Radiopathomic Graph Deep Learning for Multiscale Spatial-Contextual Modeling of Intratumoral Heterogeneity to Predict Breast Cancer Response to Neoadjuvant Therapy

Clinical Snapshot

Key Findings

• P Value: P < 0.05 for superiority over single-modality models

• Effect Size: AUC 0.91 (95% CI: 0.87, 0.94) for combined radiopathomic model in external validation

• Primary Outcome: Pathologic complete response prediction accuracy

• Nnt Or Sensitivity: Combined model significantly outperformed radiomic alone (AUC 0.84) and pathomic alone (AUC 0.83) with positive NRI and IDI

• Confidence Interval: 95% CI: 0.87, 0.94

Clinical Application

Requires specialized imaging analysis software and pathology digitization capabilities, limiting immediate widespread implementation Could complement existing Australian breast cancer guidelines (NBOCC) for neoadjuvant therapy selection. Implementation would require TGA approval for AI-based medical devices and integration with existing radiology/pathology workflows Women with invasive breast cancer being considered for neoadjuvant therapy, particularly those requiring enhanced prognostic stratification

Abstract

Purpose To develop an explainable radiopathomic graph deep learning (RPGDL) system for multiscale spatial-contextual modeling of intratumoral heterogeneity and evaluate its performance for the prediction of pathologic complete response (pCR) to neoadjuvant therapy in breast cancer. Materials and Methods The RPGDL system was developed from dual-center retrospective analysis of patients with biopsy-proven invasive breast cancer (May 2018-August 2024). For each tumor, individual radiomic and pathomic graphs were generated from pretherapeutic MRI and hematoxylin-eosin-stained biopsy slide images, respectively. These graphs were then processed by three distinct graph neural networks (GNNs): radiomic, pathomic, and radiopathomic. GNN performance was assessed with the area under the receiver operating characteristic curve (AUC), net reclassification index (NRI), and integrated discrimination improvement (IDI). A multifaceted approach was used to explain the GNNs' predictions. Results The training set included 582 (mean age, 52 years ± 9 [SD]) patients and the external test set 468 (50 years ± 10) patients from centers 1 and 2, respectively. The radiomic GNN achieved AUCs of 0.89 (95% CI: 0.85, 0.93) in the training set and 0.84 (95% CI: 0.80, 0.89) in the external test set; the pathomic GNN achieved AUCs of 0.87 (95% CI: 0.83, 0.91) in the training set and 0.83 (95% CI: 0.78, 0.88) in the external test set, with no significant difference between them (P > .05). The radiopathomic GNN outperformed both single-modality GNNs (training set: AUC, 0.95 [95% CI: 0.92, 0.98]; external test set: AUC, 0.91 [95% CI: 0.87, 0.94]; P < .05; NRI and IDI confirmed). Pathomic graphs dominated probability increases for pCR predictions, while radiomic graphs drove probability decreases for non-pCR predictions. Multifaceted analyses verified GNNs' explainability. Conclusion The developed RPGDL system enabled multiscale spatial-contextual intratumoral heterogeneity modeling for high-performance, explainable prediction of pCR to neoadjuvant therapy in breast cancer. Keywords: Dynamic Contrast-enhanced MRI, Breast, Tumor Response, Radiology-Pathology Integration, Prognosis, Principal Component Analysis, Perception, Supervised Learning, Reconstruction Algorithm Supplemental material is available for this article. © RSNA, 2026.

References

1. 1.Zhou, L. Q., Wang, X. Y., Xu, Y., Zhang, H. X., Yang, X. X., Jin, R. Q., Sang, X. Q., Zhu, Y. M., Meng, H. X., & Kuai, Z. X. (2026). Radiopathomic graph deep learning for multiscale spatial-contextual modeling of intratumoral heterogeneity to predict breast cancer response to neoadjuvant therapy. Radiology: Artificial Intelligence, 8(3), e250760. https://doi.org/10.1148/ryai.250760