarez CF, Zhang L, et al . Inference and analysis of cell-cell communication using CellChat[J ] . Nat Commun, 2021, 12(1): 1088.

Yu G, Wang LG, Han Y, et al . clusterProfiler: an R package for comparing biological themes among gene clusters[J ] . OMICS, 2012, 16(5): 284-287.

Hänzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA-seq data[J]. BMC Bioinformatics, 2013, 14: 7.

Trapnell C, Cacchiarelli D, Grimsby J, et al . The dynamics and regulators of cell fate decisions are revealed by pseudotemporal ordering of single cells[J ] . Nat Biotechnol, 2014, 32(4): 381-386.

Sun Y, Wu J, Zhang Q, et al . Single-cell hdWGCNA reveals metastatic protective macrophages and development of deep learning model in uveal melanoma[J ] . J Transl Med, 2024, 22(1): 695.

Wang X, Jin Y, Xu L, et al . Integrating Single-Cell RNA-Seq and Bulk RNA-Seq to Construct a Novel γδT Cell-Related Prognostic Signature for Human Papillomavirus-Infected Cervical Cancer[J ] . Cancer Control, 2024, 31: 10732748241274228.

Morabito S, Reese F, Rahimzadeh N, et al . hdWGCNA identifies co-expression networks in high-dimensional transcriptomics data[J ] . Cell Rep Methods, 2023, 3(6): 100498.

Tian Y, Xiao H, Yang Y, et al . Crosstalk between 5-methylcytosine and N6-methyladenosine machinery defines disease progression, therapeutic response and pharmacogenomic landscape in hepatocellular carcinoma[J ] . Mol Cancer, 2023, 22(1): 5.

Jia A, Xu L, Wang Y. Venn diagrams in bioinformatics[J]. Brief Bioinform, 2021, 22(5): bbab108.

Ishwaran H, Kogalur UB. Consistency of random survival forests[J]. Stat Probab Lett, 2010, 80(13-14): 1056-1064.

Sanz H, Valim C, Vegas E, et al . SVM-RFE: selection and visualization of the most relevant features through non-linear kernels[J ] . BMC Bioinformatics, 2018, 19(1): 432.

Dai P, Chang W, Xin Z, et al . Retrospective study on the influencing factors and prediction of hospitalization expenses for chronic renal failure in China based on random forest and LASSO regression[J ] . Front Public Health, 2021, 9: 678276.

Shen C, Chai W, Han J, et al . Identification and validation of a dysregulated TME-related gene signature for predicting prognosis, and immunological properties in bladder cancer[J ] . Front Immunol, 2023, 14: 1213947.

Lätti S, Niinivehmas S, Pentikäinen OT. Rocker: open source, easy-to-use tool for AUC and enrichment calculations and ROC visualization[J]. J Cheminform, 2016, 8(1): 45.

Wang Y, Xu J, Fang Y, et al . Comprehensive analysis of a novel signature incorporating lipid metabolism and immune-related genes for assessing prognosis and immune landscape in lung adenocarcinoma[J ] . Front Immunol, 2022, 13: 950001.

Zhao Z, Ding Y, Tran LJ, et al . Innovative breakthroughs facilitated by single-cell multi-omics: manipulating natural killer cell functionality correlates with a novel subcategory of melanoma cells[J ] . Front Immunol, 2023, 14: 1196892.

Gu X, Chen B, Zhang S, et al . The expression of CCL17 and potential prognostic value on tumor immunity in thyroid carcinoma based on bioinformatics analysis[J ] . Sci Rep, 2024, 14(1): 31580.

Zhao P, Zhen H, Zhao H, et al . Identification of hub genes and potential molecular mechanisms related to radiotherapy sensitivity in rectal cancer based on multiple datasets[J ] . J Transl Med, 2023, 21(1): 176.

Zhang X, Wu D, Zhang L, et al . Predicting the potential mechanism of radix chimonanthi pracecocis in treating osteoarthritis by network pharmacology analysis combined with experimental validation[J ] . J Ethnopharmacol, 2024, 331: 118231.

Karran JC, Moodie EE, Wallace MP. Statistical method use in public health research[J]. Scand J Public Health, 2015, 43(7): 776-782.

Jin S, Plikus MV, Nie Q. CellChat for systematic analysis of cell-cell communication from single-cell transcriptomics[J]. Nat Protoc, 2025, 20(1): 180-219.

Schmid P, Cortes J, Dent R, et al . Overall survival with pembrolizumab in early-stage triple-negative breast cancer[J ] . N Engl J Med, 2024, 391(21): 1981-1991.

Leon-Ferre RA, Jonas SF, Salgado R, et al . Tumor-infiltrating lymphocytes in triple-negative breast cancer[J ] . JAMA, 2024, 331(13): 1135-1144.

Wang X, Venet D, Lifrange F, et al . Spatial transcriptomics reveals substantial heterog eneity in triple-negative breast cancer wi th potential clinical implications[J ] . Nat Commun, 2024, 15(1): 10232.

