Assessment of Breast Tumor’s Malignancy through Developing Virtual Physiology of Human Tumor Tissue

Introduction: The objective of this study is to investigate the correlation between a tumor’s malignancy state and forces experienced by the tumor tissue components. To conduct this objective, we have developed a patient-specific image-based in silico model of breast tumor which replicates the complexity and heterogeneity as observed in a human tumor. Using our in house-developed standard and high definition Fourier Transform Infrared (FTIR) imaging, tumor tissue is imaged and classified into six distinct components as noncancerous epithelial tissue, cancerous epithelial tissue, dense stroma, loose stroma, reactive stroma, and blood vessels. Three-dimensional image-based in silico model of these components is developed implementing molecular dynamics method. We quantified the von Mises stress distribution across the tumor tissues taking into account the mechanical properties of tumors which are crucial for accurate modeling and addressing cancer progression. Our results show that the reactive stroma (CAFs) are experiencing the highest average Von Mises stress among tissue components. CAFs, known for their pro-tumorigenic functions, play a significant role in promoting tumor growth, invasion, and therapeutic resistance. Identifying regions of heightened mechanical strain can inform targeted therapies and improve treatment outcomes. Our study provides a framework to understand components and characteristics that alter with tumor malignancy and how forces experienced by the components correlate with tumor’s malignancy. This knowledge will enhance the quantitative understanding of tumor malignancy and mechanisms involved that will increase the accuracy of early diagnosis and treatment decision-making for the most prevalent type of cancer, breast tumor.

Materials and

Methods: Biopsy tissue samples were purchased from US Biomax. These samples were sourced from a microarray panel consisting of multiple types of breast cancer tissue.
A) Acquiring images of breast tumor tissue components. We used the standard and high definition FTIR imaging to image and classify 2500 tissue samples to four distinct breast tissue pathology types as normal breast tissue, hyperplasia, invasive carcinoma of no special type, and invasive lobular carcinoma. Using our FTIR imaging microscope, we first record high-definition data and apply numerical noise reduction methods to provide high spectral signal-to-noise ratio. Standard definition data were then recorded on the same breast tissue samples.
B) Reconstructing 3D geometry of breast tumor tissue components. We selected and processed 19 tissue samples to extract the relevant tissue components for further analysis. The FTIR images of tissue samples were exported as data files which were subsequently processed using MATLAB to extract individual tissue components. For instance, in the case of the malignant tissue images, a separate image was extracted specifically for each tissue component including noncancerous epithelial tissue, cancerous epithelial tissue, dense stroma, loose stroma, reactive stroma, and blood vessels.
C) Developing human image-based in silico models of breast tumor tissue. The STL files exported from COMSOL Multiphysics were then imported into Large-scale Atomic/Molecular Massively Parallel Simulator (LAMMPS) to build three-dimensional in-silico model of breast tumor tissue. A potential function is defined to model the intracomponent and intercomponent interactions between the tissue components.

Results, Conclusions, and Discussions: Our ongoing study focuses on developing in-silico models of breast tumor tissues, leveraging patient-specific data acquired through FTIR imaging. These models aim to quantify the von Mises stress magnitude and distribution within breast tumor tissues, considering diverse pathology types. Understanding the mechanical properties is paramount for accurate in-silico modeling, as these properties are recognized hallmarks in cancer initiation and progression. This study plays a critical role in developing targeted interventions and advancing our understanding of the biomechanical factors influencing cancer progression.
The data highlight reactive stroma (CAFs) as experiencing the highest average Von Mises stress among tissue components.
CAFs, known for their pro-tumorigenic functions, play a significant role in promoting tumor growth, invasion, and therapeutic resistance.
Identifying regions of heightened mechanical strain can inform targeted therapies and improve treatment outcomes. Our results show that changes in mechanical properties of tissue components significantly alter the magnitude and distribution of stresses within the breast tumor tissue. By providing a unique platform to dissect and quantify the impact of individual stromal components on the stress distribution within a tumor tissue, this study serves as an important first step in understanding of which stromal components are vital for an efficient remodeling. This knowledge will be leveraged to overcome a tumor’s resistance against the penetration of nanoparticles on a per-patient basis. To our knowledge, this is the first study on quantifying the impact of tumor tissue components, and it will serve as an important first step in establishing unique knowledge on the characteristic factors of tumor microenvironment’s components on cancer progression with an ultimate goal of leveraging this knowledge to overcome a tumor’s resistance against the penetration of nanoparticles on a per-patient basis.

Acknowledgements (Optional):