Intratumor graph neural network recovers hidden prognostic value of multi-biomarker spatial heterogeneity

The association between tumor-stromal collagen features and the clinical outcomes of patients with breast cancer: a systematic review

BACKGROUND

The tumor microenvironment (TME), particularly the extracellular matrix (ECM), plays a crucial role in regulating breast cancer progression. Among ECM components, collagen type I-accounting for over 90% of fibrillar collagen in the human body-is the primary structural component of the tumor ECM. It critically modulates tumor cell behavior, influencing migration, invasion, and therapy resistance. The structural organization of collagen type I fibers can significantly impact clinical outcomes.

METHODS

This systematic review aimed to assess the association between tumor-stromal collagen type I characteristics and clinical outcomes in breast cancer. A comprehensive search strategy identified studies from major databases, which were appraised using quality assessment tools. Data on collagen quantity, morphology, alignment, and organization were extracted and analyzed to explore their relationship with survival, metastasis, therapy resistance, and clinical characteristics of breast cancer.

RESULTS

Our analysis revealed that increased collagen density-particularly with an organized/aligned fiber orientation-was strongly associated with poor prognosis. Specifically, increased intratumoral collagen quantity was linked to reduced overall survival (HR = 7.84, p = 0.031). Stage III tumors exhibiting elevated collagen uniformity showed higher metastasis rates (p = 0.004), and HER2⁺ tumors with high collagen content correlated with resistance to HER2-targeted therapies (p < 0.05). Furthermore, higher collagen curviness was associated with better outcomes, including a reduced recurrence risk (HR = 0.77, p < 0.001). Subtype-specific trends emerged as ER/PR-negative tumors more frequently exhibited a perpendicular collagen arrangement (p = 0.02), whereas ER/PR-positive tumors showed elevated COL1A1 expression (p < 0.0001). Despite these patterns, the heterogeneity of study methodologies and the complexity of the tumor microenvironment highlight the need for unified frameworks to advance clinical translation.

CONCLUSIONS

This review highlights the prognostic significance of tumor-stromal collagen characteristics in breast cancer, suggesting that future research should focus on the molecular mechanisms underlying collagen remodeling and its potential as a cancer biomarker and therapeutic target.

Softness or Stiffness What Contributes to Cancer and Cancer Metastasis?

Beyond the genomic and proteomic analysis of bulk and single cancer cells, a new focus of cancer research is emerging that is based on the mechanical analysis of cancer cells. Therefore, several biophysical techniques have been developed and adapted. The characterization of cancer cells, like human cancer cell lines, started with their mechanical characterization at mostly a single timepoint. A universal hypothesis has been proposed that cancer cells need to be softer to migrate and invade tissues and subsequently metastasize in targeted organs. Thus, the softness of cancer cells has been suggested to serve as a universal physical marker for the malignancy of cancer types. However, it has turned out that there exists the opposite phenomenon, namely that stiffer cancer cells are more migratory and invasive and therefore lead to more metastases. These contradictory results question the universality of the role of softness of cancer cells in the malignant progression of cancers. Another problem is that the various biophysical techniques used can affect the mechanical properties of cancer cells, making it even more difficult to compare the results of different studies. Apart from the instrumentation, the culture and measurement conditions of the cancer cells can influence the mechanical measurements. The review highlights the main advances of the mechanical characterization of cancer cells, discusses the strength and weaknesses of the approaches, and questions whether the passive mechanical characterization of cancer cells is still state-of-the art. Besides the cell models, conditions and biophysical setups, the role of the microenvironment on the mechanical characteristics of cancer cells is presented and debated. Finally, combinatorial approaches to determine the malignant potential of tumors, such as the involvement of the ECM, the cells in a homogeneous or heterogeneous association, or biological multi-omics analyses, together with the dynamic-mechanical analysis of cancer cells, are highlighted as new frontiers of research.

Graph neural networks in histopathology: Emerging trends and future directions

Histopathological analysis of whole slide images (WSIs) has seen a surge in the utilization of deep learning methods, particularly Convolutional Neural Networks (CNNs). However, CNNs often fail to capture the intricate spatial dependencies inherent in WSIs. Graph Neural Networks (GNNs) present a promising alternative, adept at directly modeling pairwise interactions and effectively discerning the topological tissue and cellular structures within WSIs. Recognizing the pressing need for deep learning techniques that harness the topological structure of WSIs, the application of GNNs in histopathology has experienced rapid growth. In this comprehensive review, we survey GNNs in histopathology, discuss their applications, and explore emerging trends that pave the way for future advancements in the field. We begin by elucidating the fundamentals of GNNs and their potential applications in histopathology. Leveraging quantitative literature analysis, we explore four emerging trends: Hierarchical GNNs, Adaptive Graph Structure Learning, Multimodal GNNs, and Higher-order GNNs. Through an in-depth exploration of these trends, we offer insights into the evolving landscape of GNNs in histopathological analysis. Based on our findings, we propose future directions to propel the field forward. Our analysis serves to guide researchers and practitioners towards innovative approaches and methodologies, fostering advancements in histopathological analysis through the lens of graph neural networks.