Li C, Ji H, Zhuang S, et al . Update on the correlation between mitochondrial function and osteonecrosis of the femoral head osteocytes[J ] . Redox Rep, 2025, 30(1): 2491846.

Chaudhary S, Siddiqui JA, Pothuraju R, et al . Ribosome biogenesis, altered metabolism and ribotoxic stress response in pancreatic ductal adenocarcinoma tumor microenvironment[J ] . Cancer Lett, 2025, 612: 217484.

Lai Y, Zhong Z, Li R, et al . Unveiling the crucial role of CD8+ T cell and endothelial cell interaction in rheumatoid arthritis through the integrated analysis of spatial transcriptomics and bulk/single-cell RNA-seq[J ] . J Transl Autoimmun, 2025, 10: 100285.

Kamal KY, Trombetta-Lima M. Mechanotransduction and skeletal muscle atrophy: the interplay between focal adhesions and oxidative stress[J]. Int J Mol Sci, 2025, 26(6): 2802.

Sumaiya K, Langford D, Natarajaseenivasan K, et al . Macrophage migration inhibitory factor (MIF): a multifaceted cytokine regulated by genetic and physiological strategies[J ] . Pharmacol Ther, 2022, 233: 108024.

Jankauskas SS, Wong DWL, Bucala R, et al . Evolving complexity of MIF signaling[J ] . Cell Signal, 2019, 57: 76-88.

Grünewald TGP, Cidre-Aranaz F, Surdez D, et al . Ewing sarcoma[J ] . Nat Rev Dis Primers, 2018, 4(1): 5.

Cortes J, Rugo HS, Cescon DW, et al . Pembrolizumab plus chemotherapy in advanced triple-negative breast cancer[J ] . N Engl J Med, 2022, 387(3): 217-226.

Ruegenberg S, Horn M, Pichlo C, et al . Loss of GFAT-1 feedback regulation activates the hexosamine pathway that modulates protein homeostasis[J ] . Nat Commun, 2020, 11(1): 687.

Chang B, Liu G, Yang G, et al . REDD1 is required for RAS-mediated transformation of human ovarian epithelial cells[J ] . Cell Cycle, 2009, 8(5): 780-786.

Jin HO, Hong SE, Kim JY, et al . Induction of HSP27 and HSP70 by constitutive overexpression of Redd1 confers resistance of lung cancer cells to ionizing radiation[J ] . Oncol Rep, 2019, 41(5): 3119-3126.

Zeng Q, Liu J, Cao P, et al . Inhibition of REDD1 sensit izes bladder urothelial carcinoma to paclitaxel by inhibiting autophagy[J ] . Clin Cancer Res, 2018, 24(2): 445-459.

Foltyn M, Luger AL, Lorenz NI, et al . The physiological mTOR complex 1 inhibitor DDIT4 mediates therapy resistance in glioblastoma[J ] . Br J Cancer, 2019, 120(5): 481-487.

Jiang T, Zhu J, Jiang S, et al . Targeting lncRNA DDIT4-AS1 sensitizes triple negative breast cancer to chemotherapy via suppressing of autophagy[J ] . Adv Sci (Weinh), 2023, 10(17): e2207257.

Zheng HC, Xue H, Zhang CY, et al . The clinicopathological significances and related signal pathways of BTG3 mRNA expression in cancers: a bioinformatics analysis[J ] . Front Genet, 2022, 13: 1006582.

Zhang Q, Li Y, Zhu Q, et al . TRIM65 promotes renal cell carcinoma through ubiquitination and degradation of BTG3[J ] . Cell Death Dis, 2024, 15(5): 355.

Luo P, Li Z, He H, et al . Exploring the impact of RPN1 on tumorigenesis and immune response in cancer[J ] . FASEB J, 2025, 39(6): e70345.

Morin-Buote J, Ennour-Idrissi K, Poirier É, et al . Associatio n of breast tumour expression of cannabinoid receptors CBR1 and CBR2 with prognostic factors and survival in breast cancer patients[J ] . J Pers Med, 2021, 11(9): 852.

Caffarel MM, Andradas C, Pérez-Gómez E, et al . Cannabinoids: a new hope for breast cancer therapy?[J ] . Cancer Treat Rev, 2012, 38(7): 911-918.

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

基于机器学习构建非小细胞肺癌恶性胸腔积液的诊断和预后模型

金纳米颗粒辅助超声疗法重塑三阴性乳腺癌免疫微环境的研究进展

基于转录组学分析探讨PI3K/Akt通路在三阴性乳腺癌表阿霉素耐药中的作用

基于机器学习预测模型探索慢性丙型肝炎患者发生原发性肝癌的风险因素

叶酸修饰脂质体槲皮素经JAK2/STAT3信号通路介导的线粒体凋亡途径诱导三阴性乳腺癌细胞凋亡