Stability-Guided Identification of Compact Molecular Biomarker Signatures for Breast Cancer Subtype Classification
Keywords:
Breast Cancer, Molecular Biomarkers, Gene Expression, Stability-Guided Feature Selection, Molecular Subtype ClassificationAbstract
Breast cancer is molecularly heterogeneous and accurate classification of breast cancer subtypes is important to understand the characteristics of the tumor and to facilitate precision medicine. The objective of this study was to determine stable molecular markers, create compact gene-expression signatures and test them for breast cancer subtype classification. A secondary-data analysis was performed using the METABRIC molecular data set (1,898 BC cases and 489 gene-expression features) in a retrospective fashion. Molecular differences were evaluated by the Kruskal–Wallis test and stability guided biomarker identification was performed by 50 iterations of stratified re-sample and re-ranking of features. Multinomial logistic regression with stratified five-fold cross validation was used to compare compact signatures with 5, 10, 15 and 20 genes. There were significant differences in gene-expression between molecular subtypes. Multiple biomarkers, such as GATA3, EGFR, AURKA, CCNE1 and CHEK1, showed selection to be consistent over resampling iterations. The 20-gene signature got a classification accuracy of 73.18%, a balanced accuracy of 66.13% and a macro-F1 score of 0.6760. The full panel of 489 genes obtained a slightly higher macro-F1 score of 0.7055. Recall across molecular subtypes was variable, and lowest for Normal-like tumors. These results show that there are compact molecular signatures that can be useful in subtype classification and can significantly decrease the feature complexity. Independent validation, further stability assessment, and biological verification, however, is needed to determine the reproducibility of the biomarkers and their potential applicability to breast cancer molecular classification.
References
1.Ab Mumin, N., Hamid, M. T. R., Wong, J. H. D., Rahmat, K., & Ng, K. H. (2022). Magnetic resonance imaging phenotypes of breast cancer molecular subtypes: a systematic review. Academic radiology, 29, S89-S106.
2.Ben Rabah, C., Sattar, A., Ibrahim, A., & Serag, A. (2025). A multimodal deep learning model for the classification of breast cancer subtypes. Diagnostics, 15(8), 995.
3.Beňačka, R., Szabóová, D., Guľašová, Z., Hertelyová, Z., & Radoňák, J. (2022). Classic and new markers in diagnostics and classification of breast cancer. Cancers, 14(21), 5444.
4.Brahmareddy, A., & Selvan, M. P. (2025). TransBreastNet a CNN transformer hybrid deep learning framework for breast cancer subtype classification and temporal lesion progression analysis. Scientific Reports, 15(1), 35106.
5.Carvalho, E., Canberk, S., Schmitt, F., & Vale, N. (2025). Molecular subtypes and mechanisms of breast cancer: precision medicine approaches for targeted therapies. Cancers, 17(7), 1102.
6.Choi, J. M., & Chae, H. (2023). moBRCA-net: a breast cancer subtype classification framework based on multi-omics attention neural networks. BMC bioinformatics, 24(1), 169.
7.Dogra, A. K., Prakash, A., Gupta, S., & Gupta, M. (2025). Prognostic significance and molecular classification of triple negative breast cancer: a systematic review. European Journal of Breast Health, 21(2), 101.
8.Furtney, I., Bradley, R., & Kabuka, M. R. (2023). Patient graph deep learning to predict breast cancer molecular subtype. IEEE/ACM transactions on computational biology and bioinformatics, 20(5), 3117-3127.
9.Jin, X., Zhou, Y. F., Ma, D., Zhao, S., Lin, C. J., Xiao, Y., ... & Shao, Z. M. (2023). Molecular classification of hormone receptor-positive HER2-negative breast cancer. Nature Genetics, 55(10), 1696-1708.
10.Jung, M., Song, S. G., Cho, S. I., Shin, S., Lee, T., Jung, W., ... & Kim, S. W. (2024). Augmented interpretation of HER2, ER, and PR in breast cancer by artificial intelligence analyzer: enhancing interobserver agreement through a reader study of 201 cases. Breast Cancer Research, 26(1), 31.
11.Lafcı, O., Celepli, P., Öztekin, P. S., & Koşar, P. N. (2023). DCE-MRI radiomics analysis in differentiating luminal A and luminal B breast cancer molecular subtypes. Academic radiology, 30(1), 22-29.
12.Li, B., & Nabavi, S. (2024). A multimodal graph neural network framework for cancer molecular subtype classification. BMC bioinformatics, 25(1), 27.
