A Neural Network-Driven Fingerprint Matrix Model for the Predictive Diagnosis of Incipient Breast Cancer Progression
Keywords:
ctDNA monitoring, Deep learning prognosis, Multi-modal breast cancer, Risk stratification, SHAP interpretabilityAbstract
Background: Early-stage breast cancer prognosis remains challenging despite adjuvant therapies, necessitating multi-modal integration for refined risk stratification. This study leverages a cohort of 1200 patients with clinical, genomic, radiomic, and longitudinal ctDNA data to develop an interpretable deep learning fingerprint model for 3-year progression prediction.
Methods: Data preprocessing ensured >85% quality via imputation and normalization. A multi-branch neural network (43 layers, 41k parameters) fused modalities (clinical 15 dims, genomic 250, radiomic 200, ctDNA 5). Performance was evaluated on a 20% test set (n=240), with survival via Kaplan-Meier and Cox models. Interpretability employed permutation importance, SHAP, and pathway enrichment.
Results: Clinical features were 100% complete, genomic mutations peaked at PIK3CA (16%) and TP53 (15%), and ctDNA rose from 0.48% to 0.52% over 12 months, with 29.9% progression and 75% 3-year overall survival. The model achieved test AUC 0.861, accuracy 0.933, precision 0.933, and recall 0.389 (F1=0.549), outperforming clinical criteria (AUC 0.509) and Oncotype-like scores (0.729; ΔAUC +0.308). Risk tertiles stratified progression-free survival (low-risk 95%, high-risk 50%; HR 0.25). BRCA1 mutation (importance 0.04) and Ki67 (r=0.04) dominated features, with cell cycle (SHAP 0.082) and DNA repair defects (0.055) explaining 55% pathway attributions.
Conclusions: Multi-modal fusion yields superior, interpretable prognostics, enabling 25% intermediate reclassification (NRI=0.32) for de-escalation in 70% low-risk cases. This paradigm advances precision oncology, potentially reducing overtreatment while addressing ER- disparities.
References
C. B. Rabah, A. Sattar, A. Ibrahim, and A. Serag, “A Multimodal Deep Learning Model for the Classification of Breast Cancer Subtypes,” Diagnostics, vol. 15, no. 8, pp. 995–995, Apr. 2025, doi: https://doi.org/10.3390/diagnostics15080995
C. V. Mulder, E. Jongbloed, M. C. Liefaard, “A comparative study of four cell-free DNA assays for detecting circulating tumor DNA in early breast cancer,” Breast Cancer Research, vol. 27, no. 1, Jul. 2025, doi: https://doi.org/10.1186/s13058-025-02077-8
L. J. Esserman, I. Thompson, B. Reid, “Addressing overdiagnosis and overtreatment in cancer: a prescription for change,” The Lancet Oncology, vol. 15, no. 6, pp. e234–e242, May 2014, doi: https://doi.org/10.1016/s1470-2045(13)70598-9
I. Garcia-Murillas, “Mutation tracking in circulating tumour DNA predicts relapse in early breast cancer,” Science Translational Medicine, vol. 7, no. 302, pp. 302ra133–302ra133, Aug. 2015, doi: https://doi.org/10.1126/scitranslmed.aab0021
K. He, X. Zhang, S. Ren, and J. Sun, “Deep Residual Learning for Image Recognition,” Arxiv, Dec. 2015, doi: https://doi.org/10.48550/arxiv.1512.03385
J. Aronson, M. Bhatta, L. A. Carey, K. Jobanputra, G. P. Gupta, and Y. Abdou, “Bridging the gap: ctDNA, genomics, and equity in breast cancer care,” Npj Breast Cancer, vol. 11, no. 1, Aug. 2025, doi: https://doi.org/10.1038/s41523-025-00811-1
F. Panet, A. Papakonstantinou, M. Borrell, J. Vivancos, A. Vivancos, and M. Oliveira, “Use of ctDNA in early breast cancer: analytical validity and clinical potential,” NPJ Breast Cancer, vol. 10, no. 1, pp. 1–15, Jun. 2024, doi: https://doi.org/10.1038/s41523-024-00653-3
S. Easaw, J. Hsu, N. Steuerwald, and A. L. Heeke, “Liquid clues: tracking early-stage breast cancer with ctDNA - a mini review,” Frontiers in Oncology, vol. 15, Aug. 2025, doi: https://doi.org/10.3389/fonc.2025.1634859
H. Pan, R. Gray, J. Braybrooke, “20-Year Risks of Breast-Cancer Recurrence after Stopping Endocrine Therapy at 5 Years,” New England Journal of Medicine, vol. 377, no. 19, pp. 1836–1846, Nov. 2017, doi: https://doi.org/10.1056/nejmoa1701830