Deep Learning in Breast Cancer Imaging: A Decade of Progress and Future Directions

Breast cancer has reached the highest incidence rate worldwide among all malignancies since 2020. Breast imaging plays a significant role in early diagnosis and intervention to improve the outcome of breast cancer patients. In the past decade, deep learning has shown remarkable progress in breast cancer imaging analysis, holding great promise in interpreting the rich information and complex context of breast imaging modalities. Considering the rapid improvement in deep learning technology and the increasing severity of breast cancer, it is critical to summarize past progress and identify future challenges to be addressed. This paper provides an extensive review of deep learning-based breast cancer imaging research, covering studies on mammograms, ultrasound, magnetic resonance imaging, and digital pathology images over the past decade. The major deep learning methods and applications on imaging-based screening, diagnosis, treatment response prediction, and prognosis are elaborated and discussed. Drawn from the findings of this survey, we present a comprehensive discussion of the challenges and potential avenues for future research in deep learning-based breast cancer imaging.

Towards next-generation diagnostic pathology: AI-empowered label-free multiphoton microscopy

Diagnostic pathology, historically dependent on visual scrutiny by experts, is essential for disease detection. Advances in digital pathology and developments in computer vision technology have led to the application of artificial intelligence (AI) in this field. Despite these advancements, the variability in pathologists' subjective interpretations of diagnostic criteria can lead to inconsistent outcomes. To meet the need for precision in cancer therapies, there is an increasing demand for accurate pathological diagnoses. Consequently, traditional diagnostic pathology is evolving towards "next-generation diagnostic pathology", prioritizing on the development of a multi-dimensional, intelligent diagnostic approach. Using nonlinear optical effects arising from the interaction of light with biological tissues, multiphoton microscopy (MPM) enables high-resolution label-free imaging of multiple intrinsic components across various human pathological tissues. AI-empowered MPM further improves the accuracy and efficiency of diagnosis, holding promise for providing auxiliary pathology diagnostic methods based on multiphoton diagnostic criteria. In this review, we systematically outline the applications of MPM in pathological diagnosis across various human diseases, and summarize common multiphoton diagnostic features. Moreover, we examine the significant role of AI in enhancing multiphoton pathological diagnosis, including aspects such as image preprocessing, refined differential diagnosis, and the prognostication of outcomes. We also discuss the challenges and perspectives faced by the integration of MPM and AI, encompassing equipment, datasets, analytical models, and integration into the existing clinical pathways. Finally, the review explores the synergy between AI and label-free MPM to forge novel diagnostic frameworks, aiming to accelerate the adoption and implementation of intelligent multiphoton pathology systems in clinical settings.

Personalized decision support system for tailoring IgA nephropathy treatment strategies

BACKGROUND

The ongoing debate surrounding the use of immunosuppressive treatments for IgA nephropathy (IgAN) underscores the demand for personalized and effective strategies.

METHODS

Analyzed data from 807 IgAN patients over 5+ years using three methods: Random Forest with molecular biomarkers, network biomarkers with graph engineering, and an auto-encoder model. All models were trained using identical demographic, clinical, and pathological data, employing an 80-20 split for training and testing purposes.

RESULTS

In the comprehensive assessment of IgAN prognosis, the Random Forest model, employing molecular biomarkers, demonstrated strong performance metrics (AUC = 0.83, sensitivity = 0.51, specificity = 0.96). However, traditional graph feature engineering on patient-specific networks outperformed these results with an AUC of 0.90, sensitivity of 0.64, and specificity of 0.94. The Auto-encoder model showed the best accuracy (AUC = 0.91, sensitivity = 0.46, specificity = 0.96). The findings highlighted the superior predictive capabilities of network biomarkers over molecular biomarkers for adverse renal outcome prediction in IgAN. Consequently, we integrated Auto-encoder-derived Network Biomarkers with Random Forest Models to enhance prognostic precision in diverse IgAN treatment scenarios. The prediction for the prognosis of patients receiving supportive care, glucocorticoid therapy, and immunosuppressant treatment yielded AUC values of 0.95, 0.96, and 1, respectively, indicating high specificity. Drawing from these insights, we pioneered the development of an innovative decision support model for IgAN treatment. This model demonstrated the ability to make medical decisions comparable to those by experienced nephrologists, enabling the customization of personalized disease management strategies.