13.Li, Y. W., Dai, L. J., Wu, X. R., Zhao, S., Xu, Y. Z., Jin, X., ... & Shao, Z. M. (2024). Molecular characterization and classification of HER2-positive breast cancer inform tailored therapeutic strategies. Cancer Research, 84(21), 3669-3683.
14.Liu, T., Huang, J., Liao, T., Pu, R., Liu, S., & Peng, Y. (2022). A hybrid deep learning model for predicting molecular subtypes of human breast cancer using multimodal data. Irbm, 43(1), 62-74.
15.Lu, B., Natarajan, E., Balaji Raghavendran, H. R., & Markandan, U. D. (2023). Molecular classification, treatment, and genetic biomarkers in triple-negative breast cancer: a review. Technology in cancer research & treatment, 22, 15330338221145246.
16.Ma, M., Liu, R., Wen, C., Xu, W., Xu, Z., Wang, S., ... & Chen, W. (2022). Predicting the molecular subtype of breast cancer and identifying interpretable imaging features using machine learning algorithms. European Radiology, 32(3), 1652-1662.
17.Meyer, H. J., Wienke, A., & Surov, A. (2022). Diffusion-weighted imaging of different breast cancer molecular subtypes: a systematic review and meta-analysis. Breast Care, 17(1), 47-54.
18.Nissar, I., Alam, S., Masood, S., & Kashif, M. (2024). MOB-CBAM: A dual-channel attention-based deep learning generalizable model for breast cancer molecular subtypes prediction using mammograms. Computer Methods and Programs in Biomedicine, 248, 108121.
19.Orrantia-Borunda, E., Anchondo-Nuñez, P., Acuña-Aguilar, L. E., Gómez-Valles, F. O., & Ramírez-Valdespino, C. A. (2022). Subtypes of breast cancer. Breast Cancer [Internet].
20.Raghad Alharbi. (2021). Breast Cancer Gene Expression Profiles (METABRIC). https://doi.org/10.5281/zenodo.4768795.
21.Rajpal, S., Rajpal, A., Saggar, A., Vaid, A. K., Kumar, V., Agarwal, M., & Kumar, N. (2023). XAI-MethylMarker: Explainable AI approach for biomarker discovery for breast cancer subtype classification using methylation data. Expert Systems with Applications, 225, 120130.
22.Rakha, E. A., Tse, G. M., & Quinn, C. M. (2023). An update on the pathological classification of breast cancer. Histopathology, 82(1), 5-16.
23.Roy, M., Fowler, A. M., Ulaner, G. A., & Mahajan, A. (2023). Molecular classification of breast cancer. PET clinics, 18(4), 441-458.
24.Syrnioti, A., Petousis, S., Newman, L. A., Margioula-Siarkou, C., Papamitsou, T., Dinas, K., & Koletsa, T. (2024). Triple negative breast cancer: molecular subtype-specific immune landscapes with therapeutic implications. Cancers, 16(11), 2094.
25.Tafavvoghi, M., Sildnes, A., Rakaee, M., Shvetsov, N., Bongo, L. A., Busund, L. T. R., & Møllersen, K. (2025). Deep learning-based classification of breast cancer molecular subtypes from H&E whole-slide images. Journal of pathology informatics, 16, 100410.
26.Turova, P., Kushnarev, V., Baranov, O., Butusova, A., Menshikova, S., Yong, S. T., ... & Kotlov, N. (2025). The Breast Cancer Classifier refines molecular breast cancer classification to delineate the HER2-low subtype. NPJ Breast Cancer, 11(1), 19.
27.Weng, L., Zhou, J., Guo, S., Xu, N., & Ma, R. (2024). The molecular subtyping and precision medicine in triple-negative breast cancer---based on Fudan TNBC classification. Cancer cell international, 24(1), 120.
28.Xu, M. M., Yan, M. Y., Chen, M. Y., Chen, M. X., & Gong, M. K. (2025). Expression and subtype discordance between core needle biopsy and surgical specimen in breast cancer. Journal of Surgical Research, 307, 42-52.
29.Zhang, X. (2023). Molecular classification of breast cancer: relevance and challenges. Archives of pathology & laboratory medicine, 147(1), 46-51.
30.Zhu, J. Y., He, H. L., Jiang, X. C., Bao, H. W., & Chen, F. (2023). Multimodal ultrasound features of breast cancers: correlation with molecular subtypes. BMC Medical Imaging, 23(1), 57.