D. Kingma and J. Ba, “Adam: A Method for Stochastic Optimization,” Computer Science, 2014, doi: https://doi.org/10.48550/arXiv.1412.6980
T. Nishizawa, T. Maldjian, Z. Jiao, and T. Q. Duong, “Attention-based multimodal deep learning for interpretable and generalizable prediction of pathological complete response in breast cancer,” Journal of Translational Medicine, vol. 23, no. 1, Jul. 2025, doi: https://doi.org/10.1186/s12967-025-06617-w
A. Llinas-Bertran, M. Butjosa-Espín, V. Barberi, and J. A. Seoane, “Multimodal data integration in early-stage breast cancer,” The Breast, vol. 80, p. 103892, Jan. 2025, doi: https://doi.org/10.1016/j.breast.2025.103892
S. Lundberg and S.-I. Lee, “A Unified Approach to Interpreting Model Predictions,” arXiv.org, Nov. 24, 2017. https://arxiv.org/abs/1705.07874v2
N. Moldovan, Y. van der Pol, “Multi-modal cell-free DNA genomic and fragmentomic patterns enhance cancer survival and recurrence analysis,” Cell Reports Medicine, vol. 5, no. 1, pp. 101349–101349, Dec. 2023, doi: https://doi.org/10.1016/j.xcrm.2023.101349
A. McKenna, M. Hanna, E. Banks, “The Genome Analysis Toolkit: a MapReduce framework for analyzing next-generation DNA sequencing data,” Genome Research, vol. 20, no. 9, pp. 1297–303, 2010, doi: https://doi.org/10.1101/gr.107524.110
T. Mong, T. Hai Phan, T. Xuan Jasmine, “Multimodal analysis of genome-wide methylation, copy number aberrations, and end motif signatures enhances detection of early-stage breast cancer,” Frontiers in Oncology, vol. 13, May 2023, doi: https://doi.org/10.3389/fonc.2023.1127086
R. Yasrab, R. Agrawal, M. Mohamed Saber-Ayad, and M. El-Hadidi, “Decoding Breast Cancer Heterogeneity via Multi-Omics Integration and Language Model-Based Interpretation,” bioRxiv (Cold Spring Harbor Laboratory), Jun. 2025, doi: https://doi.org/10.1101/2025.06.26.661832
A. Maigari, C. XinYing, and Z. Zainol, “Multimodal deep learning breast cancer prognosis models: narrative review on multimodal architectures and concatenation approaches,” Journal of Medical Artificial Intelligence, vol. 8, pp. 61–61, Dec. 2025, doi: https://doi.org/10.21037/jmai-24-146
T. Sorlie , C. M. Perou, R. Tibshirani, “Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications,” Proceedings of the National Academy of Sciences, vol. 98, no. 19, pp. 10869–10874, Sep. 2001, doi: https://doi.org/10.1073/pnas.191367098
A. Dhungana, A. Vannier, F. Zhao,“Development and validation of a clinical breast cancer tool for accurate prediction of recurrence,” NPJ Breast Cancer, vol. 10, no. 1, Jun. 2024, doi: https://doi.org/10.1038/s41523- 024-00651-5
J. A. Sparano, R. J. Gray, D. F. Makower, “Prospective Validation of a 21-Gene Expression Assay in Breast Cancer,” New England Journal of Medicine, vol. 373, no. 21, pp. 2005–2014, Nov. 2015, doi: https://doi.org/10.1056/nejmoa1510764
A. L. Riediger, C. Meinzer, “Combined radiomics and liquid biopsy reflect tumor biology towards multimodal non-invasive prostate cancer risk stratification,” Medrxiv, Sep. 2025, doi: https://doi.org/10.1101/2025.09.06.25335220
T. T. V. Van, T. H. Tran, T. H. H. Nguyen, “Multimodal analysis of cell-free DNA enhances differentiation of early-stage breast cancer from benign lesions and healthy individuals,” BMC Biology, vol. 23, no. 1, Aug. 2025, doi: https://doi.org/10.1186/s12915-025-02371-z
G. Villacampa, T. Pascual., P. Pascual, “HER2DX and survival outcomes in early-stage HER2-positive breast cancer: an individual patient-level meta-analysis,” The Lancet Oncology, Jul. 2025, doi: https://doi.org/10.1016/s1470-2045(25)00276-1
H. Xue, H. Xu, and K. Wang, “Adaptive multi-modal temporal fusion network with dynamic synergistic integration for breast cancer survival prediction,” Biomedical Signal Processing and Control, vol. 112, pp. 108640–108640, Sep. 2025, doi: https://doi.org/10.1016/j.bspc.2025.108640