CONCLUSION

Our system accurately predicted IgAN prognosis and evaluated various treatment efficacies, aiding physicians in devising optimal therapeutic strategies for patients.

"Patchiness" in mechanical stiffness across a tumor as an early-stage marker for malignancy

Mechanical phenotyping of tumors, either at an individual cell level or tumor cell population level is gaining traction as a diagnostic tool. However, the extent of diagnostic and prognostic information that can be gained through these measurements is still unclear. In this work, we focus on the heterogeneity in mechanical properties of cells obtained from a single source such as a tissue or tumor as a potential novel biomarker. We believe that this heterogeneity is a conventionally overlooked source of information in mechanical phenotyping data. We use mechanics-based in-silico models of cell-cell interactions and cell population dynamics within 3D environments to probe how heterogeneity in cell mechanics drives tissue and tumor dynamics. Our simulations show that the initial heterogeneity in the mechanical properties of individual cells and the arrangement of these heterogenous sub-populations within the environment can dictate overall cell population dynamics and cause a shift towards the growth of malignant cell phenotypes within healthy tissue environments. The overall heterogeneity in the cellular mechanotype and their spatial distributions is quantified by a "patchiness" index, which is the ratio of the global to local heterogeneity in cell populations. We observe that there exists a threshold value of the patchiness index beyond which an overall healthy population of cells will show a steady shift towards a more malignant phenotype. Based on these results, we propose that the "patchiness" of a tumor or tissue sample, can be an early indicator for malignant transformation and cancer occurrence in benign tumors or healthy tissues. Additionally, we suggest that tissue patchiness, measured either by biochemical or biophysical markers, can become an important metric in predicting tissue health and disease likelihood just as landscape patchiness is an important metric in ecology.

Graph Neural Networks in Cancer and Oncology Research: Emerging and Future Trends

Next-generation cancer and oncology research needs to take full advantage of the multimodal structured, or graph, information, with the graph data types ranging from molecular structures to spatially resolved imaging and digital pathology, biological networks, and knowledge graphs. Graph Neural Networks (GNNs) efficiently combine the graph structure representations with the high predictive performance of deep learning, especially on large multimodal datasets. In this review article, we survey the landscape of recent (2020-present) GNN applications in the context of cancer and oncology research, and delineate six currently predominant research areas. We then identify the most promising directions for future research. We compare GNNs with graphical models and "non-structured" deep learning, and devise guidelines for cancer and oncology researchers or physician-scientists, asking the question of whether they should adopt the GNN methodology in their research pipelines.

GNN-surv: Discrete-Time Survival Prediction Using Graph Neural Networks

Survival prediction models play a key role in patient prognosis and personalized treatment. However, their accuracy can be improved by incorporating patient similarity networks, which uncover complex data patterns. Our study uses Graph Neural Networks (GNNs) to enhance discrete-time survival predictions (GNN-surv) by leveraging relationships in these networks. We build these networks using cancer patients' genomic and clinical data and train various GNN models on them, integrating Logistic Hazard and PMF survival models. GNN-surv models exhibit superior performance in survival prediction across two urologic cancer datasets, outperforming traditional MLP models. They maintain robustness and effectiveness under varying graph construction hyperparameter μ values, with performance boosts of up to 14.6% and 7.9% in the time-dependent concordance index and reductions in the integrated brier score of 26.7% and 24.1% in the BLCA and KIRC datasets, respectively. Notably, these models also maintain their effectiveness across three different types of GNN models, suggesting potential adaptability to other cancer datasets. The superior performance of our GNN-surv models underscores their wide applicability in the fields of oncology and personalized medicine, providing clinicians with a more accurate tool for patient prognosis and personalized treatment planning. Future studies can further optimize these models by incorporating other survival models or additional data modalities.

Detection of pathological response of axillary lymph node metastasis after neoadjuvant chemotherapy in breast cancer using multiphoton microscopy

Neoadjuvant treatment is often considered in breast cancer patients with axillary lymph node involvement, but most of patients do not have a pathologic complete response to therapy. The detection of residual nodal disease has a significant impact on adjuvant therapy recommendations which may improve survival. Here, we investigate whether multiphoton microscopy (MPM) could identify the pathological changes of axillary lymphatic metastasis after neoadjuvant chemotherapy in breast cancer. And furthermore, we find that there are obvious differences in seven collagen morphological features between normal node and residual axillary disease by combining with a semi-automatic image processing method, and also find that there are significant differences in four collagen features between the effective and no-response treatment groups. These research results indicate that MPM may help estimate axillary treatment response in the neoadjuvant setting and thereby tailor more appropriate and personalized adjuvant treatments for breast cancer patients.