J. Zhang, Y. Che, R. Liu, Z. Wang, and W. Liu, “Deep learning–driven multi-omics analysis: enhancing cancer diagnostics and therapeutics,” Briefings in Bioinformatics, vol. 26, no. 4, Jul. 2025, doi: https://doi.org/10.1093/bib/bbaf440
F. Abdullakutty, Y. Akbari, S. Al-Maadeed, A. Bouridane, and R. Rifat Hamoudi, “Enhancing the Prediction of Breast Cancer Progression Through Multi-modal Data Transformation,” Cognitive Computation, vol. 17, no. 3, Jun. 2025, doi: https://doi.org/10.1007/s12559-025-10474-6
J. Chen, T. Pan, “A deep learning-based multimodal medical imaging model for breast cancer screening,” Scientific Reports, vol. 15, no. 1, Apr. 2025, doi: https://doi.org/10.1038/s41598-025-99535-2
B. Uyar, T. Savchyn, “Flexynesis: A deep learning toolkit for bulk multi-omics data integration for precision oncology and beyond,” Nature Communications, vol. 16, no. 1, Sep. 2025, doi: https://doi.org/10.1038/s41467-025-63688-5
C. B. Rabah, A. Sattar, A. Ibrahim, and A. Serag, “A Multimodal Deep Learning Model for the Classification of Breast Cancer Subtypes,” Diagnostics, vol. 15, no. 8, pp. 995–995, Apr. 2025, doi: https://doi.org/10.3390/diagnostics15080995
H. Chai, X. Zhou, Z. Zhang, J. Rao, H. Zhao, and Y. Yang, “Integrating multi-omics data through deep learning for accurate cancer prognosis prediction,” Computers in Biology and Medicine, vol. 134, p. 104481, Jul. 2021, doi: https://doi.org/10.1016/j.compbiomed.2021.104481
H. Chai, X. Zhou, Z. Zhang, J. Rao, H. Zhao, and Y. Yang, “Integrating multi-omics data through deep learning for accurate cancer prognosis prediction,” Computers in Biology and Medicine, vol. 134, p. 104481, Jul. 2021, doi: https://doi.org/10.1016/j.compbiomed.2021.104481
Z. Wang “Deep learning-based multi-modal data integration enhancing breast cancer disease-free survival prediction,” Precision Clinical Medicine, vol. 7, no. 2, May 2024, doi: https://doi.org/10.1093/pcmedi/pbae012
E. Krasniqi, L. Filomeno, “Multimodal deep learning for predicting neoadjuvant treatment outcomes in breast cancer: a systematic review,” Biology Direct, vol. 20, no. 1, Jun. 2025, doi: https://doi.org/10.1186/s13062-025-00661-8
E. Kang, J. Cheun, “Prognostic Performance of the Next-Generation Sequencing-Based Multigene Assay in Early Breast Cancer Patients Treated According to the 21-Gene Assay Results,” Cancer Research and Treatment, vol. 57, no. 3, pp. 749–759, Dec. 2024, doi: https://doi.org/10.4143/crt.2024.1035
D. Jia, C. Chen, “Breast Cancer Case Identification Based on Deep Learning and Bioinformatics Analysis,” Frontiers in Genetics, vol. 12, May 2021, doi: https://doi.org/10.3389/fgene.2021.628136
F. Garlan, P. Laurent-Puig, D. Sefrioui, “Early Evaluation of Circulating Tumor DNA as Marker of Therapeutic Efficacy in Metastatic Colorectal Cancer Patients (PLACOL Study),” Clinical Cancer Research, vol. 23, no. 18, pp. 5416–5425, Jun. 2017, doi: https://doi.org/10.1158/1078-0432.ccr-16-3155
L. Zhang, L. Cao, “Biologically interpretable deep learning to predict response to immunotherapy in advanced melanoma using mutations and copy number variations,” Research Square, Jul. 2022, doi: https://doi.org/10.21203/rs.3.rs-1784695/v1
J. Hao, Y. Kim, T. Mallavarapu, J. H. Oh, and M. Kang, “Interpretable deep neural network for cancer survival analysis by integrating genomic and clinical data,” BMC Medical Genomics, vol. 12, no. S10, Dec. 2019, doi: https://doi.org/10.1186/s12920-019-0624-2
H. Yang, J. Wang, “MMsurv: a multimodal multi-instance multi-cancer survival prediction model integrating pathological images, clinical information, and sequencing data,” Briefings in Bioinformatics, vol. 26, no. 3, May 2025, doi: https://doi.org/10.1093/bib/bbaf209
F. Sartori, F. Codicè, I. Caranzano, “A Comprehensive Review of Deep Learning Applications with Multi-Omics Data in Cancer Research,” Genes, vol. 16, no. 6, pp. 648–648, May 2025, doi: https://doi.org/10.3390/genes16060648
